blended learning in bioinformatics - the smes instrument ......administration with computer science,...
TRANSCRIPT
1
ERASMUS+ PROGRAMME
KEY ACTION 2: COOPERATION FOR INNOVATION AND THE EXCHANGE OF GOOD PRACTICES
STRATEGIC PARTNERSHIPS IN THE FIELD OF EDUCATION, TRAINING AND
YOUTH
Blended Learning in Bioinformatics - The SMEs Instrument for Biotech Innovations
BIOTECH-GO
2016-1-BG01-KA202-023686
Biognosis, Greece
2
O3: BIOTECH-GO Joint Study Education Programme (JSTP)
Health Informatics (additional)
3
Contents
Introduction to Health Bioinformatics ....................................................................................... 4 Definition and overview of health informatics ......................................................................................................................... 4
Medical and Health Informatics ................................................................................................ 6 Clinical Big Data ................................................................................................................................................................................ 8
Big Data in Health .............................................................................................................................................................................. 12 Electronic health records (EHRs) ............................................................................................................................................... 12 Social health ...................................................................................................................................................................................... 17 Lifestyle environmental factors and public health ................................................................................................................ 18
Healthcare Informatics ............................................................................................................. 19 Clinical informatics ........................................................................................................................................................................ 19 Integrated data repository ............................................................................................................................................................. 21 Clinical research informatics ....................................................................................................................................................... 23
Common data elements (CDEs) in clinical research ........................................................................................................... 25 Sensor Informatics .......................................................................................................................................................................... 29
Wearable, implantable, and ambient sensors .......................................................................................................................... 33 From sensor data to stratified patient management ............................................................................................................. 36 Mobile health......................................................................................................................................................................................... 36
Imaging Informatics ....................................................................................................................................................................... 38 Imaging across scales ........................................................................................................................................................................ 40 From morphology to function......................................................................................................................................................... 40 Research initiatives to understand the human brain ............................................................................................................ 41
References ................................................................................................................................. 43
4
Introduction to Health Bioinformatics
The role and the importance of health care systems in the quality of life and social welfare in
modern society have been broadly well recognized. Also, healthcare is a fundamental and most rapidly
growing sector of economy; as such for a society to realize its dream and vision of sustainable
development, it is important to be ensured mental and physical health of its individuals. “Health is
wealth” goes the popular saying and therefore in every country, the health sector is critical to social
and economic growth and development, with ample evidence linking productivity to quality of health
care. Thus, any technology or system that is tailored towards improving the healthcare system is most
likely to enjoy favorable public perception.
The evolution of medicine over the centuries is always combined with the corresponding
technological developments. At the time the Human Genome Project, an international scientific
research project which triggered a revolution in biology and medicine, was declared complete in 2003,
new opportunities came in order to develop such data, analyze, and present the genomic data in a
faster, cheaper, and more accurate and reliable way. This dramatic and rapid advance in technology
has affected healthcare in term of new prevention development, better diagnosis, and improved
treatment for regular healthcare. The high volumes of various data from bioinformatics and healthcare
informatics domains coupled with analytics are expected to deliver in the near future preventive,
predictive and personalized healthcare aids. This can contribute to sustainable national growth and
development (Oyelade et al. 2015).
Definition and overview of health informatics
Health Informatics (also called health care informatics, healthcare informatics, medical
informatics, nursing informatics, or biomedical informatics) is a combination of information science
and computer science within the realm of medicine and healthcare, which deals with various fields. It
deals with the resources, devices and methods required to optimize the acquisition, storage, retrieval
and use of information in health and biomedicine. Health informatics tools not only include computers
but also clinical guidelines, formal medical terminologies, and information and communication
systems.
Health informatics includes information applications in all areas of the health sector, resulting
in a complex and diverse landscape. Thus, health informatics includes the following fields:
– Medical Informatics contains subsets such as Clinical Informatics, Pathology
Informatics, Dental Informatics and Medical Education Informatics.
– Nursing Informatics, with sub-sector Nursing Education Informatics.
What is Health informatics? Health informatics is a multidisciplinary field that includes all information technology and telecommunications applications in the field of health, including all other relevant concepts such as medical informatics, health information systems, electronic medical record, telemedicine and mobile health.
5
– Pharmaceutical Informatics.
– Bioinformaticsand Clinical bioinformatics.
– Public Health Informatics.
– Informatics for Health Research.
– Healthcare Education Informatics.
– Consumer Health Informatics.
Figure 1. Health Informatics.
There are numerous current areas of research within the field of Health Informatics, including
Bioinformatics, Image Informatics (e.g. Neuro-informatics), Clinical Informatics, Public Health
Informatics, and also Translational Bioinformatics (TBI). Research done in all subfields of Health
Informatics range from data acquisition, retrieval, storage, analytics employing data mining
techniques, and so on. Health Informatics is applied to the areas of nursing, clinical care, dentistry,
pharmacy, public health, occupational therapy, and biomedical research.
Various studies done on Health Informatics uses data from a particular level of human
existence. Bioinformatics uses molecular level data, Neuroinformatics employs tissue level data,
Clinical Informatics applies patient level data, and Public Health Informatics utilizes population data.
On the other hand, TBI exploits data from each of these levels, from molecular level to entire
population. TBI is used to mainly answer clinical level questions. These researches in various subfields
are used to improve health care system.
Health Informatics
Telemedicine
Medical Information
Systems
mhealth
Medical Informatics
Decision Support
Systems in Medicine
Internet Health
Applications
6
Medical and Health Informatics
Medical informatics can be defined as the study of medical data translated to information,
which in turn is translated to derive useful knowledge for the benefit of health care. It also involves
their storage, retrieval, and optimal use for making health care-oriented decisions and problem solving.
On more general terms,
It is derived from two words: medical and informatics where medical indicates the area of
research and informatics indicates the methodology applied to support it (Figure 2).
Figure 2. Definition of medical informatics.
MI deals with resources, devices, and methods required for optimizing the acquisition, storage,
retrieval and use of information technology in health care. MI is formed by intersection of different
scientific fields, such as computers, engineering, biology, mathematics, and physics. Hence, a medical
informatics specialist requires knowledge and skills in these different domains to survive and
contribute towards growth of the field.
For any kind of analysis and processing, medical data is required. Medical data can be further
divided into alphanumeric data, medical imaging, and pathological and physiological signals.
Biomedical informatics (Figure 3) or medical informatics is an emerging discipline underlying
the acquisition, maintenance, retrieval and application of knowledge and information in research,
education, and service in health-related basic sciences, clinical disciplines, and health care
administration with computer science, statistics, engineering, mathematics, information technologies
and management.
Medical Informatics
Medical (indicates the
area of research)
Informatics (indicates the methodology)
Medical Informatics (MI) is the application of computers, communications and information technology systems to all fields of medicine and medical care.
7
Figure 3. Biomedical Informatics at a glance.
Biomedical informatics also provides the tools and skills needed for the development and
application of new technology for improving patient care, medical education, health sciences and
management for healthcare/hospital systems.
Biomedical informatics coalesces the related fields of Medical Informatics (now being named
Health Informatics) and Bioinformatics. Health Informatics contains subsets such as Telemedicine,
Clinical Informatics, and Dental Informatics, Pharmaceutical Informatics, Nursing Informatics and
Public Health Informatics.
It is important here to define the difference between the terms data and information. Data is
collection of record from a source, for example, a subject, books, and internet. Data can be of two
types: primary and secondary. Information, on the other hand is processed data after data
classification, which means converting raw data into meaningful form after it is classified. The central
to both medical informatics and bioinformatics is the collection and analysis of information. While
medical informatics is more concerned with structures and algorithms for the manipulation of the data
and how it can be applied in healthcare, bioinformatics is more concerned with the data itself and its
biological implications.
Medical informatics is further divided into four broad categories (see below Figure 4).
Figure 4. Categories of Medical Informatics.
Health Informatics
Health Information System
Telemedicine
Medical Decision Making
Biomedical Informatics
Statistics
Informatics
Computational Analysis
Bioinformatics
Genomics
Proteomics
Sequence databases
Knowledge Management
•Journal
•Consumer health information
•Evidence-based medical information
Clinical Information Management
•Electronic medical records
•Biling transactions
•Ordering systems
Communication
•Telemedicine
•Teleradiology
•Patient e-mail
•Presentation
Decision Support
•Reminder system
•Diagnostic expert system
•Drug interaction
8
Health care is becoming an increasingly data-intensive field as doctors and researchers generate
gigabytes of medical data on patients and their illnesses. While a patient visiting the doctor before 15
years may have only generated a few data points basic information such as weight, blood pressure, and
symptoms a medical encounter today may leave a long trail of digital data from the use of high-
definition medical imaging to implantable or wearable medical devices such as heart monitors. More
importantly, as doctors and hospitals transition away from paper medical records, this data is
increasingly being collected and made available in an electronic format. The availability of large data
sets of digital medical information has made possible the use of informatics to improve health care and
medical research. Often referred to as “in silico” research, informatics offers a new pathway for
medical discovery and investigation. The field of bioinformatics has exploded within the past decade
to keep pace with advancements in molecular biology and genomics research. Researchers use
bioinformatics to gain a better understanding of complex biological processes by, for example,
analyzing DNA sequences or modeling protein structures.
With the ability to deal with large volumes of both structured and unstructured data from
different sources, big data analytical tools hold the promise to study outcomes of large-scale
population-based longitudinal studies, as well as to capture trends and propose predictive models for
data generated from electronic medical and health records. A unique opportunity lies in the integration
of traditional medical informatics with mobile health and social health, addressing both acute and
chronic diseases in a way that we have never seen before.
Clinical Big Data
The last decade has seen an exponential growth in the quantity of clinical data collected
worldwide, triggering an increase in opportunities to reuse the data for biomedical research. The
transition from paper medical records to electronic clinical systems has been accelerated by an
emphasis on modernizing our health care infrastructure. This resulted in a significant growth in the
amount of clinical data being collected.
Figure 5 illustrates the fast increase in the number of publications referring to “big data,”
regardless of disciplines, as well as those in the healthcare domain. Although the popularity of big data
is recent, the underlying challenges have existed long before and have been actively pursued in health
research.
9
Figure 5. (a) Cumulative umber of publications referring to “big data” indexed by Google
Scholar. (b) Cumulative number of publications per health research area referring to “big data”, as
indexed in IEEE Xplore, ACM Digital library, PubMed (National Library of Medicine, Bethesda, MD),
Web of Science, and Scopus. (Andreu-Perez, Poon, et al. 2015)
Big Data is a term used to describe data sets with such large volume or complexity that
conventional data processing methods are not good enough to deal with them. Big Data has been
described disparately by different people. The most popular definition of Big Data is the 5Vs, which
are Volume, Velocity, Variety, Verification/Veracity, and Value.
➢ Volume means the large amounts of data used. The amount of data being created is vast
compared to traditional data sources.
➢ Velocity means the speed at which new data is generated. Data is being generated
extremely fast - a process that never stops.
➢ Variety means the level of the complexity of data. Data comes from different sources
and is being created by machines as well as people.
➢ Veracity is used to measure the genuineness (or trustworthiness) of the data obtained.
Big data is sourced from many different places; as a result you need to test the veracity/quality of the
data.
➢ The value gives how good the quality of data is.
➢ Another factor has also been considered, which is Variability (consistency of data over
time.
These characteristics are summarized in Figure 6 along with the key features that each capture.
The definition of Big Data might be subjected to technological advances in the future. Big Data
infrastructure is a framework, which covers important components including Hadoop
(hadoop.apache.org), NoSQL databases, massively parallel processing (MPP), and others, that is used
for storing, processing, and analyzing Big Data.
10
Figure 6. Six V’s of big data (value, volume, velocity, variety, veracity, and variability),
which also apply to health data. (Andreu-Perez, Poon, et al. 2015)
What exactly is big data?
➢ A study published by The McKinsey Global Institute (MGI) in June 2011: Define Big
Data as “datasets whose size goes beyond the ability to capture, store, manage and analyze database”.
➢ A report delivered to the U.S. Congress in August 2012 defines big data as “large
volumes of high velocity, complex, and variable data that require advanced techniques and
technologies to enable the capture, storage, distribution, management and analysis of the information”.
➢ Demchenko et al. defines big data by five Vs: Volume, Velocity, Variety, Veracity and
Value.
➢ Wikipedia defines “big data” as “data sets with sizes beyond the ability of commonly
used software tools to capture, curate, manage, and process data within a tolerable elapsed time”.
➢ In 2012 Gartner defined Big Data as “information assets characterized by their high
volume, high velocity and high variety, which demand innovative and efficient processing solutions
for the improvement of knowledge and decision making in organizations”.
However, the definition extends well beyond data volume and includes issues such as data
heterogeneity and dynamism, each of which presents its own unique challenges. Beyond the technical
considerations, the potential utility of such data has led to increasing awareness across many distinct
disciplines. In oncology, for example, the term has become synonymous with the concept of evidence-
based decision-making, where integration and analysis of large volumes of data aims to improve the
quality, efficiency, cost and outcome of critical business decisions.
Big data includes information garnered from social media, data from internet-enabled devices
(including smartphones and tablets), machine data, video and voice recordings, and the continued
preservation and logging of structured and unstructured data. Big data refers to the tools; processes
and procedures, which allow an organization to create, manipulate and manage very large data sets
and storage facilities. Big data enables an opportunity for aggregation and integration leading to cost
effective and patient care.
11
Another aspect of big data in biomedicine is the use of non-traditional data sources. These were
well illustrated, both literally and figuratively, in a 2012 paper by Eric Schadt. A complex and detailed
figure (Figure 7) showed various data types that could be mined for their effects on human health:
weather, air traffic, security, cell phones, and social media among others. But strikingly to those
reading the paper just a few years later, the list did not include personal activity trackers, e.g., FitBit,
Jawbone, or even the Apple watch. This omission of such a popular technology today is indicative of
what a fast-moving field this is.
Figure 7. Heterogeneous and non-traditional sources of big data
Big data enabled by technological advances in micro- and nano-electronics, nano materials,
interconnectivity provided by sophisticated telecommunication infrastructure, massive network-
attached storage capabilities, and commodity-based high-performance computing infrastructures. The
ability to store all credit card transactions, all cell phone traffic, all e-mail traffic, video and images
from extensive networks of surveillance devices, satellite and ground sensing data informing on all
aspects of the weather and overall climate, and now to generate and store massive data informing on
our personal health including whole genome sequencing data and extensive imaging data, is driving a
revolution in high-end data analytics to make sense of the big data, drive more accurate descriptive
and predictive models that inform decision making on every level, whether identifying the next big
security threat or making the best diagnosis and treatment choice for a given patient.
12
Big Data in Health
Big Data in medicine can have many uses and applications, in areas such as epidemiology,
clinical records, clinical operation, and administrative management, among others. The health area
integrates large amounts of data, which must be stored, classified, analyzed and consulted, all of them
with systematized and structured processes generate useful information for the achievement of
advances in health discoveries and in the administration of medical resources.
In the health sector, the data on which Big Data’s analysis techniques can be applied are as
diverse as they can be (Personal data, clinical data, administrative data). Existing analytical techniques
can be applied to the vast amount of existing (but currently unanalyzed) patient-related health and
medical data to reach a deeper understanding of outcomes, which then can be applied at the point of
care. Ideally, individual and population data would inform each physician and her patient during the
decision-making process and help determine the most appropriate treatment option for that particular
patient. The information obtained once processed analyzed and classified the data allow the obtaining
of a preventive, predictive, personalized and effective medicine.
Figure 8. An applied conceptual architecture of big data analytics. (Raghupathi & Raghupathi
2014)
Electronic health records (EHRs)
The electronic health record (EHR) is also known as Electronic Medical Record (EMR),
Computerized Medical Record (CMR), Electronic Clinical Information Systems (ECIS) and
Computerized Patient Record (CPR).
In May 2008 the National Alliance for Health Information Technology released the following
definitions in an effort to standardize terms used in HIT:
13
Electronic Medical Record: “An electronic record of health-related information on an
individual that can be created, gathered, managed and consulted by authorized clinicians and staff
within one healthcare organization”.
Electronic Health Record: “An electronic record of health-related information on an
individual that conforms to nationally recognized interoperability standards and that can be created,
managed and consulted by authorized clinicians and staff across more than one healthcare
organization”.
Personal Health Record: “An electronic record of health-related information on an
individual that conforms to nationally recognized interoperability standards and that can be drawn
from multiple sources while being managed shared and controlled by the individual”.
EHR is an evolving concept defined as a systematic collection of individual patients or
populations electronically-stored health information in a digital format. It is capable of organizing
clinical data by phenotypic categories. Also, EHR enables the communication of patient data between
different healthcare professionals (specialists etc.), as it is capable of being shared across different
health care settings, by being embedded in network-connected enterprise-wide information systems.
Especially, this term refers to computer software that physicians use to track all aspects of patient care.
Typically, this broader term also encompasses the practice management functions of billing,
scheduling, etc.
EHRs are real-time, patient-centered records that make information available instantly and
securely to authorized users. While an EHR does contain the medical and treatment histories of
patients, an EHR system is built to go beyond standard clinical data collected in a provider’s office
and can be inclusive of a broader view of a patient’s care. EHRs can:
• Include a whole range of data in comprehensive or summary form, including
demographics, patient’s medical history, diagnoses, medications and allergies, treatment plans,
immunization status, laboratory test results, radiology images, vital signs, personal stats like
age and weight, and billing information.
• Allow access to evidence-based tools that providers can use to make decisions
about a patient’s care.
• Automate and streamline provider workflow in health care settings.
One of the key features of an EHR is that health information can be created and managed by
authorized providers in a digital format capable of being shared with other providers across more than
one health care organization. EHRs are built to share information with other health care providers and
organizations - such as laboratories, specialists, medical imaging facilities, pharmacies, emergency
facilities, and school and workplace clinics - so they contain information from all clinicians involved
in a patient’s care.
14
Figure 9. Samples of HERs.
15
Figure 10. Sample view of an HER by Practice Fusion, one of the largest free HER platforms,
based on the Cloud. In the above sample we observe options such as: Charts, Past Medical History
(PMH), List of prescriptions (Rx List), Drug allergies, Vaccinations, Dating and Referrals. [Οnline]
Available: https://www.practicefusion.com/
Potential benefits of an EHR?
Benefits of an EHR can be categorized as follows:
16
A major challenge for clinical bioinformatics pertains to the accommodation of the range of
heterogeneous data into a single, query able database for clinical or research purposes. HER has
recently emerged and stimulated increased research interest. An ideal EHR provides complete personal
health and medical summary by integrating personal medical information from different sources. The
inclusion of genetic imaging and population-based information in EHR has the potential to provide
patients with valuable risk assessment based on their genetic profile and family history and to carve a
niche for personalized cancer management.
EHRs describing patient treatments and outcomes are rich but underused information.
Traditional health data centers capture and store an enormous amount of structured data concerning a
wide range of information. Mining EHRs is a valuable tool for improving clinical knowledge and
supporting clinical research, for example, in discovering phenotype information. Mining local
information included in EHR data has already been proven to be effective for a wide range of
healthcare challenges, such as disease management support, pharmacovigilance, building models for
predicting health risk assessment, enhancing knowledge about survival rates, therapeutic
recommendation, discovering comorbidities, and building support systems for the recruitment of
patients for new clinical trials. Most of this work focused on the analysis of very large
multidimensional longitudinal patient data collected over many years. However, most clinical
databases provide low temporal resolution information due to the difficulty in collecting rich long-
term time-series data. To bridge this gap, current clinical databases can be enhanced by connecting
with mobile health platforms, community centres, or elderly homes such that other information can be
incorporated into the system to facilitate clinical descision-making and address unanswered clinical
questions. One interesting direction will be to build patient-specific models using data already
available in existing clinical databases, and, then, update the model with data that can be collected
•Fewer chart pulls
•Improved efficiency of handling telephone messages and medication refills
•Improved billing
•Reduced transcription costs
•Increased formulary compliance and clearer prescriptions leading to fewer pharmacy call backs
•Improved coding of visits
Potential Productivity and Financial Improvement
•Easier preventive care leading to increased preventive care services
•Point-of-care decision support
•Rapid and remote access to patient information
•Easier chronic disease management
•Integration of evidence-based clinical guidelines
Quality of Care Improvement
•Fewer repetitive, tedious tasks
•Less "chart chasing"
•Improved intra-office communication
•Access to patient information while on-call or at the hospital
•Easier compliance with regulations
•Demonstrable high-quality care
Job Satisfaction Improvement
•Quick access to their records
•Reduced turn-around time for telephone messages and medication refills
•A more efficient office leads to improved care access for patients
•Improved continuity of care (fewer visits without the chart)
•Improved delivery of patient education materials
Customer Satisfaction Improvement
17
outside the hospitals. In particular, some chronic diseases are manifested with acute events that are
unlikely to be predictable solely by sporadic measurements made within hospitals.
Taking thoracic aortic dissection, a relatively rare disease (3–4 per 100 000 people per year),
as an example, the disease is typically manifested as a tear in the intimal layer of the aorta, which can
later on develop into either type A (involving both ascending and descending aorta) or type B
dissection (involving descending aorta only). Type-A patients would require immediate surgical
intervention, whereas for type B dissection, it is generally considered as a chronic condition requiring
careful long-term control of blood pressure (BP).
Individuals with connective tissue disorders such as Marfan syndrome (MFS) are often more
susceptible to aortic aneurysms or tears. Large-scale population screening for this rare disease will,
therefore, be useful in identifying people who are at higher risk of developing aortic dissection. For a
tear to develop into type A dissection, while others into type B dissection, one hypothesis would be
that it is due to different flow patterns generated close to the tear location and across the aorta.
Although an initial model built from imaging can give good insights into the problem, this does not
take into account progressive hemodynamic variation over time and the impact of lifestyle and daily
activities. By incorporating ambulatory BP profiles, it is possible to create simulation results as a
longitudinal model spanning over a longer period of time for a better understanding of disease
progression as summarized in Figure 11.
Figure 11. Integration of imaging, modeling, and real-time sensing for the management of
disease progression and planning of intervention procedures. This example of thoracic aortic
dissection illustrates how risk stratification and subject-specific haemodynamic modeling
substantiated with long-term continuous monitoring are used to guide the clinical decision process.
Social health
One of the primary tasks of telemedicine involves connecting patients and doctors beyond the
clinic. However, this communication has been expanded, with the involvement of social networks, to
new levels of social interaction. This new feature has opened up new possibilities of patient-to-patient
communication regarding health beyond the traditional doctor-to-patient paradigm. One-fourth of
patients with chronic diseases, such as diabetes, cancer, and heart conditions, are now using social
18
network to share experiences with other patients with similar conditions, thereby providing another
potential source of big data. In addition to biological information, geolocation and social apps provide
an additional feature to understand the behaviors and social demographics of patients, while avoiding
resource intensive and expensive studies of large statistical sampling. This advantage has already been
exploited by several epidemiological studies in areas, such as influenza outbreaks, collective dynamics
of smoking, and the misuse of antibiotics. Text messages and posts on online social networks are also
a valuable source of health information, e.g., for the better management of mental health. Compared
to traditional methods, such as surveys, fluctuations and regulation of emotions, thoughts and
behaviors analyzed over social network platforms, such as Twitter, offer new opportunities for the
real-time analysis of expressed mood and its context. For example, when validating against known
patterns of variation in mood, the 2.73 × 109 emotional tweets collected over a 12-week period in a
study reported by Larsen et al. claimed to find some correlation between emotion tweets and global
health estimates from the World Health Organization on anxiety and suicide rates.
Social media and internet searches can also be combined with environmental data, such as air
quality data, to predict the sudden increase of asthma-related emergency visits. Similar models are
anticipated to help other areas of public health surveillance.
Lifestyle environmental factors and public health
Climatological data, such as heat-stress and cold-related mortality, present another dimension
to predict personal health. Recent remote sensing technologies and geographic information systems
allow climate data for global land areas to be interpolated at a spatial resolution of 500 m to 1 km.
Achieving high-resolution measurements are necessary so as to be able to monitor the real impact of
pollution on human wellbeing in urban environments. With this aim the dense grid of wireless sensor
networks facilitates the capture of spatiotemporal variability in toxic air pollutants. Such technologies
will become increasingly important for connecting epidemic intelligence with infectious disease
surveillance and launching effective heat response plans. Similarly, patterns of social factors
influencing unhealthy habits such as smoking can be studied using the collective dynamics of social
networks. As an example of this, Christakis and Fowler found that smokers mostly belonged to the
periphery of social networks, and by the time of quitting, they behaved collectively. In addition,
smokers with higher education tended to have a greater influence on their peers toward smoking
behavior, compared to less educated smokers. As regards psychological states, emotional levels
denoting hostility and stress, expressed in social media such as Twitter tweets, can serve as predictors
of heart disease mortality per geographical area.
A mobile phone is an excellent platform to deliver personal messages to individuals to engage
them in behavioral changes to improve health. Mobile phone messaging can be used as an alternative
to deliver motivational and educational advices for changing population lifestyles. Although at present,
there is limited evidence that mobile messaging-based interventions support preventive health care for
improving health status and health behavior outcomes, a better understanding of how this platform can
be used is an interesting area to explore. For example, type-2 diabetes is generally thought to be
preventable by lifestyle modification; however, successful lifestyle intervention programs are often
labor intensive. It has been shown that mobile phone messaging can be used as an alternative to deliver
motivational and educational advices for changing population lifestyles.
19
Healthcare Informatics
Healthcare informatics deals with the best use of information, through the help of technology,
in improving healthcare, public health, and biomedical research. It is more about information than the
technology. Informatics is a blend of life-health sciences and computer and informatics sciences to
make people better.
Healthcare informatics, in the simplest definition, is the division of healthcare that is involved
in providing the right and useful information about a patient’s state of health to the right person, the
healthcare professionals, at the right time it is required, thereby guiding them in making and taking
informed decisions about their patient’s treatment. There is a synergy and exchange of information in
Healthcare informatics between patients, healthcare professionals, healthcare planners, Internet
Technology providers and management of healthcare facilities.
The name healthcare informatics originated from the coalition of similar terms (though they
are sometimes used interchangeably, but there are lots of variances between them), these are: electronic
health, popularly known as eHealth, Information management and technology also known as IM&T,
Medical informatics and Telehealth.
Healthcare Informatics combines information and understanding from medical areas (pre-
clinical, clinical and post-clinical), healthcare administration and management and information
technology. Health Informatics experts understand information technology, application of technology
to solve real life challenges and managing multifaceted processes of executing novel answers into
organizations. Besides these common capabilities, an expert in Health Informatics is also conversant
with the industry, both from the standpoint of offering medical services, and from the standpoint of
managing, challenging and demanding medical organizations and procedures.
Medical Informatics makes available the essential tools to relate data and knowledge in the
process of decision-making. Information management and technology ensures collection and
information management from various sources, including the processing and delivery of the same. The
objective is to assist in health care, increase the performance within organization, resolve business
issues as well as assist service and business plans.
eHealth is a new term often used in European countries that implements electronic and digital
technologies for healthcare practices, processes and communications.
Telehealth combines technology with health services that gives opportunity to people with
medical conditions to access healthcare within the comfort of their home. A home Pod device installed
in a patient’s home is an example of telehealth that can be used to monitor the patient’s health
condition, the information that can then be communicated securely over the server to a remote
clinician.
Clinical informatics
Clinical informatics, also known as healthcare informatics, is the study and use of data and
information technology to deliver health care services and to improve patients’ ability to monitor and
20
maintain their own health. The data and clinical decision support involved in this field are developed
for and used by clinicians, patients, and caregivers.
The field includes:
• Methods to collect, store, and analyze health care data
• The study of information needs and cognitive processes, and optimal ways to meet
those needs
• Methods to support clinical decisions, including summarization, visualization,
provision of evidence, and active decision support
• Optimizing the flow of information and coordinating it with care providers’ and
patients’ workflows to maximize patient safety and care quality
• Methods and policies for information infrastructure, including privacy and security
Figure 12. Clinical informatics.
The development of electronic medical record (EMR) systems began as a means to document
clinical activities for in-patients and out-patients. They have evolved as the primary front-line patient
care clinical tool for medical professionals. The completion of the Human Genome Project opened the
era of research in genomics and proteomics. Genome research provides keys to understanding the
mechanisms of disease. In clinical informatics, the widespread adoption of the EMR system has
generated large amounts of heterogeneous clinical data-some structured and others unstructured. In
addition, the explosive health-related contents from online communities, mobile applications, and
electronic personal health records increased the availability of non-traditional data on individual
activities and life style. Integrated use of EMR and bioinformatics is beginning to influence the
changes in the research paradigm-that is, rapid introduction of new concepts into the point of care. Dr.
Want used clinical bioinformatics (CBI) with the definition of “the clinical application of
bioinformatics-associated sciences and technologies to understand molecular mechanisms and
potential therapies for human disease”, a new and important concept for the development of disease-
specific biomarkers, mechanism-oriented understanding and individualized medicine. CBI is a new
emerging science combining clinical informatics, bioinformatics, medical informatics, information
technology, mathematics, and omics science together. Also, plays an important role in a number of
clinical applications, including omics technology, metabolic and signaling pathways, biomarker
discovery and development, computational biology, genomics, proteomics, metaboliomics,
21
pharmacomics, transcriptomics, high-throughput image analysis, human molecular genetics, human
tissue bank, mathematical medicine and biology, protein expression and profiling and systems biology.
CBI aims to deal with the challenge of integrating genomic and clinical data to accelerate the
translation of knowledge into effective treatment plan development and personalized prescription. It
is to assist clinicians in various ways, including new biomarker discovery, identification of genotype
and phenotype correlations, and pharmacogenomics at the point of care.
Understanding the interaction between clinical informatics and bioinformatics is the first and
critical step to discover and develop the new diagnostics and therapies for diseases.
Figure 13. Clinical Bioinformatics for Personalized Health.
Integrated data repository
We are well into the era of ‘‘secondary use’’ of health data, of which an important part is
reusing clinical data for both publishable research and quality improvement. A 2010 survey defined
‘‘integrated data repository’’ (IDR) as a data warehouse integrating various sources of clinical data to
support queries for a range of research-like functions. An integrated data repository (IDR) containing
aggregations of clinical, biomedical, economic, administrative, and public health data is a key
component of an overall translational research infrastructure. Such a repository can provide a rich
platform for a wide variety of biomedical research initiatives. IDRs for research are usually designed
to allow ‘‘attribute-centric’’ queries: that is, queries for the set of patients who meet some criteria
based on values of clinical observations or characteristics (also called ‘‘attributes’’ in the common
Entity–Attribute–Value (EAV) model.
Current clinical and translational research increasingly relies on the existence of robust
integrated data repositories (IDRs) with administrative, clinical, and “-omics” data.
IDRs are becoming an essential resource enabling the biomedical data reuse on larger amounts
and sources of data. Several initiatives have been carried out on IDRs to provide access to biomedical
research data, either as federated query tools or as centralized repositories. In most solutions, the
adoption of a common data format was key. However, to our knowledge, the use of specific health
information standards was limited. Besides, it is agreed that the reliability of data reuse depends greatly
on its Data Quality (DQ). Certainly, DQ assessment is considered a key component to any IDR, where
some successful examples can be found in the recent literature.
22
Following clear warehouse design principles can lower long-term maintenance costs for
organizations that are currently building or significantly restructuring their data warehouses.
Maintenance of those warehouses is very costly, and architectural changes are complicated by existing
dependencies. Getting the right architecture early during the warehouse creation is crucial. The
objective to provide an integrated data repository to researchers, clinicians, and administrators can be
met in number of ways.
Integrated data repositories (IDRs) are indispensable tools for numerous biomedical research
studies. There are three large IDRs (Informatics for Integrating Biology and the Bedside (i2b2), HMO
Research Network’s Virtual Data Warehouse (VDW) and Observational Medical Outcomes
Partnership (OMOP) repository) in order to identify common architectural features that enable
efficient storage and organization of large amounts of clinical data.
Table 2. Basic requirements for an IDR
Re-use: routine care data are re-used for research purposes or clinical purposes (e.g.,
inform care of patients based on past experience with similar patients)
Integration: data are integrated to facilitate long-term lifetime analysis, such that data
from disparate sources are linked to the corresponding patient (billing data, clinical
data) and linked to the corresponding event (order entry, order fulfillment). Moreover,
semantically identical or semantically related data are also linked.
Organization: the IDR can accommodate a wide range of source systems and is easy
to use and extend. It strikes a balance between graceful evolution and stability of the
data structures. For example, major restructuring does not occur often, and most new
data sources can be integrated without major schema change.
Maintenance: the IDR is optimized for easy maintenance, especially with respect to
adapting to changes in source systems and is robust to turnover of maintenance staff
and data analysts.
Whereas the initial design of many clinical data repositories was driven by provision of
decision support or evaluation of quality of care, their research use is rapidly increasing with
significant impact on their design. As with similar efforts in informatics, adherence to general
principles will provide some immediate benefits, with the potential for future, unanticipated benefits
as well.
The results of a pilot project of the Spanish Ministry of Health, Social Services and Equality
(2015/07PN0010), envisaged to the development of a National IDR of maternal-child care
information. An IDR was developed as solution, which, based on health information standards,
23
ensured a common interface for monitoring BPs of different hospitals and regions, and having its DQ
assessed ensures a reliable data reuse.
Examples might include correlative studies seeking to link clinical observations with molecular
data, data mining to discover unexpected relationships, and support for clinical trial development
through hypothesis testing, cohort scanning and recruitment. Significant challenges exist to the
successful construction of a repository, and they include the ability to gain regular access to source
clinical systems and the preservation of semantics across systems during the aggregation process.
Clinical research informatics
The documentation, representation, and exchange of information in clinical research are
inherent to the very notion of research as a controlled and reproducible set of methods for scientific
inquiry. Clinical research is the branch of medical science that investigates the safety and effectiveness
of medications, devices, diagnostic products, and treatment regimens intended for human use in the
prevention, diagnosis, treatment, or management of a disease. Clinical research enables new
understanding and practices for prevention, diagnosis, or treatment of a disease or its symptoms.
Clinical research is critical to the advancement of medical science and public health.
Conducting such research is a complex resource and information-intensive endeavor involving
multiple actors, workflows, processes, and information resources. Ongoing large-scale efforts have
explicitly focused on increasing the clinical research capacity of the biomedical sector and have served
to increase attention on clinical research and related biomedical informatics activities throughout the
governmental, academic, and private sectors. The result is the emergence and relatively rapid
expansion of a new sub-discipline of biomedical informatics focused on clinical research, referred to
as Clinical Research Informatics (CRI).
Clinical research informatics (CRI) is the rapidly evolving and dynamic sub-discipline within
biomedical informatics that focuses on developing new informatics theories, tools, and solutions to
accelerate the full translational continuum: basic research to clinical trials, clinical trials to academic
health center practice, diffusion and implementation to community practice, and ‘real world’
outcomes. A recent factor accelerating CRI research and development efforts is the growing interest
in sustainable, large-scale, multi-institutional distributed research networks for comparative
effectiveness research. Given the large landscape that comprises translational science, CRI scientists
are asked to conceive innovative informatics solutions that span biological, clinical, and population-
based research. It is therefore not surprising that the field has simultaneously borrowed from and
contributed to many related informatics disciplines.
In 2013, Peter Embi identified 6 categories of CRI, based on several articles:
➢ Data and Knowledge Management, Discovery and Standards;
➢ Clinical Data Re-Use for Research;
➢ Researcher Support and Resources;
➢ Participant Recruitment;
➢ Patients/Consumers and CRI;
➢ Policy, Regulatory and Fiscal Matters.
24
He also concluded that the field of CRI is broad and rapidly advancing.
The description of recent advances in CRI will be structured according to three use cases of
clinical research: Protocol feasibility, patient identification and recruitment, and clinical trial
execution. Basically, these three use cases cover the full range of clinical research.
– Protocol Feasibility
A key process in clinical research is protocol feasibility. The task is to estimate how many
patients are available according to a set of feasibility criteria (e.g. diabetes type II patients, aged 18-
60, HbA1c >8%) in a defined setting (e.g. hospitals A, B and C) and time frame (e.g. within past 12
months). A clinical study can only be successful, if a patient cohort of adequate size is existing. Patient
counts are usually sufficient to answer this question, i.e. aggregated, irreversibly de-identified data.
– Patient Identification and Recruitment
Once a clinical study is initiated, eligible patients need to be identified. It is well-known that a
large proportion of clinical trials are delayed or not successful due to issues with patient recruitment.
In contrast to protocol feasibility, aggregated patient counts are not sufficient to support patient
identification and recruitment. Candidate patient lists need to be generated and communicated to
treating physicians. In a second step, local study teams get involved.
– Clinical Trial Execution
Data management in clinical trials is costly due to the high documentation workload – on
average 180 pages per patient in a trial – and the need for high data quality. To support clinical trial
execution, data can be transferred from EHR systems into electronic data capture (EDC) systems.
Given the rapid advances in biomedical discoveries, the growth of the human population, and
the escalating costs of health care, the need to accelerate and improve clinical research is essential. In
parallel, there is a strong need for global collaboration to address the huge challenges of efficient and
effective data capture in clinical research. As such, the field of CRI has emerged as critical to solving
many of the current challenges faced by clinical researchers and the research enterprise. There is little
doubt the field is poised for continued and rapid growth, and that growth will be reflected in the
biomedical informatics literature in the years to come. Open metadata, content standards with semantic
annotation and computable eligibility criteria are key success factors for the future of CRI.
25
Figure 14. Clinical research informatics.
Figure 15. A conceptual model for clinical research informatics consisting of an informatics-
enabled clinical research workflow, heterogeneous data sources, and a collection of informatics
methods and platforms. EHR, electronic health records; IDR, integrated data repositories; PHR,
personal health records (Kahn & Weng 2012).
Common data elements (CDEs) in clinical research
Consistency in data collection is a fundamental principle of scientific research in general and
clinical trials in particular. In any given study, each opportunity for data collection is expected to meet
specifications independent of time, location, or people involved. While consistency of data collection
within an individual study is essential for maintaining data quality and enabling analysis, consistency
of data collection across multiple studies brings additional value. As biomedical research becomes
more data-intensive, and as policy and practices promote increased data sharing, greater scientific
26
opportunities emerge from the comparison and secondary use of biomedical research data. Data
sharing to support the combination of data across datasets for strengthening inferences and performing
new analyses is rapidly becoming a general expectation. One empiric approach for achieving
consistency in data collection within and across research studies is the use of Common Data Elements
(CDEs).
What are CDEs?
The term ‘‘Common Data Element’’ was initially developed by Silva and Wittes in 1999 for
case report forms used in National Cancer Institute clinical trials and has continued to evolve. As used
currently, a CDE is a combination of a precisely defined question (variable) paired with a specified set
of responses to the question that is common to multiple datasets or used across different studies. The
primary context for CDEs is in research where precision, reproducibility, and cross-study comparison
are priorities. A CDE can stand alone as a single variable or may be included in a structured collection
of elements such as a multi-item scale or index or a complex case report form.
One critical characteristic of CDEs is the use of a defined value set, where, for a question that
is designated as a variable for data collection, the permissible responses are restricted to a fixed list.
For example, if the variable is current pregnancy status, the fixed value set could be limited to yes or
no. If the variable is type of brain tumors that are gliomas of the highest grade, the fixed value set
could be, based on the current classifications, glioblastoma multiforme, gliosarcoma, or gliomatosis
cerebri.
For some CDEs, precision in defining the method of assessment may be part of the
specification. For example, if a CDE for a clinical study is defined as the result of an immunoassay,
the CDE may specify the specific way in which the assay is to be conducted. For example, with the
enzyme-linked immunospot (ELISPOT) assay to detect either antibody or cytokine secretion, the
results of several studies show ELISPOT results vary from laboratory to laboratory but can be
harmonized through rigorous training, quality assurance, and quality control measures
In practice, CDEs are identified by research communities from variable sets currently in use or
are newly developed to address a designated data need. CDE development and selection is an iterative
process guided by feasibility, utility, and acceptability that benefits from multiple stakeholders
including clinicians, informaticists, terminologists, statisticians, patients, and others. CDEs that are
specified using standardized vocabularies, codesets, and terminologies can ease the burden of data
collection and data exchange and promote discovery and interoperability between systems, including
patient registries and electronic health records.
There are no formal international specifications governing the construction or use of CDEs.
Consequently, CDEs tend to be made available by research communities on an empiric basis.
What is the value of CDEs?
CDE use has some advantages within a single study if they are perceived and implemented as
a standard or specification. CDEs can provide consistency and efficiency in establishing data collection
infrastructure and minimize variability in training and implementation. Consequently, the use of CDEs
27
can increase the efficiency, quality, clarity, and reproducibility of the overall research process and
results.
CDEs can be used to design the logic of data collection and can be embedded in case report
forms, patient registries, and integrated into collected and analytic datasets. CDEs can be expressed in
machine-readable formats to be used in data analytic plans and structured routines and scripts to
incorporate the CDE variables.
Enhanced value of CDEs is across studies to pool and combine data for meta-analyses,
modeling, and post hoc construction of synthetic cohorts for exploratory analyses. CDEs can also be
a tool to link datasets and examine relationships even if there is not a one-to-one mapping across all
data elements in multiple datasets. CDEs can be used to link and aggregate variables across multiple
datasets by identifying the CDEs and pulling the associated values into a new hybrid analytic dataset.
CDEs can also be used to map associations across datasets. Well-constructed and implemented CDEs
increase the precision and can eliminate the errors that come with other methods such as ad hoc
transformations, conversion, and manual linking.
CDEs that are used in multiple studies are a tool to leverage the substantial investment made
to collect quality data from clinical trials by increasing the consistency of data collection across studies.
The use of CDEs, especially when they conform to accepted standards, can facilitate cross-study
comparisons, data aggregation, and meta-analyses; simplify training and operations; improve overall
efficiency; promote interoperability between different systems; and improve the quality of data
collection.
Sources of CDEs
CDE are stored in collections, repositories, or libraries. Several are readily available and free
although registration may be required for access. Researchers can find pre-defined CDEs already
developed and available in sets for use with case report forms (CRFs). The National Institutes of Health
(NIH) has an extensive repository of CDE collections, and researchers are encouraged to use them in
human subject research, such as clinical studies, registries, or biorepositories. CDEs are developed
within many disciplines and it is recommended they be used to establish uniformity when collecting
or using clinical data.
NIH has an extensive collection of CDE repositories, which can be found on their Resource
portal:
CDE Repositories
General Categories: These include CDEs used across several research disciplines. Examples
from NIH listings include:
• NIH Common Data Element (CDE) Resource Portal - Collections and Efforts
• PhenX Toolkit
• PROMIS - validated patient-reported outcome measures
28
• NIH Toolbox - Validated measures of cognitive, emotional, sensory and motor
functions
Focused Categories: These are specific CDEs are those that are less common to all subjects
but more specific to research subcategories. Some examples include:
• NINDS Common Data Elements for neurological disorders and stroke clinical research
• NCI - biomarker harmonization
• NEI-AREDS - harmonize data across biorepositories
• NIDA - electronic health record extracted
• NCATS - Global Rare Diseases Patient Registry
• NIH - National Institute on Drug Abuse
Other CDE Repositories:
• Health & Human Services HHS.gov Common Data Element Repository (CDER)
Library
• A federal-wide, online searchable repository for grants-specific data standards,
definitions, and context.
Simple examples of CDEs:
• Patient age
• Gender
• Marital status
• Diagnosis
• Tumor grade
• Number of alcoholic beverages consumed over past 30 days
Development of CDEs:
The development process generally includes four basic steps:
1. A need for a CDE or group of CDEs is identified. This may generate from an
organization, an individual investigator, sponsor, or funding entity; it may initiate from a
regulatory agency such as the FDA, or from a professional society or an entire research
community.
2. Stakeholders & expert groups are identified and meet to develop or select CDE
for an identified purpose
3. Iterations & updates are made, in initial development and with ongoing input
from broader community
29
4. CDEs are endorsed by the developing body or organization. Their use is then
required, recommended, encouraged, or acknowledged as an option for use.
Sensor Informatics
Sensor Informatics (SI) is an emerging technology that is concerned with the extraction of
information from data generated by sensors, usually as part of a system designed to solve a larger
application problem.
Global healthcare systems are struggling with aging population, prevalence of chronic diseases,
and the accompanying rising costs. Also, escalated incidence and costs associated with lifestyle
induced poor health (e.g. obesity), and non-communicable diseases such as cancer and cardiovascular
diseases are major healthcare challenges globally. In response to these challenges, researchers have
been actively seeking for innovative solutions and new technologies that could improve the quality of
patient care meanwhile reduce the cost of care through early detection/intervention and more effective
disease/patient management. Rather than relying on delayed intervention and expensive treatments,
the future of a sustainable global healthcare system is one that is specifically focused on prevention,
early detection and minimally invasive management of diseases. Thus, the future healthcare system
should be preventive, predictive, preemptive, personalized, pervasive, participatory, patient-centered,
and precise, i.e., p-health system (Zheng et al. 2014). Health informatics, which is an emerging
interdisciplinary area to advance p-health, mainly deals with the acquisition, transmission, processing,
storage, retrieval, and use of different types of health and biomedical information. The two main
acquisition technologies of health information are sensing and imaging.
30
With increasing availability of sensor enriched smart, wearable devices, the knowledge about
our own health and wellbeing is also evolving. It is no longer just limited to disease progression or the
effect of therapeutic measures provided in clinical settings. Such a trend in sensor informatics has
given rise to the big personal data, which is set to influence the future of healthcare. Preventing disease
through promotion of healthy lifestyle choice is a potentially cost-effective approach to modern
healthcare challenges. Choices such as diet, physical activity, sleep, smoking and alcohol, have all
been associated with many medical conditions.
With the future of healthcare is shifting from reactive to preventive medicine, made possible
by systems approach to disease through integrated diagnosis, treatment, and prevention, we are more
focused on the quality of life and our individual wellbeing. As sensors get smaller, smarter and
increasingly pervasive, the possibility of personalised, predictive, preventive and participatory
medicine becomes increasingly realistic. Novel sensor informatics can allow the detection of disease
at an earlier stage, stratify patient management with optimised and individualised treatment, and
directly involve patients in both immediate and long-term continuous monitoring of therapeutic
responses. In addition to device level developments, the myriad of data generated continuously in real-
time represents one of the major challenges in sensor data analytics.
Advances in sensing hardware have been accelerating in recent years and this trend shows no
signs of slowing down (Poon & Zhang 2008). In Table below, we summarize some of the state-of-the-
art developments in sensing hardware covering devices used in research (Andreu-Perez, Leff, et al.
2015).
31
Recent advances in sensing technologies have made it possible to monitor health in an
unobtrusive and seamless manner, transforming episodic, largely manual sampling processes to
continuous, context-aware monitoring and intelligent intervention. Figure 24 outlines the evolution of
allied technologies in the last 10 years. Three factors in particular have contributed to these advances:
1) increased data processing power, 2) faster wireless communications with higher bandwidth, and 3)
improved design of microelectronics and sensor devices. The first two represent general trends in
computing, whereas the third is of particular interest to pervasive health. Advances in sensor
electronics have supported the development of a wide range of embedded systems, as well as devices
that are small, lightweight and can be comfortably worn by an individual or ubiquitously placed in the
environment with minimal power consumption.
32
Figure 24. Evolution of the allied technologies for pervasive healthcare in the last from 2004
to 2015.
According to the analysis in the BI Intelligence report (Garner) published at the end of 2014,
the price of one MEMS sensor has decreased by half during the last. This has partly driven a paradigm
shift of future Internet applications toward what is termed “the Internet of Things” (IoT). The Internet
of Things (IoT) is a new concept, providing the possibility of healthcare monitoring using wearable
devices. The IoT is defined as the network of physical objects, which are supported by embedded
technology for data communication and sensors to interact with both internal and external objects
states and the environment (Haghi et al. 2017). Moreover, enabling technologies ranging from nano-
and microelectronics, advanced materials, wearable/mobile computing, and telecommunication
systems, as well as remote sensing and geographic information systems have made it possible for
sensing health information to be collected pervasively and unobtrusively as illustrated in Figure 25
(Andreu-Perez, Poon, et al. 2015).
33
Figure 25. Big sensing data in health are all aroud us, enabled by techologies ranging from
nano- and microelectronics, advanced materials, wearable/mobile computing, and telecommunication
systems as well as remote sensing and geographic information systems. The inner loop presents
technologies for sensor components, while the middle loop presents devices and systems potentially
own by each individual or household. The outer loop presents sensing technologies required at the
community and public health level.
Wearable, implantable, and ambient sensors
In the last decade, wearable devices have attracted much attention from the academic
community and industry and have recently become very popular. The most relevant definition of
wearable electronics is the following: “devices that can be worn or mated with human skin to
continuously and closely monitor an individual’s activities, without interrupting or limiting the user’s
motions” (Gao et al. 2016).
Three factors have contributed to the rapid uptake of wearable devices (Andreu-Perez, Leff, et
al. 2015):
1. Increased data processing power,
2. Faster wireless communications with higher bandwidth, and
3. Improved design of microelectronics and sensor devices.
Advances in sensor electronics have supported the development of a wide range of embedded
systems, as well as devices that are small, lightweight and can be comfortably worn by an individual
or ubiquitously placed in the environment with minimal power consumption. Thus far, wearable
devices are widely used to measure key health indicators such as electrocardiogram (ECG), heart rate,
blood pressure, blood oxygen saturation (SpO2), body temperature, postures and physical activities
(Andreu-Perez, Leff, et al. 2015).
34
Today, the range of wearable systems, including micro-sensors seamlessly integrated into
textiles, consumer electronics embedded in fashionable clothes, computerized watches, belt-worn
personal computers (PCs) with a head mounted display, glasses, which are worn on various parts of
the body are designed for broadband operation. The field of wearable health monitoring systems is
moving toward minimizing the size of wearable devices, measuring more vital signs, and sending
secure and reliable data through smartphone technology. Although there has been an interest in
observing comprehensive bio/non-bio medical data for the full monitoring of environmental, fitness,
and medical data recently, but one obvious application of wearable systems is the monitoring of
physiological parameters in the mobile environment. The majority of commercially available wearable
devices are one-lead applications to monitor vital signs. However, most of such recreational devices
are not suitable for the medical monitoring of high-risk patients (Haghi et al. 2017).
Example platforms include earlier systems with limited connectivity and single sensing
elements developed solely for use in research laboratories to more recent ambient sensors as well as
easy-to-wear wearable/implantable devices equipped with continuous multi-modal sensing
capabilities and support for data fusion deployed in a wide range of clinical applications (Zheng et al.
2014). Furthermore, parallel developments in miniaturized sensor embodiment, microelectronics and
fabrication processes, and the availability of wireless power delivery have made miniaturized
implantable sensors increasingly versatile.
Implantable sensors address the challenges of both acute and chronic disease monitoring by
providing a means of capturing critical events and continuous streamlining of health information.
Recent advances in microelectronics and nanotechnology have greatly improved the sensitivity of
different sensors. For example, based on metal nanoparticle arrays and single nanoparticles, the
sensitivity of localized surface plasmon resonance optical sensors can be pushed toward the detection
limit of a single molecule. This has enabled the development of the next generation of high-throughput
sequencing technologies, as well as the detection of biomolecules, such as glucose, lactate, nitric oxide,
and sodium ions. For diabetic patients, a myriad of new sensors for both wearable and implantable
applications have been developed, which provide continuous monitoring and corresponding response
to the time-varying glucose level, which is well known to be diet dependent.
There is a clear trend of moving from the scenario where a centralized large computing
infrastructure is shared between multiple users toward one where each individual possesses multiple
smart devices, most of which are sufficiently small to be wearable or implantable such that the use of
these sensing devices will not affect normal daily activities. These sensor systems have the potential
to generate datasets, which are currently beyond our capabilities to easily organize and interpret.
Meanwhile, healthcare services delivered via ambient intelligence consisting of ambient sensors and
objects interconnected into an integrated IoT represent a promising and supportive solution for the
ageing society. It is important that such systems should take into account the sensor, service, and
system integration architecture. Such distributed systems require decentralized inference algorithms,
which are frequency explored, either in the framework of parametric models, in which the statistics of
phenomena under observation are assumed to be known by the system designer, or nonparametric
models, when the underlying data is sparse and prior knowledge is limited (Andreu-Perez, Poon, et al.
2015).
35
Figure 26. Four popular motion tracker wearable devices.
Figure 27. Unobtrusive wearable devices for various physiological measurement developed by
different groups: watch-type BP device, PPG sensors mounted on eyeglasses, motion assessment with
sensors mounted on shoes, wireless ECG necklace for ambulatory cardiac monitoring (courtesy of
IMEC, Netherlands), h-Shirt for BP and cardiac measurements, ear-worn activity recognition sensor,
glove-type pulse oximeter, strain sensors mounted on stocking for motion monitoring, and ring-type
device for pulse rate and SpO2 measurement. (Zheng et al. 2014)
36
From sensor data to stratified patient management
Physiological sensing by these smart devices can be long term and continuous, imposing new
challenges for interpreting their clinical relevance. For example, the current clinical practice defines
hypertension based on measurements taken during infrequent hospital visits. Although automated
oscillometric BP measurement devices are now available, studies in these areas are often limited to
taking BP once every hour over a 24-hr period. With the newly emerging ambulatory devices, a
comprehensive BP-related profile of an individual can be made available. Nevertheless, the
interpretation of these data is non-trivial, since in many situations, they may not be equivalent to the
clinical BP readings that are currently being used by practitioners. The signals, however, carry
underlying physiological meanings that, if properly processed and managed, can be used as additional
information for understanding uncontrolled hypertension or to enhance the current hypertension
management schemes. In addition to vital sign monitoring, smart implantable sensors provide a
promising technology to monitor postoperative complications, such as slow tissue healing and
infections. Moreover, smart implants can also have a reactive role by delivering drugs for chronic pain
and acting as brain stimulators for neurological diseases including refractory epilepsy and Parkinson’s
disease. This makes smart implants not just another resource for data collection but also an integral
part of early intervention.
With increased volume and acquisition speed of data from both wearable and implantable
sources, new automated algorithms are needed to reduce false alarms such that they are sufficiently
robust to support large-scale deployment, particularly for free-living environments. Automatic
classifications are necessary since the dataset sizes are beyond the capability of manual interpretation
within a reasonable time period. New compression-based measures are, therefore, proposed as high-
quality cloud computing services to reduce the computation time for the automated classification of
different types of cardiac arrhythmia. In many situations, measurements must be interpreted together
with the context under which the data is collected. For example, many physiological parameters, such
as BP or episodes of gastroesophageal reflux disease are posture dependent, which can be captured by
inertial sensors. Therefore, multimodal integration and context awareness are essential to the analysis
of pervasive sensing data. (Andreu-Perez, Poon, et al. 2015)
Mobile health
Mobile health technology (mHealth), which is the applying mobile communication technology
to healthcare and patient wellness, has become increasingly popular. Nowadays, smart phones have
become an inseparable companion for about 3.2 billion unique mobile users worldwide with over
30,000 available healthcare apps (Ross et al. 2014). Mobile health (m-Health) proposes to deliver
healthcare services, surpassing geographical, temporal, and even organizational barriers. M-Health
solutions address emerging problems on health services, including, the increasing number of chronic
diseases related to lifestyle, high costs of existing national health services, the need to empower
patients and families to self-care and handle their own healthcare, and the need to provide direct access
to health services, regardless of time and place (Silva et al. 2015).
M-Health uses sensors, global positioning satellite receivers, and accelerometers that
continuously monitor data. Once risk factors are determined for various conditions, the hope is to
37
integrate these factors into the EHR to assist physicians in identifying at-risk patients and build
predictive models (Ross et al. 2014). Typical m-Health services architectures (presented in Figure 28)
use the Internet and Web services to provide an authentic pervasive interaction among doctors and
patients. A physician or a patient can easily access the same medical record anytime and anywhere
through his/her personal computer, tablet, or smartphone.
M-Health services are even becoming popular in developing countries where healthcare
facilities are frequently remote and inaccessible. Mobile applications for healthcare systems are rapidly
growing and evolving. Research interest in this topic is expanding every day, as well as the
diversification of the impact areas (Silva et al. 2015).
Figure 28. Illustration of a typical architecture of m-Health services.
Due to the ongoing developments of network, mobile computing and computer storage, the
storage and retrieval of big health data have attracted great attentions. With the increasing use of
various kinds of long-term monitoring devices, there is an urgent demand for the intelligent
management of big health data. These data could be integrated into the Personalized Health Record or
EHRs, and would contribute to building an automated system to identify at-risk populations and send
automated health messages to patients (Ross et al. 2014). A large number of cloud-based storage and
retrieval systems are emerging in recent years (Zheng et al. 2014).
Many mobile applications are available for translating personal health and fitness signals into
big data (e.g. The Apple ResearchKit). Add-on technologies and so-called “wearables” are also
empowering the consumer with mobile diagnostics that can measure blood pressure level, monitor
heart rate, identify cervical cancer, and even perform an eye exam (Taglang & Jackson 2016). Many
sensors or wearable devices have a companion mobile application that uses Bluetooth, Zigbee, infrared
waves, ultra band wireless communication or a USB-based sync service to update health-monitoring
data from the wearable monitor to a connected computer or data aggregation database. As part of the
mobile Health initiative (http://www.hhs.gov/open/initiatives/mhealth/), mobile applications have
38
been designed that provide software modules to harness the internal sensors in mobile phones for the
capture of health data (Shameer et al. 2017).
The new generation of smartphones has a wide range of health apps with standardized protocol
to connect to sensors provided by different companies. They can potentially serve as a platform to
centralize health data, from which additional new information that was previously untraceable by
individual sensors can now be mined. In fact, newer models of mobile phones are packed with
sophisticated sensors that facilitate the extraction of different types of vital signs, even without the
need for external devices. These sensors can provide valuable health information for the management
of many long-term illnesses. For example, the video cameras of mobile phones can be used to collect
heart rate and heart rate variability, embedded accelerometers and gyroscopes to track energy
expenditure. Furthermore, the pulse transmission time as measured by time delays between
electrocardiographic and photoplethysmographic sensors can be used as a surrogate measure for BP.
This information can be calculated from two devices that connect with a mobile phone independently,
one with an electrocardiographic sensor and the other one with a photoplethysmographic sensor. When
connected to health providers, a closer level of interaction in healthcare can be maintained toward
greater personalization and responsiveness (Andreu-Perez, Poon, et al. 2015).
The effective management of this huge inflow of mobile-generated data calls for the
implementation of a big data solution. Therefore, healthcare organizations leverage big data solutions
to manage all of the health information, improve care and increase access to healthcare. National
Institutes of Health (NIH), awarded $10.8 milion grant to the 11 national big data centers to develop
innovative approaches in order to collect, analyze and manage health data generated by mobile and
wearable sensors. The Mobile Sensor Data-to-Knowledge (MD2K) center is a part of NIH Big Data
to Knowledge (BD2k), which aims to support advances in research, policy, and training needed for
the effective use of big data in healthcare informatics [MD2K 2015] (Fang et al. 2016).
Imaging Informatics
Imaging MS is a new technology for direct mapping and imaging of biomolecules present in
tissue sections. Imaging MS shows potential for several applications, including biomarker discovery,
biomarker tissue localization, understanding of the molecular complexities of tumor cells, and
intraoperative assessment of surgical margins of tumors.
Imaging informatics, also referred to as radiology informatics or medical imaging informatics,
is the discipline that stands at the intersection of biomedical informatics and imaging, bridging the two
areas to further our comprehension of disease processes through the unique lens of imaging, and
improve clinical care. Imaging informatics addresses not only the images themselves, but encompasses
the associated data to understand the context of the imaging study, to document observations and to
correlate and reach new conclusions about a disease ant the course of a medical problem.
Imaging informatics aims to improve the efficiency, accuracy, usability and reliability of
medical imaging services within the healthcare enterprise. Imaging informatics concerns how
information about and contained within medical images is retrieved, analyzed, enhanced, and
exchanged throughout complex healthcare systems. When clinicians have immediate electronic access
39
to medical images, precious time is saved, allowing for timely medical decisions, reducing unnecessary
repetition of exams and driving costs down.
Molecular imaging enables visualization of cellular and molecular processes that may be used
to infer information about the genomic and proteomic profiles. As a result, the bioinformatic analysis
of genomic and proteomic profiles may be valuable to assist the interpretation of images using
molecular probes. Molecular diagnostics and molecular imaging can provide the two aspects of the
disease: molecular diagnostics can provide the information of the exact mutation of a particular gene
and classify the exact type of cancer, while molecular imaging can target the very same type of cells
with that particular mutation in order to provide diagnostic information and disease staging (Schaffer
et al. 2006).
The ever-increasing amount of annotated and real-time medical imaging data has raised the
question of organizing, mining, and knowledge harvesting from large-scale medical imaging datasets.
While established imaging modalities are getting pervasive, new imaging modalities are also
emerging. These modalities are rapidly filling up the entire EM spectrum as shown in Figure 29. Many
of these imaging techniques are now geared toward real-time in situ or in vivo applications, making
multimodality imaging an exciting yet challenging big data management problem.
Recent developments in imaging are progressing in multiple frontiers. First, there is relentless
effort in making existing imaging modalities faster, higher resolution, and more versatile. Take
cardiovascular magnetic resonance imaging (MRI) as an example, imaging sequences are no longer
limited to morphological and simple tissue characterization (e.g., via T1, T2/T2∗ relaxation times).
Details concerning vessel walls, myocardial perfusion and diffusion, and complex flow patterns in vivo
can all be captured. When facilitated with new minimally invasive interventional techniques, novel
drugs and other forms of treatment, MRI now serves as a therapeutic and interventional aid, rather than
solely a diagnostic modality. Similar advances can also be appreciated for ultrasound, computed
tomography (CT), and other imaging modalities. Moreover, extensive efforts in combining different
imaging modalities, not by postprocessing, but at the hardware level, e.g., MRI/PET and PET/CT,
open up a range of new opportunities, particularly for oncological imaging and targeted therapy
(Andreu-Perez, Poon, et al. 2015).
Figure 29. Different imaging modalities across the electromagnetic spectrum.
40
Imaging across scales
There have been extensive research efforts for developing new technologies that probe deeper
into the biological system, from tissue (up to micrometer) to the protein level (micronanometer). In
particular, recent advances in stimulated emission depletion fluorescence microscopy allow the
generation of 3-D superresolution images of living biological specimens. It overcomes the classical
optical resolution limit of light microscopy and pushes the spatial resolution of optical microscope
toward the nanoscale. This opens up the possibility of imaging not only the fine morphological
structure of many organ systems (e.g., microfibrils that form blood vessels), but also subcellular
behavior and molecular signaling. The use of quantum dots or qdots also pushes the boundary of
imaging resolution, allowing the study of intracellular processes at molecular levels (20-40 nm).
Another class of fluorescent labels is made by conjugating qdots with biorecognition molecules, which
emission wavelength can be tuned by changing the particle size such that a single light source can be
used for simultaneous excitation of all different-sized dots. These technologies have already been used
for immunofluorescence labeling of tissues, fixed cells, and membrane proteins, such as cancer
markers, the hybridization of chromosomes, the labeling of DNA, and contrast-enhanced image-
guided resection of tumors (Andreu-Perez, Poon, et al. 2015).
From morphology to function
The understanding of many biological processes requires the identification and representation
of structure–function relationships. This expands across different spatial scales, namely proteins, cells,
tissues, and organs. For instance, haemodynamic analysis combined with contractile analysis,
substantiated with myocardial perfusion data, can be used to elucidate the underlying factors associated
with cardiac abnormalities. Starting with modeling, the tissue and scaling up toward a more specific
description of organ behaviors has made it possible to create integrative models of heart function.
These architectural models fuse information such as fibrous-sheet geometrical models of tissue and
membrane currents from ion channels at the subcellular level.
Amongst all organs that have been studied to define their function from its morphology, the
brain is the one that has received the most attention recently. This is motivated by the fact that brain
structure and function are keys to understand cognitive processes, hence the need for unveiling
neuronal behavior from the molecular level up to the functioning of neural circuits. Super-resolution
fluorescence microscopy has been applied to study neural morphology and their subcellular structures.
These techniques may enable us to achieve a resolution as high as 20 nm. Needless to say, the myriad
of markers necessary for each single type of cell and synapse would result in an enormous database.
Methods, such as functional MRI (fMRI) and functional diffusion tension imaging provide
flexible information in the form of macrostructural, microstructural, and dense connectivity matrices.
Improved fMRI sampling methods produce time-series data of multiple blood oxygenation-level-
dependent volumes of the brain. In addition, there is an increasing trend in making neuroimaging
multimodal. In some studies, several modalities are used to compensate the benefits and tradeoff of
one another. Furthermore, information from lower cost and rapid noninvasive methods, such as
wearable electroencephalography and functional near-infrared spectroscopy allows gathering brain
functional data for examining cortical responses due to more complex tasks.
41
An indirect way of inferring function consists of a combination of imaging modalities as well
as medical records, demographics, and lab test results. In order to maximize the information contained
in these heterogonous sources, linking different metadata with features extracted from image
modalities is key to characterize the structure, function, and progression of diseases. Solving this
challenge presents a unique opportunity for bridging the semantic gap between images and more
effective prediction, diagnosis, and treatment of diseases. However, this issue entails many
independent yet interrelated tasks, such as generating, segmenting, and extracting enormous amounts
of quantifiable spatial objects and features (nuclei, tissue regions, blood vessels, etc.). This requires
the implementation of effective and optimized querying systems in order to reduce the computational
complexity of handling these data. Fig. 9 represents a schema of what big data means for imaging, as
defined by both structural and functional data.
Figure 30. Processing schema of imaging toward big data. Nonfunctional medical imaging is
acquired and processed to serve as a model to register organ activity in the resulting functional
imaging. Results from image processing and functional imaging are stored in databases with specific
metadata protocols. Large-scale big data analysis is performed in these databases linking then the
features extracted through medical imaging processing and functional imaging.
Existing efforts in improving the spatiotemporal constraints of brain imaging are significantly
increasing the computational resources needed for neuroimaging studies. RAM memory is an
important resource for neuroimaging analysis. For instance, to perform subject-, voxel- and trial-level
analysis, a significant amount of fMRI images needs to be loaded into memory (Andreu-Perez, Poon,
et al. 2015).
Research initiatives to understand the human brain
Another active topic in imaging is to study the functional connectivity of the human brain,
which is fundamental to both basic and applied neurobiological research. Both U.S. and European
Union (EU) have launched large-scale Human Brain projects in recent years with an aim to unravel
42
the organ’s complexity. The NIH-funded Human Connectome Project (HCP) aims at leveraging the
latest advances in DTI to study brain areas in relation to their functional, structural, and
electrophysiological connectivity. The idea behind the HCP is that neural connectivity is as unique as
the fingerprint to each individual. Genetics, environmental influences, and life experience are factors
contributing to the formation of each individual’s neural circuitry. This is supported by genome-wide
association studies that link genetic variants with neurological and psychiatric disorders that have
abnormal brain connectivity, e.g., variants at human clusterin (CLU) on chromosome 8 and
complement receptor 1 on chromosome 1 are associated with Alzheimer’s disease as well as those
specific markers associated with schizophrenia and dementia.
Recently, the EU commission supported financially the human brain project, which aims to
develop a biological model of the brain that simulates different aspects of the nervous system,
including point neuron models, neural circuitry, and cellular models at different scales. The main idea
is to provide a simulation platform for theoretical neuroscientists to study how the brain processes
information. For this purpose, it would require simulating all functions, architecture, and chemical
properties for the 86 billion neurons and trillions of synapses of the human brain as estimated by
Azevedo et al.. Therefore, the aim of the project is both ambitious and controversial (Andreu-Perez,
Poon, et al. 2015).
43
References
Altman, R.B., 2012. Translational Bioinformatics: Linking the Molecular World to the
Clinical World. Clinical Pharmacology & Therapeutics, 91(6), pp.994–1000. Available at:
http://doi.wiley.com/10.1038/clpt.2012.49.
Andreu-Perez, J., Poon, C.C.Y., et al., 2015. Big data for health. IEEE journal of biomedical
and health informatics, 19(4), pp.1193–208. Available at:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7154395.
Andreu-Perez, J., Leff, D.R., et al., 2015. From Wearable Sensors to Smart Implants--Toward
Pervasive and Personalized Healthcare. IEEE transactions on bio-medical engineering, 62(12),
pp.2750–62. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25879838.
Anhøj, J., 2003. Generic Design of Web-Based Clinical Databases. Journal of Medical
Internet Research, 5(4), p.e27. Available at: http://www.jmir.org/2003/4/e27/.
Aronson, S.J. & Rehm, H.L., 2015. Building the foundation for genomics in precision
medicine. Nature, 526(7573), pp.336–42. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/26469044.
Bain, J.R. et al., 2009. Metabolomics applied to diabetes research: moving from information
to knowledge. Diabetes, 58(11), pp.2429–43. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/19875619.
Ban, T.A., 2006. The role of serendipity in drug discovery. Dialogues in clinical
neuroscience, 8(3), pp.335–44. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17117615.
Baro, E. et al., 2015. Toward a Literature-Driven Definition of Big Data in Healthcare.
BioMed Research International, 2015(1), pp.1–9. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/6137488.
Barretina, J. et al., 2012. The Cancer Cell Line Encyclopedia enables predictive modelling of
anticancer drug sensitivity. Nature, 483(7391), pp.603–7. Available at:
http://www.nature.com/nature/journal/v483/n7391/full/nature11003.html%3FWT.ec_id%3DNATU
RE-20120329.
Baskar, S. & Aziz, P.F., 2015. Genotype-phenotype correlation in long QT syndrome. Global
Cardiology Science and Practice, 2015(2), p.26. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=4614326&tool=pmcentrez&rendertype=
abstract.
Bates, D.W. et al., 2014. Big Data In Health Care: Using Analytics To Identify And Manage
High-Risk And High-Cost Patients. Health Affairs, 33(7), pp.1123–1131. Available at:
http://content.healthaffairs.org/cgi/doi/10.1377/hlthaff.2014.0041.
Benson, G., 2015. Editorial: Nucleic Acids Research annual Web Server Issue in 2015.
Nucleic Acids Research, 43(Web Server issue), pp.W1–W2. Available at:
https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkv581.
Berg, J.S. et al., 2017. Newborn Sequencing in Genomic Medicine and Public Health.
Pediatrics, 139(2). Available at: http://www.ncbi.nlm.nih.gov/pubmed/28096516.
44
Bhatia, D., 2015. Medical Informatics PHI Learni., PHI Learnign Private Limited, Delhi: PHI
Learnign Private Limited, Delhi.
Boland, M.R. et al., 2013. Discovering medical conditions associated with periodontitis using
linked electronic health records. Journal of clinical periodontology, 40(5), pp.474–82. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23495669.
Cancer Genome Atlas Network, 2012. Comprehensive molecular portraits of human breast
tumours. Nature, 490(7418), pp.61–70. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23000897.
Cars, T. et al., 2013. Extraction of electronic health record data in a hospital setting:
comparison of automatic and semi-automatic methods using anti-TNF therapy as model. Basic &
clinical pharmacology & toxicology, 112(6), pp.392–400. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23374887.
Chen, J. et al., 2013. Translational Biomedical Informatics in the Cloud: Present and Future.
BioMed Research International, 2013, pp.1–8. Available at:
http://www.hindawi.com/journals/bmri/2013/658925/.
Choi, I.Y. et al., 2013. Perspectives on clinical informatics: integrating large-scale clinical,
genomic, and health information for clinical care. Genomics & informatics, 11(4), pp.186–90.
Available at: http://dx.doi.org/10.5808/GI.2013.11.4.186.
Christakis, N.A. & Fowler, J.H., 2008. The collective dynamics of smoking in a large social
network. The New England journal of medicine, 358(21), pp.2249–58. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/18499567.
Christensen, K.D. et al., 2016. Are physicians prepared for whole genome sequencing? a
qualitative analysis. Clinical genetics, 89(2), pp.228–34. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/26080898.
Collins, F.S. & Varmus, H., 2015. A New Initiative on Precision Medicine. New England
Journal of Medicine, 372(9), pp.793–795. Available at:
http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:New+engla+nd+journal#0.
Collymore, D.C. et al., 2016. Genomic testing in oncology to improve clinical outcomes
while optimizing utilization: the evolution of diagnostic testing. The American journal of managed
care, 22(2 Suppl), pp.s20-5. Available at: http://www.ncbi.nlm.nih.gov/pubmed/26978033.
Costello, J.C. et al., 2014. A community effort to assess and improve drug sensitivity
prediction algorithms. Nature biotechnology, 32(12), pp.1202–12. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/24880487.
Danciu, I. et al., 2014. Secondary use of clinical data: the Vanderbilt approach. Journal of
biomedical informatics, 52(1), pp.28–35. Available at: http://dx.doi.org/10.1016/j.jbi.2014.02.003.
Delaney, S.K. et al., 2016. Toward clinical genomics in everyday medicine: perspectives and
recommendations. Expert review of molecular diagnostics, 16(5), pp.521–32. Available at:
https://www.tandfonline.com/doi/full/10.1586/14737159.2016.1146593.
Denny, J.C., 2014. Surveying Recent Themes in Translational Bioinformatics: Big Data in
45
EHRs, Omics for Drugs, and Personal Genomics. IMIA Yearbook, 9(1), pp.199–205. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/25123743.
Dudley, J.T. et al., 2011. Computational repositioning of the anticonvulsant topiramate for
inflammatory bowel disease. Science translational medicine, 3(96), p.96ra76. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/21849664.
Dugas, M., 2015. Clinical Research Informatics: Recent Advances and Future Directions.
Yearbook of medical informatics, 10(1), pp.174–7. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/26293865%5Cnhttp://www.pubmedcentral.nih.gov/articlerend
er.fcgi?artid=PMC4587057.
Eichstaedt, J.C. et al., 2015. Psychological language on Twitter predicts county-level heart
disease mortality. Psychological science, 26(2), pp.159–69. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/25605707.
Embi, P.J., 2013. Clinical research informatics: survey of recent advances and trends in a
maturing field. Yearbook of medical informatics, 8, pp.178–84. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23974569.
Embi, P.J. & Payne, P.R.O., 2009. Clinical research informatics: challenges, opportunities
and definition for an emerging domain. Journal of the American Medical Informatics Association :
JAMIA, 16(3), pp.316–27. Available at: http://dx.doi.org/10.1197/jamia.M3005.
Eriksson, R. et al., 2014. Dose-specific adverse drug reaction identification in electronic
patient records: temporal data mining in an inpatient psychiatric population. Drug safety, 37(4),
pp.237–47. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24634163.
Fang, R. et al., 2016. Computational Health Informatics in the Big Data Age. ACM
Computing Surveys, 49(1), pp.1–36. Available at:
http://dl.acm.org/citation.cfm?doid=2911992.2932707.
Fernández-Suárez, X.M. & Galperin, M.Y., 2013. The 2013 nucleic acids research database
issue and the online molecular biology database collection. Nucleic Acids Research, 41(D1), pp.1–7.
Forrest, G.N. et al., 2014. Use of electronic health records and clinical decision support
systems for antimicrobial stewardship. Clinical infectious diseases, 59(Suppl 3), pp.S122-33.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/25261539.
Friedl, M.A. et al., 2010. MODIS Collection 5 global land cover: Algorithm refinements and
characterization of new datasets. Remote Sensing of Environment, 114(1), pp.168–182. Available at:
http://dx.doi.org/10.1016/j.rse.2009.08.016.
Friedman, C. et al., 2004. Automated Encoding of Clinical Documents Based on Natural
Language Processing. Journal of the American Medical Informatics Association, 11(5), pp.392–402.
Available at: https://academic.oup.com/jamia/article-lookup/doi/10.1197/jamia.M1552.
Friedman, C.P., 2009. A “fundamental theorem” of biomedical informatics. Journal of the
American Medical Informatics Association : JAMIA, 16(2), pp.169–70. Available at:
http://dx.doi.org/10.1197/jamia.M3092.
Galperin, M.Y. & Fernandez-Suarez, X.M., 2012. The 2012 Nucleic Acids Research
46
Database Issue and the online Molecular Biology Database Collection. Nucleic Acids Research,
40(D1), pp.D1–D8. Available at: https://academic.oup.com/nar/article-
lookup/doi/10.1093/nar/gks1297.
Gao, W. et al., 2016. Fully integrated wearable sensor arrays for multiplexed in situ
perspiration analysis. Nature, 529(7587), pp.509–514. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/26819044.
Garnett, M.J. et al., 2012. Systematic identification of genomic markers of drug sensitivity in
cancer cells. Nature, 483(7391), pp.570–5. Available at:
http://www.nature.com/doifinder/10.1038/nature11005.
Geeleher, P., Cox, N.J. & Huang, R., 2014. Clinical drug response can be predicted using
baseline gene expression levels and in vitro drug sensitivity in cell lines. Genome Biology, 15(3),
p.R47. Available at: http://genomebiology.biomedcentral.com/articles/10.1186/gb-2014-15-3-r47.
Griffith, M. et al., 2013. DGIdb: mining the druggable genome. Nature methods, 10(12),
pp.1209–10. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24122041.
Hagar, Y. et al., 2014. Survival analysis with electronic health record data: Experiments with
chronic kidney disease. Statistical Analysis and Data Mining: The ASA Data Science Journal, 7(5),
pp.385–403. Available at: http://doi.wiley.com/10.1002/sam.11236.
Haghi, M., Thurow, K. & Stoll, R., 2017. Wearable Devices in Medical Internet of Things:
Scientific Research and Commercially Available Devices. Healthcare Informatics Research, 23(1),
p.4. Available at: https://synapse.koreamed.org/DOIx.php?id=10.4258/hir.2017.23.1.4.
Hall, M.J. et al., 2015. Understanding patient and provider perceptions and expectations of
genomic medicine. Journal of Surgical Oncology, 111(1), pp.9–17. Available at:
http://doi.wiley.com/10.1002/jso.23712.
den Hartog, A.W. et al., 2015. The risk for type B aortic dissection in Marfan syndrome.
Journal of the American College of Cardiology, 65(3), pp.246–54. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/25614422.
Hayes, D.F., Khoury, M.J. & Ransohoff, D., 2012. Why Hasn’t Genomic Testing Changed
the Landscape in Clinical Oncology? American Society of Clinical Oncology educational book.
American Society of Clinical Oncology. Meeting, 1, pp.e52-5. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/24451831.
Hayward, J. et al., 2017. Genomics in routine clinical care: what does this mean for primary
care? British Journal of General Practice, 67(655), pp.58–59. Available at:
http://bjgp.org/lookup/doi/10.3399/bjgp17X688945.
He, K.Y., Ge, D. & He, M.M., 2017. Big Data Analytics for Genomic Medicine.
International journal of molecular sciences, 18(2), p.412. Available at: http://www.mdpi.com/1422-
0067/18/2/412.
Herland, M., Khoshgoftaar, T.M. & Wald, R., 2014. A review of data mining using big data
in health informatics. Journal Of Big Data, 1(1), p.2. Available at:
http://www.journalofbigdata.com/content/1/1/2.
47
Hersh, W., 2009. A stimulus to define informatics and health information technology. BMC
medical informatics and decision making, 9(1), p.24. Available at:
http://www.biomedcentral.com/1472-6947/9/24.
Hijmans, R.J. et al., 2005. Very high resolution interpolated climate surfaces for global land
areas. International Journal of Climatology, 25(15), pp.1965–1978. Available at:
http://doi.wiley.com/10.1002/joc.1276.
Hoyt, R.E., Sutton, M. & Yoshihashi, A., 2009. Medical Informatics Practical Guide for the
Healthcare Professional,
Huser, V. & Cimino, J.J., 2013. Desiderata for healthcare integrated data repositories based
on architectural comparison of three public repositories. AMIA ... Annual Symposium proceedings.
AMIA Symposium, 2013(1), pp.648–56. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/24551366%5Cnhttp://www.pubmedcentral.nih.gov/articlerend
er.fcgi?artid=PMC3900207.
Hutchinson, S. et al., 2003. Allelic variation in normal human FBN1 expression in a family
with Marfan syndrome: a potential modifier of phenotype? Human molecular genetics, 12(18),
pp.2269–76. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12915484.
Iyer, G. et al., 2012. Genome sequencing identifies a basis for everolimus sensitivity. Science
(New York, N.Y.), 338(6104), p.221. Available at: http://www.ncbi.nlm.nih.gov/pubmed/22923433.
Jennings, L. et al., 2009. Recommended principles and practices for validating clinical
molecular pathology tests. Archives of pathology & laboratory medicine, 133(5), pp.743–55.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/19415949.
Jensen, P.B., Jensen, L.J. & Brunak, S., 2012. Mining electronic health records: towards
better research applications and clinical care. Nature reviews. Genetics, 13(6), pp.395–405.
Available at: http://www.nature.com/doifinder/10.1038/nrg3208.
Kahn, M.G. et al., 2012. A pragmatic framework for single-site and multisite data quality
assessment in electronic health record-based clinical research. Medical care, 50 Suppl(0), pp.S21-9.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/22692254.
Kahn, M.G. & Weng, C., 2012. Clinical research informatics: a conceptual perspective.
Journal of the American Medical Informatics Association, 19(e1), pp.e36–e42. Available at:
http://www.scopus.com/inward/record.url?eid=2-s2.0-84863552740&partnerID=tZOtx3y1.
Kamesh, D.B.K., Neelima, V. & Ramya Priya, R., 2015. A review of data mining using big
data in health informatics. International Journal of Scientific and Research Publications, 5(3), pp.1–
7. Available at: http://www.ijsrp.org/research-paper-0315/ijsrp-p3913.pdf.
Karczewski, K.J., Daneshjou, R. & Altman, R.B., 2012. Chapter 7: Pharmacogenomics F.
Lewitter & M. Kann, eds. PLoS Computational Biology, 8(12), p.e1002817. Available at:
http://dx.plos.org/10.1371/journal.pcbi.1002817.
Khatri, P. et al., 2013. A common rejection module (CRM) for acute rejection across multiple
organs identifies novel therapeutics for organ transplantation. The Journal of experimental medicine,
210(11), pp.2205–21. Available at: http://www.jem.org/lookup/doi/10.1084/jem.20122709.
48
Kouskoumvekaki, I., Shublaq, N. & Brunak, S., 2014. Facilitating the use of large-scale
biological data and tools in the era of translational bioinformatics. Briefings in bioinformatics, 15(6),
pp.942–52. Available at: https://academic.oup.com/bib/article-lookup/doi/10.1093/bib/bbt055.
Kovats, R.S. & Hajat, S., 2008. Heat stress and public health: a critical review. Annual review
of public health, 29(1), pp.41–55. Available at:
http://www.annualreviews.org/doi/10.1146/annurev.publhealth.29.020907.090843.
Kreso, A. et al., 2013. Variable clonal repopulation dynamics influence chemotherapy
response in colorectal cancer. Science (New York, N.Y.), 339(6119), pp.543–8. Available at:
http://www.sciencemag.org/cgi/doi/10.1126/science.1227670.
Ku Jena, R. et al., 2009. Soft Computing Methodologies in Bioinformatics. European
Journal of Scientific Research, 26(2), pp.189–203.
Laakko, T. et al., 2008. Mobile health and wellness application framework. Methods of
information in medicine, 47(3), pp.217–22. Available at:
http://www.schattauer.de/index.php?id=1214&doi=10.3414/ME9113.
Lamb, J. et al., 2006. The Connectivity Map: using gene-expression signatures to connect
small molecules, genes, and disease. Science (New York, N.Y.), 313(5795), pp.1929–35. Available at:
http://www.sciencemag.org/cgi/doi/10.1126/science.1132939.
Larsen, M.E. et al., 2015. We Feel: Mapping Emotion on Twitter. IEEE Journal of
Biomedical and Health Informatics, 19(4), pp.1246–1252. Available at:
http://ieeexplore.ieee.org/document/7042256/.
Lee, J., Kuo, Y.-F. & Goodwin, J.S., 2013. The effect of electronic medical record adoption
on outcomes in US hospitals. BMC health services research, 13(1), p.39. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23375071%5Cnhttp://www.pubmedcentral.nih.gov/articlerend
er.fcgi?artid=PMC3568047.
Lobitz, B. et al., 2000. Climate and infectious disease: use of remote sensing for detection of
Vibrio cholerae by indirect measurement. Proceedings of the National Academy of Sciences of the
United States of America, 97(4), pp.1438–43. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/10677480.
Londin, E.R. & Barash, C.I., 2015. What is translational bioinformatics? Applied &
Translational Genomics, 6, pp.1–2. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S2212066115000174.
Luber, G. & McGeehin, M., 2008. Climate change and extreme heat events. American
journal of preventive medicine, 35(5), pp.429–35. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/18929969.
Lussier, Y.A. & Liu, Y., 2007. Computational approaches to phenotyping: high-throughput
phenomics. Proceedings of the American Thoracic Society, 4(1), pp.18–25. Available at:
http://pats.atsjournals.org/cgi/doi/10.1513/pats.200607-142JG.
MacKenzie, S.L. et al., 2012. Practices and perspectives on building integrated data
repositories: results from a 2010 CTSA survey. Journal of the American Medical Informatics
49
Association, 19(e1), pp.e119–e124. Available at: https://academic.oup.com/jamia/article-
lookup/doi/10.1136/amiajnl-2011-000508.
Marcos, M. et al., 2013. Interoperability of clinical decision-support systems and electronic
health records using archetypes: a case study in clinical trial eligibility. Journal of biomedical
informatics, 46(4), pp.676–89. Available at: http://dx.doi.org/10.1016/j.jbi.2013.05.004.
Mccauley, M.P. et al., 2017. Genetics and Genomics in Clinical Practice : The Views of
Wisconsin Physicians. WMJ, 116(2), pp.69–75.
McMurry, A.J. et al., 2013. SHRINE: enabling nationally scalable multi-site disease studies.
PloS one, 8(3), p.e55811. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23533569.
Moltchanov, S. et al., 2015. On the feasibility of measuring urban air pollution by wireless
distributed sensor networks. The Science of the total environment, 502, pp.537–47. Available at:
http://dx.doi.org/10.1016/j.scitotenv.2014.09.059.
Murphy, S.N. et al., 2010. Serving the enterprise and beyond with informatics for integrating
biology and the bedside (i2b2). Journal of the American Medical Informatics Association : JAMIA,
17(2), pp.124–30. Available at: https://academic.oup.com/jamia/article-
lookup/doi/10.1136/jamia.2009.000893.
Nadkarni, P.M. & Brandt, C., 1998. Data Extraction and Ad Hoc Query of an Entity--
Attribute--Value Database. Journal of the American Medical Informatics Association, 5(6), pp.511–
527. Available at: https://academic.oup.com/jamia/article-lookup/doi/10.1136/jamia.1998.0050511.
Nagalla, S. & Bray, P.F., 2016. Personalized medicine in thrombosis: back to the future.
Blood, 127(22), pp.2665–2671. Available at:
http://linkinghub.elsevier.com/retrieve/pii/S2468171717300029.
Nunes, M. et al., 2015. Evaluating patient-derived colorectal cancer xenografts as preclinical
models by comparison with patient clinical data. Cancer research, 75(8), pp.1560–6. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/25712343.
Okada, Y. et al., 2014. Genetics of rheumatoid arthritis contributes to biology and drug
discovery. Nature, 506(7488), pp.376–81. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/24390342.
Oyelade, J. et al., 2015. Bioinformatics, Healthcare Informatics and Analytics: An Imperative
for Improved Healthcare System. International Journal of Applied Information Systems, 8(5), pp.1–
6. Available at: http://research.ijais.org/volume8/number5/ijais15-451318.pdf.
Pandey, A.S. & Divyasheesh, V., 2016. Applications of Bioinformatics in Medical
Renovation and Research. International Journal of Advanced Research in Computer Science and
Software Engineering, 6(3), pp.56–58.
Payne, P.R.O., Embi, P.J. & Sen, C.K., 2009. Translational informatics: enabling high-
throughput research paradigms. Physiological Genomics, 39(3), pp.131–140. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2789669&tool=pmcentrez&rendertype=
abstract.
Plunkett-Rondeau, J., Hyland, K. & Dasgupta, S., 2015. Training future physicians in the era
50
of genomic medicine: trends in undergraduate medical genetics education. Genetics in medicine :
official journal of the American College of Medical Genetics, 17(11), pp.927–34. Available at:
http://www.nature.com/doifinder/10.1038/gim.2014.208.
Poon, C.C.Y. & Zhang, Y.-T., 2008. Perspectives on high technologies for low-cost
healthcare. IEEE engineering in medicine and biology magazine : the quarterly magazine of the
Engineering in Medicine & Biology Society, 27(5), pp.42–7. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/18799389.
Prahallad, A. et al., 2012. Unresponsiveness of colon cancer to BRAF(V600E) inhibition
through feedback activation of EGFR. Nature, 483(7387), pp.100–3. Available at:
http://www.nature.com/doifinder/10.1038/nature10868.
Raghupathi, W. & Raghupathi, V., 2014. Big data analytics in healthcare: promise and
potential. Health information science and systems, 2(1), p.3. Available at:
http://www.hissjournal.com/content/2/1/3.
Ram, S. et al., 2015. Predicting Asthma-Related Emergency Department Visits Using Big
Data. IEEE Journal of Biomedical and Health Informatics, 19(4), pp.1216–1223. Available at:
http://ieeexplore.ieee.org/document/7045443/.
Ramachandran, A. et al., 2013. Effectiveness of mobile phone messaging in prevention of
type 2 diabetes by lifestyle modification in men in India: a prospective, parallel-group, randomised
controlled trial. The lancet. Diabetes & endocrinology, 1(3), pp.191–8. Available at:
http://dx.doi.org/10.1016/S2213-8587(13)70067-6.
Ramírez, M.R. et al., 2018. Big Data and Health “Clinical Records.” In Innovation in
Medicine and Healthcare 2017. Springer International Publishing AG 2018, pp. 12–18. Available at:
http://link.springer.com/10.1007/978-3-319-39687-3.
Rehm, H.L., 2017. Evolving health care through personal genomics. Nature reviews.
Genetics, 18(4), pp.259–267. Available at: http://www.nature.com/doifinder/10.1038/nrg.2016.162.
Relling, M. V. & Evans, W.E., 2015. Pharmacogenomics in the clinic. Nature, 526(7573),
pp.343–350. Available at: http://www.nature.com/doifinder/10.1038/nature15817.
Richesson, R.L. & Andrews, J.E., 2012. Introduction to Clinical Research Informatics. In
Health Informatics. Springer-Verlag London Limited 2012, pp. 3–16. Available at:
http://link.springer.com/10.1007/978-1-84882-448-5_1.
Rose, P.W. et al., 2011. The RCSB Protein Data Bank: redesigned web site and web services.
Nucleic Acids Research, 39(Database), pp.D392–D401. Available at:
https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkq1021.
Rose, P.W. et al., 2015. The RCSB Protein Data Bank: views of structural biology for basic
and applied research and education. Nucleic acids research, 43(Database issue), pp.D345-56.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/25428375.
Ross, M.K., Wei, W. & Ohno-Machado, L., 2014. “Big data” and the electronic health
record. Yearbook of medical informatics, 9(1), pp.97–104. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/25123728%5Cnhttp://www.pubmedcentral.nih.gov/articlerend
51
er.fcgi?artid=PMC4287068.
Sáez, C. et al., 2017. A Standardized and Data Quality Assessed Maternal-Child Care
Integrated Data Repository for Research and Monitoring of Best Practices: A Pilot Project in Spain.
Studies in health technology and informatics, 235, pp.539–543. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/28423851.
Safran, C. et al., 2007. Toward a National Framework for the Secondary Use of Health Data:
An American Medical Informatics Association White Paper. Journal of the American Medical
Informatics Association, 14(1), pp.1–9. Available at:
http://jamia.oxfordjournals.org/content/14/1/1.full.
Sanseau, P. et al., 2012. Use of genome-wide association studies for drug repositioning.
Nature biotechnology, 30(4), pp.317–20. Available at:
http://www.nature.com/doifinder/10.1038/nbt.2151.
Scanfeld, D., Scanfeld, V. & Larson, E.L., 2010. Dissemination of health information
through social networks: twitter and antibiotics. American journal of infection control, 38(3),
pp.182–8. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20347636.
Schadt, E.E., 2012. The changing privacy landscape in the era of big data. Molecular Systems
Biology, 8(612), pp.1–3. Available at: http://msb.embopress.org/cgi/doi/10.1038/msb.2012.47.
Schaffer, J.D., Dimitrova, N. & Zhang, M., 2006. Chapter 26 BIOINFORMATICS. In
Advances in Healthcare Technology. pp. 421–438.
Semenza, J.C. & Menne, B., 2009. Climate change and infectious diseases in Europe. The
Lancet. Infectious diseases, 9(6), pp.365–75. Available at: http://dx.doi.org/10.1016/S1473-
3099(09)70104-5.
Shameer, K. et al., 2017. Translational bioinformatics in the era of real-time biomedical,
health care and wellness data streams. Briefings in Bioinformatics, 18(1), pp.105–124. Available at:
https://academic.oup.com/bib/article-lookup/doi/10.1093/bib/bbv118.
Shameer, K., Readhead, B. & Dudley, J.T., 2015. Computational and experimental advances
in drug repositioning for accelerated therapeutic stratification. Current topics in medicinal chemistry,
15(1), pp.5–20. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25579574.
Sheehan, J. et al., 2016. Improving the value of clinical research through the use of Common
Data Elements. Clinical Trials, 13(6), pp.671–676. Available at:
http://journals.sagepub.com/doi/10.1177/1740774516653238.
Signorini, A., Segre, A.M. & Polgreen, P.M., 2011. The use of Twitter to track levels of
disease activity and public concern in the U.S. during the influenza A H1N1 pandemic. PloS one,
6(5), p.e19467. Available at: http://www.ncbi.nlm.nih.gov/pubmed/21573238.
Silva, B.M.C. et al., 2015. Mobile-health: A review of current state in 2015. Journal of
biomedical informatics, 56, pp.265–72. Available at: http://dx.doi.org/10.1016/j.jbi.2015.06.003.
Simon, R., 2005. Roadmap for developing and validating therapeutically relevant genomic
classifiers. Journal of clinical oncology : official journal of the American Society of Clinical
Oncology, 23(29), pp.7332–41. Available at: http://www.ncbi.nlm.nih.gov/pubmed/16145063.
52
Sirota, M. et al., 2011. Discovery and preclinical validation of drug indications using
compendia of public gene expression data. Science translational medicine, 3(96), p.96ra77.
Available at: http://www.ncbi.nlm.nih.gov/pubmed/21849665.
Sun, J. et al., 2014. Predicting changes in hypertension control using electronic health records
from a chronic disease management program. Journal of the American Medical Informatics
Association, 21(2), pp.337–344. Available at: https://academic.oup.com/jamia/article-
lookup/doi/10.1136/amiajnl-2013-002033.
Taglang, G. & Jackson, D.B., 2016. Use of “big data” in drug discovery and clinical trials.
Gynecologic oncology, 141(1), pp.17–23. Available at:
http://dx.doi.org/10.1016/j.ygyno.2016.02.022.
Tenenbaum, J.D., 2016. Translational Bioinformatics: Past, Present, and Future. Genomics,
proteomics & bioinformatics, 14(1), pp.31–41. Available at:
http://dx.doi.org/10.1016/j.gpb.2016.01.003.
Teutsch, S.M. et al., 2009. The Evaluation of Genomic Applications in Practice and
Prevention (EGAPP) Initiative: methods of the EGAPP Working Group. Genetics in medicine :
official journal of the American College of Medical Genetics, 11(1), pp.3–14. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/18813139.
The Eurowinter Group, 1997. Cold exposure and winter mortality from ischaemic heart
disease, cerebrovascular disease, respiratory disease, and all causes in warm and cold regions of
Europe. The Eurowinter Group. Lancet (London, England), 349(9062), pp.1341–6. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/9149695.
Toh, S. et al., 2011. Comparative-effectiveness research in distributed health data networks.
Clinical pharmacology and therapeutics, 90(6), pp.883–7. Available at:
http://doi.wiley.com/10.1038/clpt.2011.236.
Toubiana, L. & Cuggia, M., 2014. Big Data and Smart Health Strategies: Findings from the
Health Information Systems Perspective. IMIA Yearbook, 9(1), pp.125–127. Available at:
http://www.schattauer.de/en/magazine/subject-areas/journals-a-z/imia-
yearbook/archive/issue/1973/manuscript/22305.html.
Vilardell, M., Civit, S. & Herwig, R., 2013. An integrative computational analysis provides
evidence for FBN1-associated network deregulation in trisomy 21. Biology open, 2(8), pp.771–8.
Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3744068&tool=pmcentrez&rendertype=
abstract.
Vodopivec-Jamsek, V. et al., 2012. Mobile phone messaging for preventive health care. The
Cochrane database of systematic reviews, 12(12), p.CD007457. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23235643.
Wade, T.D. et al., 2014. Using patient lists to add value to integrated data repositories.
Journal of biomedical informatics, 52, pp.72–7. Available at:
http://dx.doi.org/10.1016/j.jbi.2014.02.010.
53
Wade, T.D., Hum, R.C. & Murphy, J.R., 2011. A Dimensional Bus model for integrating
clinical and research data. Journal of the American Medical Informatics Association, 18(Supplement
1), pp.i96–i102. Available at: https://academic.oup.com/jamia/article-lookup/doi/10.1136/amiajnl-
2011-000339.
Walker, K.L. et al., 2014. Using the CER Hub to ensure data quality in a multi-institution
smoking cessation study. Journal of the American Medical Informatics Association, 21(6), pp.1129–
1135. Available at: http://jamia.oxfordjournals.org/cgi/doi/10.1136/amiajnl-2013-
002629%5Cnhttp://www.scopus.com/inward/record.url?eid=2-s2.0-
84929044127&partnerID=40&md5=2c0c1e46853824a8779ebce39d9aabd8.
Wang, X. & Liotta, L., 2011. Clinical bioinformatics: a new emerging science. Journal of
clinical bioinformatics, 1(1), p.1. Available at: http://www.jclinbioinformatics.com/content/1/1/1.
Weber, G.M. et al., 2011. Direct2Experts: a pilot national network to demonstrate
interoperability among research-networking platforms. Journal of the American Medical Informatics
Association, 18(Supplement 1), pp.i157–i160. Available at: https://academic.oup.com/jamia/article-
lookup/doi/10.1136/amiajnl-2011-000200.
Weiner, M.G. & Embi, P.J., 2009. Toward reuse of clinical data for research and quality
improvement: the end of the beginning? Ann Intern Med, 151(5), pp.359–360.
Weinstein, J.N. et al., 2013. The Cancer Genome Atlas Pan-Cancer analysis project. Nature
genetics, 45(10), pp.1113–20. Available at:
http://www.nature.com/ng/journal/v45/n10/abs/ng.2764.html.
Weiskopf, N.G. & Weng, C., 2013. Methods and dimensions of electronic health record data
quality assessment: enabling reuse for clinical research. Journal of the American Medical
Informatics Association, 20(1), pp.144–151. Available at: https://academic.oup.com/jamia/article-
lookup/doi/10.1136/amiajnl-2011-000681.
Westfall, J.M., Mold, J. & Fagnan, L., 2007. Practice-based research--“Blue Highways” on
the NIH roadmap. JAMA, 297(4), pp.403–6. Available at:
http://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.297.4.403.
Wilhelm, M. et al., 2014. Mass-spectrometry-based draft of the human proteome. Nature,
509(7502), pp.582–7. Available at: http://www.nature.com/doifinder/10.1038/nature13319.
Wishart, D.S. et al., 2013. HMDB 3.0--The Human Metabolome Database in 2013. Nucleic
acids research, 41(Database issue), pp.D801-7. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/23161693.
Wynden, R. et al., 2010. Ontology mapping and data discovery for the translational
investigator. AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on
Translational Science, 2010, pp.66–70. Available at:
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3041530&tool=pmcentrez&rendertype=
abstract.
Xu, H. et al., 2015. Validating drug repurposing signals using electronic health records: a
case study of metformin associated with reduced cancer mortality. Journal of the American Medical
54
Informatics Association : JAMIA, 22(1), pp.179–91. Available at:
https://academic.oup.com/jamia/article-lookup/doi/10.1136/amiajnl-2014-002649.
Zheng, Y.-L. et al., 2014. Unobtrusive Sensing and Wearable Devices for Health Informatics.
IEEE Transactions on Biomedical Engineering, 61(5), pp.1538–1554. Available at:
http://ieeexplore.ieee.org/document/6756983/.
Zlotta, A.R., 2013. Words of wisdom: Re: Genome sequencing identifies a basis for
everolimus sensitivity. European urology, 64(3), p.516. Available at:
http://dx.doi.org/10.1016/j.eururo.2013.06.031.