pn healthcare 100

8/12/2019 PN Healthcare 100

http://slidepdf.com/reader/full/pn-healthcare-100 1/222

© 2010 IBM Corporation

Putting the pieces together at point of impactcan be life changing

S y m p t o m

s

UTIDiab

etesInfluenza

Hypokalemia

Renal

Failure

no abdominal pain no back pain no cough no diarrhea

(Thyroid Autoimmune) Esophagit

ispravastatinAlendronate

levothyroxinehydroxychloroquine

Diagnosis Models

frequent UTI

cutaneous lupus

hyperlipidemiaosteoporosis

hypothyroidism

Confidencedifficulty swallowing

dizziness

anorexia

feverdry mouththirst

frequent urination

F a m i l y

H i s t o r y

Graves‘ Disease

Oral cancerBladder cancerHemochromatosisPurpura

P a t i e n t

H i s t o r y

M e d i c a t i o n s

F i n d i n g s

supine 120/80 mm HG

urine dipstick:leukocyte esterase

urine culture: E. Coliheart rate: 88 bpm

SymptomsA 58-year-old

woman

complains ofdizziness,

anorexia, dry

mouth,increasedthirst, andfrequent

urination. She

A 58-year-oldwoman

presented toher primary

care

physicianafter severaldays of

dizziness,

anorexia, dry

FamilyHistoryHer family

historyincluded oraland bladder

cancer in hermother , Graves'

disease in two

sisters,

PatientHistoryHer history

was notablefor cutaneous

lupus,hyperlipidemi

a,osteoporosis,

frequent

urinary tract

Hermedications

werelevothyroxine,hydroxychlor

oquine,pravastatin,and

alendronate.

MedicationsFindingsA urine

dipstick was

positive forleukocyte

esterase and

nitrites. Thepatientgiven a

prescription

fo

• Extract Symptoms from record• Use paraphrasings mined from text to handlealternate phrasings and variants

• Perform broad search for possible diagnoses• Score Conf idence in each diagnosis based on

evidence so far

• Identify negative Symptoms • Reason with mined relations to explain away

symptoms (thirst is consistent w/ UTI)

• Extract Family History • Use Medical Taxonomies to generalize medical

conditions to the granularity used by the models

• Extract Patient History

• Extract Medications • Use database of drug side-effects• Together, multiple diagnoses may best explain

symptoms• Extract Findings: Confirms that UTI was present

Most Confident Diagnosis: Di

Most Confident Diagnosis: UMost Confident Diagnosis: Eso

Most Confident Diagnosis: Inf




Administrative Simplification

PhysicianPatient

Physician records symptoms,medical history, and vitals

Physician determines patient is at risk forbreast malignancy and wants a breast MRI.Staff requests MRI pre-authorization from

WellPoint Utilization Management (UM)

UM RN reads the pre-auth request(fax/email/transcribed vm) and creates asummary. RN queries Watson and reviews

Watson’s recommendation

Clinicalstaff

UtilizationManagement

RegisteredNurses (RN)





Clinicalstaff

RN makes a decision: authorize orforward request to WLP MD

Administrative Simplification Process with Watson

MedicalPolicies / Care

Guidelines




Oncology Advisor Use Case Workflow

Pre-Authoriza

tion

Pati

ent

Oncologist

Nurse Office Staff

Treatment

Analyzer

LongitudinalPatient Health Information

CareGuidelinesGeneticTesti

ng

CostInformation

Med

icalLiter atur

e

Morbidityand Mortality

Info

Medical

Poli

Query for patientspecific treatment

options (e.g.,mortality, morbidity,

likelihoods)

Treatmentoptions

DiscussTreatment

options

Create tailoredtreatment programand adjust as needed

Presumptive Diagnosis

CareMana

gement

Comprehensivecase management

Authorize fulltreatment program instead ofeach episode ofcare

Notifyadditional

care giversof treatmentauthorization

Claims Management

Automated claimsauthorization forproviders in patient‟scare delivery team

SURGER

Y

EMR Claims

Summary Benefits

Increase evidence based treatment

Streamline pre-auth / claims administration

Improve coordination of collaborative care

Coordinated collaborative care

THER

A

PY

Clinical

PatientData

PCP



© 2010 IBM Corporation© 2010 IBM Corporation

August 10, 2010

Fraud and Forensics:New Techniques, Better Results

National Association of State Auditors, Comptrollers, and Treasurers




Fraud, waste, and abuse costs governments hundreds ofbillions in revenues each year

“Medicaid fraud increasing throughout state, but

economy forcing cutbacks in investigations… lossescould be $2 billion a year .” Naples (FL) Daily News, 1/4/2009

Medicare &Medicaid

Tax &Revenue

Worker‘s

Compensation

UnemploymentInsurance

Food & NutritionPrograms

“…the FTB has pegged California’s tax gapassociated with the PIT and CT at $6.5 billion

annually .” California Legislative Analyst’s Office, 2008-09 Budget

“…workers’ compensation fraud is the fastest

growing insurance scam in the nation. Today, 10cents out of each premium dollar is wasted onfraud .” Connecticut Conference of Municipalities, 11/12/2008

“…sophisticated shell games are costing Michigan'sunemployment trust fund up to an estimated $95million a year. .” Michigan Unemployment Insurance Agency

“Every year, food stamp recipients exchangehundreds of millions of dollars in benefits for cash

instead of food with retailers across the country...”Lincoln (NC) Tribune, 10/21/2006

ImproperPayments

“…agencies reported a total improper paymentestimate of about $55 billion for fiscal year 2007 .”

US Government Accountability Office, 1/23/2008


http://slidepdf.com/reader/full/pn-healthcare-100 7/222© 2010 IBM Corporation

Program Estimated Improper Payments($ in billions) (error rate)

• Medicaid $18.6 10.5%• Earned Income Tax Credit $12.1 25.4%• Medicare Fee-For Service $10.4 3.6%•

Medicare Advantage $6.8 10.5%• Supplemental Security Income $4.6 10.7%• Unemployment Insurance $4.0 10%• Old Age, Survivors and Disability Insurance $2.0 0.3%• Food Stamps $1.7 5.6%• Temporary Assistance for Needy Families $1.7 9.3%• School Lunch $1.4 16.5%

Source: Office of Management and Budget: Improving the Accuracy and Integrity of Federal Payments, June 8, 2009

Date provided to OMB by the Federal Departments and Agencies.

Fraud, waste, and abuse costs governments hundreds ofbillions in revenues each year



The problem of fraud and abuse has process, organization,technology, and analytics dimensions

Fraud &Abuse

Process

Organizat ionAna ly t ics

Process Dimension

• Are proper business controls andprocedures in place to detect and deterfraud and abuse?

• Do laws and policies constrain fraudrelated processes?

• Are fraud detection and investigationprocesses optimally aligned toorganizational goals?

Organization Dimensio n

• Are there organizational barriers toimplementing an effective frauddetection and investigation

program?• Is there a single focal point (person

or team) that is responsible forfraud and abuse activities?

• Are key performance indicators inplace that measure and promoteexcellence in fraud and abusepursuit?

Analyt ics Dimension• What detection analytics, if any,

are in place?• Do detection analytics allow for

both immediate andretrospective analysis?

• Are analytics used to controland optimize fraud workload?

Technology

Technology Dimension

• What technologies are used tosupport fraud detection andinvestigation?

• Does fragmented data createchallenges in having a completepicture of behavior?

• Does technology supportmeasurement and reporting of

fraud exposure?



PredictiveModels

Who is behaving well? Which entities are likely to behave ―badly‖ in the

future? What are the indicators that an entity‘s behavior

is getting ―better‖ over time? ―Worse‖ over

time?

Data Mining &Clustering

What are patterns of non-compliant (andcriminal) behavior that I don‘t know about? If I catch a ―bad‖ entity, how can I find out who

else is behaving like that? Are there groups of entities who behave the

same way?

Which entities are behaving differently thanothers (in a suspicious way)?

How ―good‖ or ―bad‖ is a entity behaving,

relative to other entities? What is ―normal‖ behavior?

OutlierDetection

Analytics are transforming how governments are tacklingfraud, waste, and abuse



New York State Department of Taxation and Finance – Income tax refund fraud and abuse

–Debt collection

North Carolina Department of Health and Human Services –Medicaid fraud detection

Case Studies



5,000 total employees

Approximately $60 billion collected annually (2009)

Highly sophisticated taxpayer population (most Fortune 500

companies have a presence in New York) Wide range of taxpayers (demographic and cultural)

New York State Department of Taxation and Finance



The project objective was to build a system to enhance existing auditcase selection methods for fraud detection of pre-processed Returns

The Questionable Refund Detection unit wanted…

A better way to identify questionable returns

To question suspect returns before issuing refunds

To improve collect ability of audit cases

To issue refunds in a timely manner

To make program management more flexible

To leverage investments in data warehousing and business intelligencetechnologies

To scientifically predict good audit candidates utilizing return filing patterns, casehistory, and other external indicators

To improve their ability to detect new areas of fraud




CASE STUDY – State of New YorkPredicting tax compliance

The Case Identification and Selection System (CISS) appliesbusiness rules and predictive models to categorize and scorereturns nightly and identifies the „next best case‟ for audit

selection. In addition, a separate web based portal providesscreening and resolution of cases.

Solution

Benefits

Challenge New York wanted to enhance current audit case selectionmethods for detection of audit issues at the time a return isprocessed. Specific audit programs include Earned IncomeCredit, Dependent Child Care Credit, Itemized Deductions,Wage/Withholding, and Identity Theft.

$889 million increase in revenue in the first five years

Increased screener and auditor productivity

Enhanced taxpayer correspondence

Improved audit program management

13



100 employees in Program Integrity Unit

Approximately $14 billion in annual paid claims (2009)

$25 million in recoupment letters issued annually

North Carolina Department of Health and Human Services



Solution components:

IBM Fraud Analysis Center InfoSphere Identity Insight IBM Global Business Services:

Business Analytics andOptimization

IBM Software Group: LabServices

“I think we are going

to save tens of

millions of dollars.”

– Beverly Perdue, Governor

State of North Carolina

The Need:

This large state social services agency faces a significant exposureto healthcare fraud and abuse. The current business process andtechnology used to fight fraud, waste, and abuse in Medicaid isineffective – producing only around $25 million annually in recoveries.This leakage, combined with a significant state budget deficit,motivated the state to aggressively pursue cost takeout projects.

The Solution: The state implemented a comprehensive health analytics solution.This solution examines claims for suspicious patterns of behavior,quickly identifying providers and recipients for investigation. Inadditional, the solution identifies organized criminal rings andcollusive behaviors by uncovering suspicious relationships amongproviders and between providers and recipients.

Benefits:

$60m - $100m in recoupments in a 12 month period (expected)

CASE STUDY – State of North CarolinaDetecting and pursuing Medicaid fraud




Claim Data Identified $140M = 18% in suspect claims

Personal Care Services ($86M of $555M = 15%)• Schemes identified within PCS include:

High payments per patient, high home health aid visit utilization

Billing for services on Sundays/Holidays that may not have beenrendered

Out of sequence billing

Durable Medical Equipment ($55M of $235M = 23%)• Schemes identified within DME suppliers include:

Expensive orthotics for amputees

Respiratory equipment unbundling





PCS Analysis Results





CASE STUDY – Predicting the ―next best‖

case and technique for debt collections

The Collections Optimizer applies business rules andpredictive models to prioritize the „next best debt‟ for

collection. The solution will also create a customized“collection technique map” for each individual collection

case, rescoring as new events occur.

Solution

Benefits

(expected)

Challenge New York assigns collections cases based on dollar valueand uses a standard series of techniques to pursuepayment. This approach has led to a substantial backlog ofcollections cases. As new events occur that affect

collectability, cases are not reprioritized. As a result,collectors may spend time on cases that have lowprobability of collection.

$100 million increase in revenues collected over a 3 yearperiod

Substantial reduction in backlog of delinquent debts owedto the state

19





What lessons can you take away from these case studies?

1. We are in the ―early adopter‖ phase of sophisticated, real-time analytics in government2. Learn more about these methods and technologies

– Helps to provide proper oversight – Can recommend proactively in your audits

3. Analytics uses math and statistics to examine your existing data sources smarter, faster, andbetter1. Not an exercise in “magical” data sources

2. Privacy concerns (if any) can be overcome

4. Using analytics is plain ol‘ good government – Great approach for helping to close budget gaps and be responsible stewards of public funds – Can enable positive ROI in same budget cycle – Allows you to do more with the same or less – Sentinel effect

5. Adopt analytics in YOUR audits and day-to-day work6. Ways to get started

– Pick a business risk and jump in – Government Commission – Fraud Club




Frieda [email protected]

(914) 474-6606

Shaun [email protected]

(516) 203-6063

Contact Information

mailto:[email protected]






Regional Health Information Exchange ..Laying the Foundation for a NewRevolution in Patient Care

Mark D. McCou rt

Global Solut ion Sales Leader

Interoperabil i ty and Clinical Decis ion Intell igence

http://w3-03.ibm.com/marketing/branding/do/ebusinessod.html




In a 1932 visit to London, a reporter asked Mohandas Gandhi whatHe thought of Western Civilization, he replied…

“I think it would be a good idea.”

When asked the same question of the US Healthcare System, HankMcKinnell, CEO of Pfizer, comes to the same conclusion…

“I think it would be a good idea.”

A Call To Action, Hank McKinnell, CEO Pfizer




Several major trends will have a have profound effecton the delivery and financing of healthcare in 2010

Value Based Purchasing and Reimbursement – Consumers and payors (employers, health plans, and governments) willincreasingly direct their health care purchasing or reimbursement dollarbased on value

Information Liquidity – Advancements in pervasive technologies, interoperability and genomics will

increase the availability of meaningful health information. Such informationand new decision support approaches will enhance the healthcareecosystem‟s ability to manage and improve the health of populations and

individuals across geographies and the continuum of care

The Informed Consumer – The increasing thirst for healthier lives and preserving youth, increased out-

of-pocket costs and the proliferation of the health information will transformmany patients into active health care purchasers




Change is scary to many people, even the positive outcomes

“My goal is to die before there's a technology breakthrough thatforces me to live to a hundred and thirty.”




Global Healthcare Challenge: Deliver high quality,service, and access, with a sustainable cost structure

Current pressur es on the healthcare industry raise the urgency to act

Costs – Cost of care delivery continues to rise; linked to demand, inefficiency, and errors

– Government and private industry demanding reform in Healthcare to slow inflation

Resources – Limited capital for investment

– Limited pool of resources to meet anticipated demand for service

Demographics and Capacity – Aging population projections will strain the current infrastructure

– Public health deficiencies being thrust to the forefront of healthcare agenda

Technology – The maturation of collaborative technologies and analytical tools

– Public agencies focus on informatics as tool for governmental reform

– Investment in Information Technology for improved collaboration has proven to drive cost down inother industries

Prologue: Interoperability breaks the barrier to achieving greater productivity

Healthcare: All Signs Point to a need to Increase Quality of Care




Healthcare: All Signs Point to a need to Increase Quality of Carewhile Reducing the Cost of Delivery

Medical errors, many of which can be prevented, are too common - the Institute ofMedicine estimates that 44,000 to 98,000 people die each year from medical errors inthe hospital[1]

Medication errors are found in 1 of every 5 doses given in the typical hospital andskilled nursing facility, and 7% of those are potentially harmful (more than 40 per dayin a typical 300-patient facility)[2]

Health insurance costs have risen over 10% in each of the past three years.[3] Betterinformation systems are essential to reducing health care costs

One study concludes that 14% of hospital admissions occur because physicians donot have access to complete patient information

17 to 49% of diagnostic laboratory tests are performed needlessly, because medicalhistory and the results of earlier studies are not available when the new tests areordered.[4],[5]

We have no nationwide monitoring system to identify bio-terrorism in a timely

manner, to identify potential epidemics at an early stage, to identify patterns ofadverse drug reactions, or to integrate a geographically disperse Pt record

[1] ―To Err Is Human: Building a Safer Health System (2000).‖ Institute of Medicine (IOM) http://www.nap.edu/openbook/0309068371/html/

[2] Barker KN, Flynn EA, Pepper GA, et al. Medication errors observed in 36 healthcare facilities. Arch Int Med. 2002;162:1897-1903.

[3] The 2003 Kaiser Family Foundation and the Health Research and Educational Trust Employ er Health Benefits 2003 Annual Survey found that increases in health insurancepremiums were 10.9%, 12.9%, and 13.9% for 2001, 2002, and 2003 respectively. See http://www.kff.org/insurance/ehbs2003-1-set.cfm for details.

[4] Tierney WM, McDonald CJ, Martin DK, Hui SL, Rogers MP. Computerized display of past test results.: effect on outpatient testing. Ann Intern Med. 1987;107:569 –74.

[5] HIMSS. ―EHR and the Return on Investment.‖ 2003. http://www.himss.org/content/files/EHR-ROI.pdf

http://www.nap.edu/openbook/0309068371/html/

http://www.kff.org/insurance/ehbs2003-1-set.cfm

http://www.himss.org/content/files/EHR-ROI.pdf









http://www.nap.edu/openbook/0309068371/html/




Consumerism

Individuals are increasingly responsible for the cost oftheir own care, requiring new value propositions &markets

New incentives are created for customer service,information mgmt, & efficiency

Increased demand for new drugs technology & units ofservice are fueled by the aging and expandingpopulation and informed consumers

Pay for Performance

• Practitioners and providers are responding positively tostructured incentives

• Payers are using new incentives to assess providersand promote cost efficiency and high performance

Suboptimal Quality of Care

• Poor healthcare quality is at epidemic levels in the US,resulting in as many as 3M injuries each year & more than300,000 deaths

• Patients often fail to receive recommended treatments forcommon conditions

• CMS have begun linking quality of services to paymentincentives based on MMA metrics

Increasing Complexity of Medicine

• Clinical instruments and treatments, many of which cannot beimplemented without computerization, are growing increasinglycomplex

• Practitioners are employing predictive modeling and analyticsto improve diagnoses and effective treatments

Pervasive Medicine

• Pervasive devices and technologies are demonstratingsignificant impacts throughout the value chain, allowingfor real-time EHR updates and promoting efficientcollaboration

• Healthcare providers are expanding their reach beyondhospitals walls with the use of handheld devices &wireless infrastructures

Healthcare System Inefficiency

• Inefficiencies are pervasive driven largely by paper-basedprocesses and poor communications across ecosystemplayers, resulting in redundancy and labor-intensiveadministrative processes

• Information technology is widely recognized as offeringpotential solutions to inefficiency, however barriers existincluding availability of capital and the need for associatedprocess & culture change

Regulatory Compliance

Regulatory oversight is impacting healthcare at the federal,state, and local level

HIPAA standards are requiring careful handling, use, andmonitoring of patient information

FDA is increasing in regulatory stringency for standards inBiotechs and Pharmas

MMA provides for a wide variety of payment options forproviders

Sources: Forester Trends 2005: Health Plans,, Gartner PreA dictions for Healthcare Market; Internal HCLS Level 0 and Level 1 Strategy Docs

Consumers

Providers

Public Health

Payers

P h a r m a

Increasing Healthcare Cost

• Healthcare costs are rising annually at 2-4 times the rate ofinflation• In 2004, employer health insurance premiums increased by

11.2 percent• The US spends more than 15% of GDP on healthcare, at

least 50% more other developed nations, and rank lower onoutcomes such as life expectancy

• Chronic Disease represents 80% of cost for 20% population

MitigatorsContributorsCore Issues

Key forces impacting the current healthcare ecosystem

Geometrically IncreasingHealthcare Cost

Suboptimal Quality of CareDifficult to Measure

Regulatory Compliance

Increasing Complexityof Medicine

Pay for Performance

Pervasive Medicine

Consumerism

Delivery Model MismatchWith Business Model

Enabling Technology Adding both Value and Complexity




The traditional institutional approach to systemsand data management make it extremely challengingto deliver and interpret information




Enhancing PtSafety

ReducingCosts

Moving Toward Improving Quality of Care

Departmental AutomationI n

c r e a si n gA

ut o m ati o n

Increasing Value

QualityTarget

Information Delivery

Information Linkage

Information Access

Clinical UsabilityClosing the loop

Automation isKey to MeasuriQuality

LIS, RIS, Rx, ADT,HIS, CPOE, EMR

MPI, Lexicon, Entity Analytics

Portals, Context Mgt

Community Integration

Data Warehousing,Mining

Modify Care Standards

HL7, IE, HUBs

Finding the data




·

Globally national healthservices and

departments areundertaking efforts toimprove sharing ofclinical information …

… and locally manyinteroperability effortsare launching in the US

The global healthcare challenge to improve quality and productivity inhealthcare is driving initiatives world wide




Inform Clinical Practice

Strategy 1. Incentivize EHR adoption

Strategy 2. Reduce risk of EHR investment

Strategy 3. Promote EHR diffusion in rural and underservedareas

Personalize Care

Strategy 1. Encourage use of Personal Health Records

Strategy 2. Enhance informed consumer choice

Strategy 3. Promote use of telehealth systems

Improve Population Health

Strategy 1. Unify public health surveillance architectures.

Strategy 2. Streamline quality and health status monitoring

Strategy 3. Accelerate research and dissemination ofevidence

Interconnect Clinicians

Strategy 1. Foster regional collaborations

Strategy 2. Develop a national health information network

Strategy 3. Coordinate federal health information systems

IBM endorses the Health and Human Services ―Decade of Health‖strategic framework which gives the industry a starting point onthe road to a national health policy

Goals from the HHS Framework for Strategic Action – July 21, 2004




Interoperable healthcare networks can improve the efficiency,cost effectiveness, and safety of healthcare, while delivering thegreatest possible health benefits and value to consumers

Sources: Center for Information Technology Leadership, Partners Health Care, Harvard (2004)

The Network Effect

Payers

Providers

Consumersof Care

Employers

Sample Health InformationExchange Benefits

Public Health

Improved population health

Improved wellness

Improved monitoring and safety

Payers and Employers

Reduced costs

Reduced MLR

Lower absenteeism

Efficiency

Providers

Reduced errors

Improved quality

Efficiency

Other (e.g., Life Sciences)

Faster routes-to-market

Potential U.S. netefficiency gain from use ofIT in health care:> $86B per yearOr 5% of total healthcareexpenditures of $1.7 T

Community Health

Information Exchange

$55B

Outpatient EHR

$25B

Inpatient EHR

$6B




Forrester recently reported the following as a list of activecommunity clinical sharing projects, but we already know thatmany others are brewing

From 2005 HIMSS Presentation: View From HIMSS: Building Regional Health Information NetworksEric G. Brown, Vice President Forrester Research




While a common vision is developing, the approaches to clinicalinformation sharing vary widely

Grass Roots Sharing Speed to Value

From formation intended to allow forrobust clinical information sharing ofdifferent points eHR components fromdisparate sources

Examples:

Santa Barbara County Care Data Exchange: eHRsharing across multiple stakeholders with federatedstorage of information.

Taconic Health Information Network and Community(THINC): Physician driven effort to connect officeeHRs with Hospitals. Broad stakeholderinvolvement.

SAFE Health: eHR-lite sharing across community ofWorcester, MA. Joint input from Fallon HP, Fallon

Clinic, and UMMC.

Massachusetts eHealth Collaborative: Broadlyfocused effort to develop interoperability model forthe State. Still in planning phase.

Initial goals on delivering morefocused functionality and services to

jump start community sharing

Examples:

The Indiana Health Information Exchange (IHIE):Query ability of 13 institutions and their medicalstaffs to aggregated CDR. Ability to deliverphysician reports electronically.

Delaware Health Information Network (DHIN):Creation of eHR-lite from available claims dataavailable first in emergency setting then wideraccess.

Michigan Health Information Infrastructure;

MA-SHARE/MedsInfo ED:

Access to prescription fulfillment information fromPharmacy Benefits Management firms with plansto include ePrescribing

We view the RHIO focus from the U.S. Federal Government as a




Information

Sources• Clinician Offices• Hospitals• Labs• Payers/PBMs• Pharma• Patients• Other

Information

Users• Clinicians• Hospitals• Payers• Pharma• Public Health• Patients• Other

Business Value

Productivity Quality Economy

Infrastructure Boundaries

Enterprise Local Aggregate Regional Aggregate Macro

Data Exchange• Mapping• Protocols• Traffic

Management

InformationManagement

Functions

• Warehouse

• Analytics• Reports• Query

Network Access&Management

• MPI• Pointers to data• Security• Lexicon• Audit

• Help Desk• Portal

Interoperability Relationship Model

catalyst to realize the network effect of information in healthcare

Core Services




Consumers of Care

Providers

Public Health

Payers

For stakeholders to maximize benefits in the new ecosystemadditional value must be obvious at the end-user application level

Services-Oriented Architecture

Decision SupportProtocols

Disease and CareManagement Transaction

Engine forReferrals & Auth.

PredictiveModeling

Case / UtilizationManagementProtocols

CDR

Acute Care HIS

Primary Care EMR

CDR

LocalIntegration

Patient/MemberPortal

PersonalHealth Record

Health/WellnessInformation

Chat

Macro Integration Enginefor

• Business services• Message handling• Security• Information management• Application interface

BiosurveillanceQuality of CareManagement

LocalIntegration

Current ExchangeLocalized by domain ormanually driven

Future Exchange• Real time• Clinical focus• Aggregated, and• Automated

MedicalManagement

AdministrativeEfficiency

LocalIntegration

Clinical Research

Program

Management

•eRX•Lab•eHR/ eMR•Messaging•Ref & Auths




There are many entry points to begin realizing the benefits of

Interoperability

InteroperabilityInfrastructure

C D I e - P r e s

c r i b i n g

E H R

E D S h a r i n g B i

o - s u

r v e i l l e n c e

R e g

u l a t o r y

R e p o r t i n

g

C l i n i c

a l

M e s

s a g

i n g

C l i n i c

a l

R e s e a r c h

R e g u l a

t o r y

R e p o r t i n

g

C o n

s u m

e r

P o r t a l

DiseaseMgmt

C a r e M g m t

F r a u d

a n d A b u s e M g m

t

M e d i c a l

M g m

t

P r e d i c t i v e

M o d e l i n g

C l i n i c

a T

r i a l

C a n

d i d

a t e

I D

A d v e

r s e

E v e

n t

A l e

r t s

Key to entry points

Provider

Payer

Pharma

D M / C M

D e c i s i o n

S u p p o r t

What Interoperability enables – pick your starting point(s)




For interoperability to succeed you must conceptualize the roadmap that ultimately will deliver you to a sustainable operationalmodel in which stakeholders find clear benefits for participation

Reason for Being

Vision

Mission

Guiding Principals

By-Laws

Goals

Objectives

Success Criteria

Business Modeling

Systems Modeling

Business Plan

Financial Plan

Systems Plan

Macro Design

Policy andProcedureDefinition

Design GoalPrioritization

Use CaseDevelopment

RequirementsDefinition

Micro Design

TechnologySelection

Build

Unit Test

Systems Test

User Test

Roll Out

Repeat for newstakeholders andfunctionality

StrategicOperations

BusinessOperations

AdministrativeOperations

User Technical

Support

Network Operations

Governanceand Strategy Business andSystems Design RequirementsDefinition and SolutionDesign

SolutionImplementation On-goingOperations andTechnicalSupport

Healthcare leaders backed the HCN demonstration, to enable




Leverage Acceptable Data Sharing Protocols

Promote Clinical Reporting Standards

Healthcare leaders backed the HCN demonstration, to enablerapid sharing of health data and improved bio-surveillance, whichincluded federal agencies and nationally recognized institutions

Healthcare Collaborative Network (HCN) Solution

• Enables rapid detection and response to adverse healthcareevents including bio-surveillance

• Creates lower cost capabilities for collecting, aggregating,analyzing and reporting clinical information at near real time

• Establishes a common electronic healthcare information highwaythat supports government, non-profit, and private industry needs

IBM

CDC

CMS

FDA

NY PresbyterianHospitals

Vanderbilt UniversityMedical Center

Wishard MemorialMedStar Health*

* MedStar hospitals connected are: Franklin Square Hospital Center; Good Samaritan Hospital;Harbor Hospital Center; Union Memorial Hospital

Initial goals

Demonstrationparticipants

HCN A hit t




HCN Architecture

Components: Portal,Broker, Gateway

Leverages legacysystems

HCN uses existingopen data standardsand a non-proprietaryimplementationapproach (e.g. ICD,CPT, LOINC, viaHL7)

Meets highestsecurity standards(authentication,encryption, HIPAA)

As soon as a healthtopic is satisfied allinformation availableon the related patientevent can be shared

EHR

Key Design Elements




Specific examples of HCN in practice give a picture of howorganizations can benefit

Clinical Data Repository Population (EHR)

Aggregate electronic health records from multiple sources for patient care

For intelligent data warehouse population

Quality of Care Analysis

Prompt identification/notification of patients for whom care departs from established carepaths

Development of best practice guidelines for specific patient populations facilitated through

easier aggregation and analysis of outcomes across disparate systems Evaluation of the practices pattern for specific patient population to support outcomes

assessments

Detection of Adverse Drug Events

Evaluation of drug effects relative to outcome indicators like lab values

– Patient with prescription for Coumadin exhibiting International Normalized Ratio (INR)/ Prothrombin Time labresult greater than 8

Monitoring across larger populations for lower frequency adverse events

Public Health Alerts

Rapid notification of sentinel events related to direct diagnosis of public health concerns

Rapid notification of syndromic indicators which could warrant deeper epidemiologicinvestigation as precursors of outbreaks




Across the healthcare ecosystems HCN supports theneeds of many stakeholder relationships

Hospitals - Internal collaboration for :

Research/ Outcomes Analysis/ Quality Improvement Studies Adverse Event Detections

Hospitals and PayersImproved coordination for case managementImproved identification of disease management candidatesSupport for quality incentive programs

Hospitals and Pharmaceutical ResearchersIdentification of candidates for clinical trialsPost market population analysisCompliance observation for outcomes analysis

States (and Local) - Public Health ReportingBio-surveillance/ Situational awarenessHealth/disease management program candidate evaluation, outcomes analysis and resource planning

Population monitoring

HCN is ready today to enhance the capabilities for




y y pcollaboration among various stakeholders in support ofimproved quality and reduced medical errors

Summary of HCN solution benefits:

Enables monitoring groups (e.g. Local health jurisdictions, FDA,CMS, CDC) to improve detection and response time for bio-surveillance, adverse drug reactions, quality of care, and diseaseoutbreaks

Enables rapid ability to aggregate and share data

Enhances ability to judge quality of care

Facilitates and improves efficiency of mandatory reporting and collaboration with business partners

Provides a secure environment for clinical data transmission using SSL and the highest level of encryption

Leverages existing applications minimizing barriers toimplementation




PatientPortal

PhysiciansPortal(PCP)

ERDocPortal

Patient ID.

ResultsHC Data EntryClinical MsgsCollaboration

Patient ID.

Ins. FinancialHealth StatusClinical Msgs

AlertsCollaboration

EnterpriseService Bus

Medical Mgmt

QualityPrograms

Collaboration

MED LocationServices

PartnersDirectories• Patient• Provider• Location

ContextManagement CDR Security

ProviderSystems

Consumer Portal Provider Portal Payer Portal

IBM believes the service oriented architecture model “the bus”is best to meet the needs of multi-stakeholder solutions

The Enterprise Healthcare Service Bus can offer stakeholders




The Enterprise Healthcare Service Bus can offer stakeholdersbasic as well as enhanced functions, both clinical andtechnical

Business Services

Authentication and Access Control

Patient Index and Cross Reference

Location Services

Decision Support Engine

Risk Modeling Engines

Repository – Consumer or Provider

Workflow and Context Management MED – Medical Entities Dictionary

NLP – Natural Language Processing

Business Partner Connections –Lab –Pharmacy –Eligibility

Audit Logging Monitoring, Reporting, and Alerting

Technology Foundation

Integration Engine – Industry Specific Connectors and DataHandlers

– HL7, CDA, CCR – DICOM – XML – ebXML – EDI – Connectors

– JDBC – FTP, SMTP – ODBC – Text, etc. – CIT System application adapters

Mapping, Transformation, De-Identification- Pub/Sub

Applications supporting services (e.g. EMR,

Clinical Messaging, eRX, Decision Support)




RHIO (shared infrastructure)

CentralizedStorage

DocumentRegistry

DocumentRepository

DocumentConsumer

A n n o t a t i o n

L o c a t o r

S u m m a r i z a t i o n

( E H R E x t r a c t )

W o r k f l o w

HISRIS

Large Healthcare Enterprise

DocumentSource

PIX

Rural Practice

Portal

NHIN Bus

N H I N I n t e r f a c e

E

n t er pr i s e

S er v i c e

B u s

Healthcare enterprise

Vendor components

IHE XDS Components

NHIN level components

Enterprise Service Bus

MPI Server

IHII Architecture Clinical Affinity Domain

PC or Browser

HospitalInformationSystem

IHE XDS Standard

IBM is extending this model through the development of the

Integrated Healthcare Infrastructure (IHII)

Expectations for realizing the goal of interoperability need to




Technology

• Spend more time up front on appropriatedesign to mitigate implementation churn

• Technology for this solution can be complex- think component and standards based tofacilitate integration

• Assure technical design captures thenetwork goals but can be componentized toallow for manageable phases

• Hire skilled and technically knowledgeableresources to manage projects AND day today operation or a technology partner toprovide same

Process

• Clearly define how decisions will be made for the initiative and the organization

• Set conservative operating budgets (sunny day vs rainy day planning)

• Be realistic about attaining financial targets. Sustainability will come from making the numbers work for yourcommunity

• Address legal and policy issues early as challenges to design but not deal breakers

People

Technology

Process

People

• Committed leadership from primarystakeholders

• Make communications a cornerstone ofyour operations - among stakeholdersand the community

• Establishment of realistic goals and aplan to progressively add value tostakeholders as functionality matures

• Define clear and measurable success

criteria• Don‟t take short cuts especially onfoundational governance tasks

• Be as inclusive as possible withoutgetting unwieldy

Expectations for realizing the goal of interoperability need tobalance the input of People, Process and Technology




IBM has made much progress helping our clients achieve

interoperability through various stages

Consortium

IntraOperability

BioSurvelliance / Public

InterOperability

http://www.connectingforhealth.org/index.html

http://www.ehealthinitiative.org/




Mark McCourtGlobal Solutions Sales LeaderClinical InteroperabilityIBM Healthcare & Life SciencesOffice (520) 575-9589mobile (520) 247-3042Email: [email protected]

http://www.ibm.com/solutions/healthcare




Aligning the Nature and Management of Health Care by maximizing thebenefits of information technology adoption

CDHM Conference, 07 Oktober 2011, BucharestBDent, Bart de Witte, MAppSc – IBM Central & Eastern Europe, Healthcare Industry Leader

IBM Healthcare




–

Outcomes & Knowledge




–

Medical Books




–

Medical Journals




–

Individual Knowledge




–

Organisational Knowledge




–

Knowledge about our Patients




–

Components of a System for Delivering Healthcare

health care is essent ial ly a proc ess of apply ing knowledg e to so lve pat ients'

heal th pro blems Health carepractitioners

Health careoutcomes

OrganizationsProcesses

of care

Knowledgefor

care

Knowledge ofPatientHistory




–

Evolution of knowhow & outcome – example diabetes Type 2

Source: NHS Choices - Healthguides care maps A-Z




–

The experimental nature of careThe distribution of uncertainty throughout a patient‘s care

Source – Prof. Dr. Richard Bohmer – Designing Care – Harvard Business Press 2008




–

The effect of learning on the clinical process

Peak of uncertaintyassociated with

diagnosis and treatmentselection is lowered

The time and number ofiterations required to solve anyindividual's health problem is

reduced




–

"evidence-creating medicine."

“High Performing Healthcare Organizations

simul taneous ly imp lement best pract ices for wel l -

understoo d health iss ues, create the new knowledge

required to so lve complex heal th problems, and embed

that knowledge into

next-generat ion cl inical pract ice:"

Richard Bohmer, Designing Care (2009)




–

Aligning the Nature and Management of Health Care by maximizing the benefits of information technology

adoption




–71 6/1/2014




–

Using Technology and Information systems to manage care

Source – Prof. Dr. Richard Bohmer – Designing Care – Harvard Business Press 2008




–

Over 80% of information today is unstructured and based on natural

language

1997




–

1997IBM Deep Blue beat G. Kasparov

2011




–

2011IBM Prepares to Send Watson Supercomputer

into Health Care




–

Business challengeGet Value out of your unstructured data

Objective:• IBM and University of Ontario Institute of Technology Collaborate With Canadian Hospital to Help Predict

Changes in Infants' Condition”.

Watson Analytics Healthcare Onramp Solution (WAHOS) Concept




–

Public

Private

WAHOSComplimentary

AnalyticsExpand Watson

Usage andValue

Standalone Solution Concepts1. Advance clinical decision making, accessing

risks and benefits of treatment in real time.

2. Greater insight and understanding into clinicalresults and medical research.

3. Optimize the process and improve outcomeaccuracy of identifying the disease or disorder.

Reduced mortality ratefor cardiac surgery

from 3.8% to 1.7%

360 view of patient'streatment patterns and

outcomes

Unlocking bio-medicalinformatics answers in

hours, not weeks

WAHOSCore Use Cases• Diagnostic Assistance

• Clinical Treatment Effectiveness• Critical Care Intervention• Research for Improved Disease

Management

Treatment &DiagnosticAssistance

Watson HC

Internal Medicine: Harrison‟s, Cecil‟s, ACP, MERCK

Medical Dictionaries: Stedman‟s, Taber‟s, Jablonski‟s

General Resource: National Clearinghouse, UMLS

Specific Resources: American Heart Ass.,

Journals, Magazines: NEJM, JAMA, BJM,, Lancet

Web Resources: NIH, NIM, Wikipedia, WWW

Books

Clinical

Guidelines

Journals, Magazines,

Web Resources

MoreReliable

MoreRecent

Watson Analytics Healthcare Onramp Solution (WAHOS) Concept

Public

Private

CustomerData &

Systems

77

http://www.bjc.org/?id=1&sid=1




–

Watson for HealthcareHow to best treat HIM or do we just treat an average 30 year old Afro-American?




–

Analyze the PastUse Content Analytics to understand

patterns, trends, deviations, anomalies,context and more from large corpuses of

information

Why Adding Retrospective, Prospective and Predictive Analytics to

Watson Makes Sense

Question What is Known

Use DeepQA to get confident accurateanswers from trusted knowledge base.

Predict the Future

Use Predictive Analytics to makemore informed decisions through

predictive and future businessscenario modeling

See the Present

Analyze and extract text from in-processdocuments or other content with

Content Analytics … feed the results toother systems.

Health professionals need to determine

what happened in he past

Health professionals need to know what

is happening now

Health professionals want to

optimize and predict outcomes

Health professionals want

confident and accurate

answers to questions

79




–80

The ChallengeSequoia Hospital wanted to gather patient-specificintelligence on the effectiveness of its cardiovasculartreatments and protocols but was hampered by toolsthat couldn’t adequately manage the quantity andcomplexity of the data.

The Solution

Sequoia Hospital replaced its previous solution withIBM SPSS Statistics, enabling the hospital to reshapeits treatment protocols and cut its overall mortalityrate from 2.9 to 1.3 percent.

Smarter Business OutcomesAutomated statistical and analysis tools have made itpossible to easily manage and analyze surgical andtreatment data on 14 years of data on 10,000 cardiac

patients for better patient outcomes.

Reduced mortality associated with cardiac surgeryfrom 2.9 percent to 1.3 percent.

The ChallengeExisting Biomedical Informatics (BMI) resources weredisjointed and non-interoperable, available only to asmall fraction of researchers, and frequently redundant.No capability to tap into the wealth of researchinformation trapped in unstructured clinical notes,diagnostic reports, etc.

The SolutionCapitalizing on the untapped, unstructuredinformation of clinical notes and reports by using IBMContent Analytics with IBM InfoSphere Warehouse.

Smarter Business OutcomesResearchers now able to answer key questionspreviously unavailable. Examples include Does thepatient smoke?, How often and for how long?, If smoke

free, how long? What home medications is the patienttaking? What is the patient sent home with? What wasthe diagnosis and what procedures performed onpatient?Locating answers in hours, not weeks or months … a

“gamechanger”

Case Study 1 & 2: IBM Providing Analytics Solutionsto Healthcare Today

http://www.sequoiahospital.org/Medical_Services/Heart_and_Vascular_Institute/index.htm

http://www.sequoiahospital.org/Medical_Services/Heart_and_Vascular_Institute/index.htm




–

Case Study 3

University of Ontario Institute of Technology, Neonatal intensive care units (NICU) atToronto's Hospital for Sick Children and two other international hospitals




–




–

IBM Personalized Healthc are 2010 - Omics and Molecular

Medicine are a Major Force for Shaping the Future ofPersonalized MedicinePersonalized Health Care

TranslationalMedicine

Health CareToday Digital Imaging

Episodic Treatment Electronic Health Records Artificial Expert Systems

Clinical Genomics

Genetic Predisposition Testing

Molecular Medicine

Diagnosis

Pre-symptomatic Treatment

Lifetime Treatment

Evolutionary Practices

R e v o l u t i o n a r y T e c h n o l o g y

AutomatedSystems

Non-specific(Treat Symptoms)

InformationCorrelation

1st Generation

Diagnosis

Organized(Error Reduction)

Personalized(Disease Prevention)

D i s t r i b u t e d H i g h - T h r o u g h

p u t A n a l y t i c s

Data and Systems Integration




–

Use Cases We Simultaneously Expand the Value of Watson

and our Existing Portfolio of Analytics OfferingsWAHOS ―Land‖ Use Cases

Diagnostic Assistance (DA),Clinical Treatment Effectiveness (CTE),Critical Care Intervention (CCI),Research for Improved Disease Management(RIDM)

Watson for Healthcare Focus Use Cases

Patient Inquiry (PI),Patient Workup (PW),Differential Diagnosis (DD),Treatment Protocol Options (TPO),Automated Treatment Authorization (ATA),

Treatment Protocol Analysis (TPA),Second Opinion (SO)

* Retrospective to Prospective to Predictive Care Management

Retrospective Predictive

Watson for HC

WAHOS

DD

PW

TPO

DACTE

CCI

RIDM

PI

Prospective

TPA

ATASO

86




Healthcare and L ifesc iences – Changing the World …Everyday

China Lifesciences Partnering Forum

January, 2006

Dr. Caroline A. Kovac

General Manager

IBM Healthcare and Life Sciences




The Next 10 Years

We are at the beginning of a period of transformational change in human healthcare

The ‗tipping point‘ will be accelerated by:

Changing demographics New economic models

Technological innovation

Globalization

The transformation of healthcare in the next ten years will threaten some existingbusiness models, but offer extraordinary opportunities as well.




The Human Genome – A Starting Point Not an Endpoint

But now the real challenge – translating this base of

knowledge into meaningfuldiagnostics and innovative

new therapies and treatments.

The l i fe sciences, pharmaceut ical and healthcare

systems mu st und ergo fun damental , st ructura l changeto achieve the prom ise of personal ized

medicine….Information technology will play a decisive,t ransform at ive role.

New Drivers of R&D




New Drivers of R&D

R&D

Manufacturing

Biologics

Technology Change

Systems Biology

Regulatory Change

Drug/indication>Mutation

Therapeutics>Theranostics (Rx/Dx)

Market Place ChangeGenerics

New Markets(obesity/memory/aging)>Old Markets(cancer/CV)

Personalized>Predictive>Preventative

medicine

Source: Burrill and Company, 2005

Challenges in the Life Sciences In




Safety-Based Drug Withdrawals

Combined peak sales was in excess of $11 Billion

Sources: Goldman Sachs, Med Ad News

Neurontin ®

Increasingly Rapid Generic Erosion

Pipeline Failures and Delays

Blockbuster US Patent Expiries, 2002-2007

VIOXX

2004

Challenges in the Life Sciences In

http://www.taxol.com/index.html

http://www.prozac.com/home.jsp




Cost of Development has more than doubled between 1980

and 1990

$4,300

$4,893 $4,838

$5,322

$5,860 $5,733

$5,891

$6,454

$6,924

$0

$1,000

$2,000

$3,000

$4,000

$5,000

$6,000

$7,000

1989 1990 1991 1992 1993 1994 1995 1996 1997

Cost per Patient in Phases I-III* ClinicalTrials, 1989-1997

*Out-of-pocket investigator and central laboratory cost per patient.

Source: Tufts, CSDD “Outlook2001”; Parexel Pharma R&D Sourcebook 2001, p. 113

$318

$802

$0

$100

$200

$300

$400

$500

$600

$700

$800

$900

1980 approvals 1990 approvals

Average Cost to Develop a New Drug, fromDiscovery to Approval ($millions)

CAGR 5.4%




Market Cap Comparison: Biotech vs Big Pharma

Company 9/05 12/04 12/03 9/02 12/01 12/00 12/99

Pfizer 183 199 280 192 251 290 124

J&J 188 184 154 112 181 146 129

Merck 60 69 103 165 133 216 153

Eli Lilly 61 65 77 50 88 105 75

BMS 47 47 58 65 112 145 125

Pfizer/Merck 243 268 383 357 384 506 277

Total USBiotech

466 399 342 213 366 425 312

Industry 1.9x 1.5x 0.9x 0.6x 1.0x 0.8x 1.1x

Source: Burrill and Company, 2005

Challenges in the Healthcare Indus



© 2010 IBM Corporation True North, Advisory Board Council, Advancing the Patient Safety Agenda, 2004www.healthalliant.com

:

Bureau of Economic Analysis 202-606-9900CMS 877-267-2323 Cms.gov/statistics/nhe

Challenges in the Healthcare IndusPatient Safety

Cost

Compliance

Productivity

Ag ing Populat ion in the US wil l increase the number




Ag ing Populat ion in the US wil l increase the number

of patients in Heart Disease

BBBB

AGE WHENINCIDENCE OFCHD INCREASES

AMERICAN HEARTASSOCIATION 2002

Many New Players in an Emerging Market




y y g g

Health andWellnessMarket

Employers andGovernment



Healthcare and Life Sciences are part of the same ecosystem




Healthcare and Life Sciences are part of the same ecosystem

Basic Researchand Discovery

Pharmaceuticaland Biotech

Diagnosticsand Medical

Devices

Academic Medical Research Centers

Integrated Care Delivery Networks

Community Hospitals and Practices

Payors

Private, Government, Employers

Life Sciences Healthcare Providers

PATIENT

Information Based Medicine




A system of medical care which supplements traditionalopinion-based diagnoses with new insights gleaned throughcomputerized data acquisition, management and analysis.

The goal of Information-Based Medicine (IBM) is toimprove treatment outcomes by improving the accuracy ofdiagnostic decisions.

Information Based Medicine will require unprecedented




Access to DiverseHeterogeneousDistributed Data

Expression Arrays(various tissues)

Personalgenomics

X-rays, MRI,

mamograms,etc

Clinical Record

Analysislab notes

Hospital events ....admission,surgery, recovery, discharge

1.PatientInformation

q paccess to diverse, integrated information

Challenges• Volume and complexity of data

• Integrating massive volumes ofdisparate data•Need for sophisticated analytics•Growing collaboration across ecosystem




The tools of 'Systems Biology'

Sequenceassemblyalgorithms

Data miningand patterndiscovery

Molecularmodeling andstructure

prediction

Complex celland organsimulation

# SEQLET# SEQLET

1 G..G.GK[STG]TL

2 H.....HRD.K..N

3 SGG[QEMRY]..R[VLIA].[IGLMV]R.L4 V.I.G.G..G...A

6 G.GLGL.I

...

T t d T t t




TPMTGenotypeTest forLeukemiaPatients

CYP Test todeterminedrugmetabolismHER2 Test

+

Rise of Dx/Rx Bundles―Theranostics‖

Targeted Treatments




NationalDigitalMammographyArchive

North CarolinaBioGrid

eDiamondProject

Smallpox GridProject

Projects in Grid based medical research

Medical Imaging is a Growing Market




Medical Imaging is a Growing Market

Nuclear M ed

Expanding role of

PET in oncology

and radiation

oncology

Hospit al End Users

Medical Imaging Moving

Rapidly beyond

Rad io l ogy Depa r tm en t

Orthopedics

Use of novel imaging

design I.e. stand-up

MRI

Women's Heal th

Increasing focus onbreast and bone

screening

Cardiologists

I/T Cardio and

Cardiac CT, MRI

and Ultrasound

US imagingmarketgrowing 10-12%

Chinaimagingmarket

growing25%+

Mobile Health Monitoring: Solutions for Chronic Disease




Management?

•Chronic disease – diabetes,congestive heart failure, etc.

– accounts for 75% ofmedical costs in the US

•Disease management cansubstantially improve patient

outcomes and lower cost

•Medical devices andtelecommunications

technologies are converging

The ―global-ness‖ of Biotech




Science/technologyIntellectual property/patents/FTO

People

Communications

Competition

Capital

Markets—diseases know no borders

Even the smallest biotech is a globalplayer

Convergence of IT and Biology




Scientific discovery

New drugs and treatments

Revolution in healthcare

Life Sciences Information Technology




Second US-China Computer ScienceL d hi S it



Harbin Institute of Technology

Computer Science andBioinformatics

Wang Yadong

Leadership Summit

Computer science: a multidisciplinary science




Seeking to develop intelligent machine


Chip(2 processors)

Compute Card(2 chips, 2x1x1)

Node Board(32 chips, 4x4x2)

16 Compute Cards

System(64 cabinets, 64x32x32)

Cabinet(32 Node boards, 8x8x16)

2.8/5.6 GF/s4 MB

5.6/11.2 GF/s0.5 GB DDR

90/180 GF/s8 GB DDR

2.9/5.7 TF/s256 GB DDR

180/360 TF/s16 TB DDR

Blaise Pascal:1642 Charles Babbage:1833 ENIAC:1944

IBM: Blue

Gene A massivelyparallelsupercomput

er

Fig.from www.sciencemuseum.org.uk

From:http://www.resear

ch ibm com





• Seeking to make machine more intelligent


―Boolean logic

―Finite state machines

―Formal grammars

―Turing machines

―Recursion

―Garbage collection

―Complexity theory

―Machine learning

The emerging multidisciplinary research in C S




114

Grid computingPervasive computingintelligent searchingService computing

…

Human Genome Map

Genome informaticsBioinformatics

The emerging multidisciplinary research in C.S

Internet and WWW(World Wide Web)

from http://discovermag

From http://www.fotose

Bioinformatics on demand




115

Bioinformatics on demand

Challenge• Genome Annotation• Sequence Assembly• Physical Mapping• Protein folding / Docking• Protein-protein interaction• Finding Binding Site Motifs

• Transcriptional Regulation• Post-transcriptional Regulation• Histone modification• DNA methylation• Drug discovery

Then the ―omics‖

revolution

• Genomes• Proteomes

• Transcriptomes

• Epigenetics(-omics)

• “Whatever”omes (...?)

• Human Genome Project

• GenBank database• EMBL database• BLAST

• Al orithm for hi h-

Bioinformatics

Wh t‘ t?




116

What‘s next? • From ―omics‖ to system biology • From bench to bedside

• From data to knowledge

• Computational System biology

• Translational bioinformatics• Bio Knowledge engineering

Computational system biology: on demand




Computational system biology: on demand

Challenge: how to construct descriptive and predictive

models and interacting systems from the “omics” data.

Gene regulatory networkProtein-Protein

interaction networkMetabolic Network

Signaling network

Cell Simulation

Key Areas: represenattion, visualization, modeling, networks ofbiological entities, and construct interacting systems .

Translational bioinformatics



© 2010 IBM CorporationPage 118

The Human Genome – A Starting Point Not an Endpoint.But now the real challenge – translating this base ofknowledge into meaningful diagnostics and innovativenew therapies and treatments, and to ach ieve the

promise of personalized medicine….Informat iontechnology w il l play a decis ive, transfo rmative role.





Challenge 2: Reducing the knowledge gap






The goal of translational bioinformatics is to enable the

transformation of increasingly voluminous genomic andbiological data into diagnostics and therapeutics for the clinician

Key research areas:• Data : Integration of increasingly voluminous biologicaland clinical data•Knowledge: Newly found knowledge from theseintegrative efforts that can be represented, stored,retrieved, and disseminated• Algorithm: To integrate genetic, genomic, proteomic,animal model, and clinical measurements are needed forthe next step

A human phenome-interactome network of protein




A human phenome-interactome network of proteincomplexes implicated in genetics disorders

The Bioinformatics In Harbin Institute of Technology




gy

2001 established the Bioinformatics Division,

2001 Set up the bachelor program. This program provides three

areas of study

• computer science• molecular biology

• bioinformatics

2001 established the center for bio-computing, research focus on

•Biometrics and Computer aided traditional

Tongue diagnoses system



© 2010 IBM CorporationPage 125

Tongue diagnoses system

Pulse diagnoses system

Computer aidedtraditional Chinesemedicine diagnoses

Bioinformatics in HIT




MiR2Diseas

e: Databaseandpredicationsystem

MotifModeler

:Identifytranscriptionfactor and

Gene expression levels

Transcription factors

Transcriptional initiation

5’ upstream regulatory region

microRNA-inducedmRNA degradation

mRNA degradation

3’ un-translated region

Bioinformatics in HIT




Gene Mining: A Novel and Powerful Ensemble

Decision Approach to Hunting for Disease GenesUsing Microarray Expression Profiling.

Virtual heart Petri nets based e cell

Bioinformatics Technology supported by




128

gy pp yNational High Technology Program Supported

Seven Research Directions: 2001-2005)

Technology for Bio-Data Acquisition and Mining

Technology for Bio-Data Application

Structure Genomics and Proteomics

Molecular and Drug Design

Bio-Chip

High-Throughput Drug Screening

Novel Drugs

New Budget Launch Into Bioinformatics Field from




129

the 863 program 2006-2010)

Topic Program : Bioinformatics and Bio-computing

Key Program: Drug design and Molecular design

National Program:

– Functional genomics and protoemics

– Major disease related molecular genotype and personal

health care

Total budget: ~70 million USD



IBM Content Analytics

Delivers



© 2010 IBM Corporation132

Delivers:

…Deep, valuable business insight extracted fromunstructured content

…Linguistic modeling and concept extractionthrough classification, NLP and text analytics

…Dynamic, secure analytics-driven search withaccurate and timely results

…extract value and business success from

unstructured content with Content Analytics!

Analytics Text Miner

Organize, Analyze and Visualize structured, semi-

structured and unstructured enterprise content Identify trends, patterns, correlations, anomalies

and business context from collections.

Natural Language Processing (NLP)

Utilize linguistic modeling to enable semantic /entity / concept extraction from business content

Employ advanced classification to categorizecontent deliver business insight

Secure Enterprise Search

Crawl, index and analyze data and content forsecure enterprise search

Search and explore to derive business value

Business Integration Integrate Content Analytics with business

processes and other analytics solutions to deliverfull business context.

Enterprise

Search

Business

Integration

TextAnalytic

Content

Analytics




CustomerService

ProductManagement

ServiceManagement

Marketing

Churn Alerts

Monitor FAQsMarket Sentiment

CSR LogsMarket

ResearchTranscripts Web Blogs

Email

Sales

Partners

IndustryReports

Internal Docsand Reports

Corporate Reputation

Voice of the Customer CSR Training and Monitoring

ICA Delivers Insight to Multiple Lines of Business




IBM Content Analytics adds value to…

Automotive Quality Insight

• Analyzing: Tech notes, call logs, online media•For: Warranty Analysis, Quality Assurance

• Benefits: Reduce warranty costs, improve customersatisfaction, marketing campaigns

Crime Analytics

• Analyzing: Case files, police records, 911 calls… • For: Rapid crime solving & crime trend analysis

• Benefits: Safer communities & optimized force deployment

Healthcare Analytics

• Analyzing: E-Medical records, hospital reports

•For: Clinical analysis; treatment protocol optimization

•Benefits: Better management of chronic diseases;optimized drug formularies; improved patient outcomes

Insurance Fraud

• Analyzing: Insurance claims• For: Detecting Fraudulent activity & patterns

• Benefits: Reduced losses, faster detection, moreefficient claims processes

Customer Care

• Analyzing: Call center logs, emails, online media

•For: Buyer Behavior, Churn prediction •Benefits: Improve Customer satisfaction / retention, marketing campaigns, new revenue opportunities

Social edia for arketing

• Analyzing: Call center notes, SharePoint, multiple

content repositories• For: churn prediction, product/brand quality

• Benefits: Improve consumer satisfaction, marketingcampaigns, find new revenue opportunities or product/brand quality issues




Dynamically search and explore content fornew business insight

New Connections and Dashboard views to easily detectinsights; plus add your own custom views

Interactively assess for content preservationand decommissioning to reduce storage costsand risk

Powerful solution modeling and support foradvanced classification tools for more accurateand deeper insight

Enhanced analytics configuration tools

Deliver rapid insight to other systems, usersand applications for complete business view

Quickly generate Cognos BI reports, link betweenCognos reports and ICA views; deliver analysis to IBM

Case Manager solutions

IBM Content Analytics v2.2 HighlightsA platform for rapid insight

Search and Content Analytics: How it works




UIMA Pipeline + Annotators

136

…”car” that “smelled” “like smoke” and

had “half a tank of gas” ...

Source InformationCorporate (Contact Center, Test Data,Dealer notes, ECM, etc.) and External (NHTSA, Edmunds, Consumer Reports,

MotorTrend etc.)

Noun Verb Noun Phrase Prep Phrase

Vehicle Scent 1st issue 2nd issue

Primary Issue: “Odor” Related Issue: “Fuel Level”

ExtractedConcepts

Analyzed Content(and Data)

Fine grain control over the entities and facets that are created

IBM Master Data Mgmt

RDB

Tailor your text analysis with flexible, easy-to-use tooling




1 Develop your Custom Text Analysis with ToolingBuild language and domain resources into a LangaugeWare dictionary.

Develop rules to spot facts, entities and relationships.Create and test UIMA annotators with a collection of documents.

2 Export your Custom Text AnalysisEasily generate the annotators to be Content Analytics ready

3 Deploy your Custom Text Analysis with in ICAImport newly created annotators via Content Analytics administration console andassociate it to a collection.

View ofProject Resources

Easy to test and verifyyour tailored text analysis

Easy to exportyour custom text

analysis




Text Analytics / Natural Language Processing (NLP)

The simplest text analytics scenario is to scan a set of documents written ina natural language, then:

• model the document set for predictive classification purposes,or• populate a database or search index with the information extracted.

Text analytics also describes that application of text analytics to respond tobusiness problems, whether independently or in conjunction with query andanalysis of fielded, numerical data.

It is a truism that 80 percent of business-relevant information originates in

unstructured form, primarily text.

These techniques and processes discover and present knowledge – facts,business rules, and relationships – that is otherwise locked in textual form,impenetrable to automated processing.




Text Analytics

The term text analytics describes a set of techniques: – Linguistic – Statistical – Machine learning

These techniques model and structure the information content of textual

sources for – business intelligence – exploratory data analysis – research – investigation.

The term is roughly synonymous with text mining

Text Analytics is now more frequently in business settings while "textmining" is used in – life-sciences research – government intelligence.




Text Analytics

Text analytics involves – information retrieval – lexical analysis to study word frequency distributions – pattern recognition – tagging/annotation

– information extraction – data mining link and association analysis – Visualization – predictive analytics

The overarching goal is, essentially, to turn text into data for analysis viaapplication of

– natural language processing (NLP) – analytical methods.

What is Text Mining?

Text mining technology has been developed to acquire




Text mining technology has been developed to acquireuseful knowledge from large amounts of textual data



Unstructured Information Mgmt Architecture (UIMA)




UIMA stands for Unstructured Information Management Architecture.

Structure of UIMA

The UIMA architecture can be thought of in four dimensions:

It specifies component interfaces in an analytics pipeline

It describes a set of Design patterns

It suggests two data representations: – an in-memory representation of annotations for high-performance analytics – an XML representation of annotations for integration with remote web services.

It suggests development roles allowing tools to be used by users withdiverse skills

C l a s s i f i c a t

i o n

C u s t o m A n a l y t i c s

T o k e n i z a t i o n

W o r d A n a l y t i c s

N a m e d E n t i t y

R e c o g n i t i o n

M u l t i - w o r d

A n a l y t i c s

L a n g u a g

e

I d e n t i f i c a t i o n

UIMA

http://en.wikipedia.org/wiki/Pipeline_(software)

http://en.wikipedia.org/wiki/Design_pattern_(computer_science)

http://en.wikipedia.org/wiki/XML

http://en.wikipedia.org/wiki/XML

http://en.wikipedia.org/wiki/Design_pattern_(computer_science)

http://en.wikipedia.org/wiki/Pipeline_(software)




ICA, NLP, LRW & UIMA?annotator- a software component that performs linguistic analysis tasks and produces and recordsannotations




Annotators

IBM® Content Analytics provides a number of UIMA annotators for advanced text analysis.

When documents are processed through the document processing pipeline, the annotators extractfrom unstructured content – Concepts – Words – Phrases – Classifications

– named entities The annotators mark these extractions as annotations.

The annotations are added to the index as tokens or facets

The annotations are used as the source for content analysis.

Some annotators support user-defined dictionaries, user-defined rules, and customconfigurations.

When configuring the document processing pipeline for a collection, an administrator selects theannotators to be used. Some of the key functions the annotators support include: – Populating the common analysis structure to a relational database with specific text analysis results. – Capturing special words of interest as the subject of text analytics. – Capturing patterns of words as the subject of text analytics. – Capturing named entities, such as persons, places, and organization names. – Categorizing documents. – Fundamental text analytics, such as parsing content to identify parts of speech. – Multilingual text analytics capabilities. The results of analytics can vary based on the language of the

input document.

LRW Develops UIMA annoators, isn't that enough?




Developing a model is an iterative process between LRW andICA

ICA




ICACrawlers

ICADocumentProcessing

ICAText

Miner

ICACustom

AnnotatorConfiguration

ICAis the platform

for findinginsights using

the annotations!!




Communities

On-line communities, User Groups, Technical Forums, Blogs, Social networks, andmore

– Find the community that interests you … • Information Management ibm.com/software/data/community • Business Analytics ibm.com/software/analytics/community • Enterprise Content Management ibm.com/software/data/content-management/usernet.html

IBM Champions – Recognizing individuals who have made the most outstanding contributions to

Information Management, Business Analytics, and Enterprise Content Managementcommunities

• ibm.com/champion

http://www.ibm.com/software/data/community


http://www.ibm.com/software/analytics/community


http://www.ibm.com/software/data/content-management/usernet.html


http://www.ibm.com/software/data/champion















Let‘s Build a Smarter Planet Changing Landscape in Health Care Delivery:

The Opportunity and Possibilities

Chicago

November, 2009




Focus onValue

Global trends and their implications

Information

liquidity andtransparency

Escalating Costs

Smarter Healthcare

http://www.treehugger.com/us-money-photo.jpg




+ + =

Improve

operat ional

effect ivene

Deliver

col laborat ive

care for

Achieve

better qual i ty

and

A smarter heal th system fo rges partnersh ips in o rderto del iver better care, predic t and p revent dis ease and

empower indiv iduals to make smarter choices.

IntelligentInstrumentedInterconnected

Solution focus and investments




Healthcare Solution Ar

Content, Images, Records, FormsLifecycle Management

DataIntegration

Public ReportingQuality Standards

Drug / Patient SafetyChronic Disease Management

QualityOutcomes

OperationalEfficiency

DataManagement

Business Process andWorkflow Optimization

Health Integration Framework (SOA, Data Management and Business Process Framework)

Collaborative careand wellness

Better quality

and outcomes

Operationaleffectiveness

Health AnalyticsClinical Quality Management and Performance

Member and Care Management AnalyticsPredictive and Preventive Care

Patient-Centered Collaborative CareHealth Information Exchanges

Portals for Care Teams, Patients, MembersEnterprise mobility, Public Health

Infrastructure OptimizationClaims optimization and ICD-10 compliance SOA

for flexible business processesPrivacy, security and business resilience

Application management and hostingStorage virtualizationData center optimization & energy management

We‘re enabling connectivity and integrationof medical device information

http://www.s3group.com/uploads/RTEmagicC_continua_logo_96dpi_margins.jpg.jpg




PHR

IBM Continua

ReferenceImplementation

Devices Device Manager

PHM Report

TelehealthService

Wired Weighing Scale

Independent Living Hub Medication Monitor

Wired Glucose Meter

Wired Glucose Meter

Care Management

care management platform for monitoring

chronic patients andsupporting behavior

change

aggregate data andmap into Google Health

formatcollect device data

* Device Mgr also

Avail on IBM Platform

Third PartyDevice

Manager

Portals for connecting Patients

http://images.google.com/imgres?imgurl=http://www.trustedreviews.com/images/article/inline/5405-IMG7832s.jpg&imgrefurl=http://www.trustedreviews.com/notebooks/review/2007/09/10/Dell-Latitude-D531/p4&h=430&w=500&sz=55&hl=en&start=5&usg=__6EMYuWER9fnbRdQQ6fkUWkxI4yM=&tbnid=bgNEJHWYZ9A5bM:&tbnh=112&tbnw=130&prev=/images?q=Intel+PC&gbv=2&hl=en




NCQA Diabetes Scorecard




Anatomical representation of patient and health information




Cli i l T i l Pharmaceuticals beginning to explore how to

http://www.flickr.com/photos/ibm_research_zurich/3811882716/in/set-72157621885848049/




Clinical Trialsg g p

leverage social networks – Emerging trend

Complements previous scenario – Similar integration and deployment patterns

http://www.newsweek.com/id/187882/page/1

161Copyright IBM - 2009

World Community Grid

•Launched in Nov ‘04






•Mobilize the community•Technology solving problems

•Advisory Board of experts in healthsciences, technology and philanthropy•RFP for new projects

•Human Proteome Folding•FightAIDS@Home•Help Defeat Cancer

•Help Cure Muscular Dystrophy•1,350,000 computers•271,000 years of computation

Building a smarter planet …




Teva USA: Improved product visibility and authenticity toensure a safe and secure pharmaceutical supply chain.Enhanced operational efficiency by reducing manual processes.

Honda Italia: Implemented RFID component trackingsolution to improve production efficiency and quality.

DHL: Developed and implemented a real-time monitoringsystem using RFID and sensing technology to monitortemperature sensitive shipments while in transit.

Airbus: In a strategic move, RFID is being utilized toautomate the tracking of aircraft segments while improvingsupply chain visibility and reducing costs.

Matiq: Employs RFID tags to trace meat and poultry fromthe farm to store shelves to ensure safety and freshnessand provide more transparency to consumers.

Metro: Using RFID technology throughout its entire supplychain to help them get the merchandise the customers wanton the shelves when they want them.




D i i h i th h lth t

If we could make a step change in just the




165

Driving change in the healthcare system If we could make a step change in just thetreatment of chronic conditions like

coronary artery disease, congestive heartfailure and diabetes, we‘d attack 80 percentof all healthcare costs

Benefits of optimising data – Integrates patient information from electronic

medical records, claims, medication, lab dataand clinical decision support

– Allows for Evidence based Medicine

– Enables Chronic Disease Management – preventive treatments before the disease hits

– Doctors can deliver more complete and accuratedecisions, reducing medical mistakes and

unnecessary tests and treatments – Gives healthcare organisations access to all the

tools and services without making big individualinvestments

– Could alert people to catastrophic medicalevents before they happen




Using (Information) Technology to Improve PatientOutcomes John Crawford, Healthcare Industry Leader, IBM Europe

West of England AHSN First Annual Conference

17 September 2013

Around the world, the transformation of care systems is under way…enabled by information technology




Focus on the individual

Access to unprecedentedamounts of data createsan opportunity for deeperinsight and earlierintervention and engagement

Emphasis on value

Buyers of care (includinggovernments, private insurers,employers and individuals)expect greater value,improved quality and betteroutcomes – at a moreaffordable cost

New care modelsIncreasing demand foraffordable health and socialcare services is driving theformation of new models andpartnerships

Big Data/Analytics

Personalisedmedicine

Targeted socialinterventions

Health apps andconnected devices

Flat healthcarebudgets

Reduced coverage

More transparency

• Safety

• Quality

• Outcomes

Local Government+ Academia +Care Providers +Industry

Social Enterprises

New private sector

entrants

• Telco, utility, retail

• Start-up / Venture



Prescriptive Analytics is moving us towards predictionand optimisation of outcomes




Foundational

• What happened?

• When and where?

• How much?

Advanced, Predictive

• What will happen?• What will be the impact?

• Dashboards

• Clinical data repositories

Data integrationData warehouse

• Basic reporting

• Spreadsheets

Transactionreporting

• Enterprise analytics

• Evidence-based medicine

Decision supportanalytics

• Personalised healthcare

• Population risk models

• Optimising care systems

Predictiveanalytics

• What are potential scenarios?

• What is the best course?

• How can we pre-empt and

mitigate risk?

Applying new insights from lifestyle, social and clinical data toassess and anticipate population health needs




PreventionEngage citizens toimprove health literacyand inform lifestylechoices to prevent theonset of health issues

• Deliver evidence-

based healthinformation usingsocial media andmobile platforms

• Improve inadequatehousing to preventasthmaexacerbations

Segment populations by risk profiles

Base care approaches on evidence

Healthy low risk At risk High risk Early clinicalsymptoms

Active disease

Care ManagementDeliver the right care services to support theindividual with the right programs and treatmentsto improve quality of life and optimise resources

• Prevent admissions and readmissions by

sensing deterioration with telecare &telehealth services, and use lower cost careenvironments

• Provide support for rehabilitation and returntowork programs

Early interventionPromote routine screening and coaching forhealthy lifestyles to defer disease onset andmanage risk

• Build population risk stratification models,

and use these to target interventions moreeffectively

• Enroll individuals and families at high risk fordiabetes in lifestyle programs E

x a m p l e

s

Personalised medicine is becoming a reality




Access to Clinical Knowledge(e.g. Diagnostic tools, Comparative Effectiveness)

Access to

PatientInformation

PoorPoor

Good

Good

Individual clinician knowledge & experience

Clinician consensus

Evidence-Based(Based on large-scale trials)

Personalised(Based on „people like me‟)

More art than science

More science than artSource: IBM Global Business Services and IBM Institute for Business Value

Protocol-Based(Based on best practice)



University of OntarioInstitute of Technology(UOIT) uses Big Data toimprove quality of care for




p q yneonatal babies

Need

• Performing real-time analytics usingphysiological data from neonatal babies

• Continuously correlates data from medicalmonitors to detect subtle changes and alert

hospital staff sooner

• Early warning gives caregivers the ability toproactively deal with complications

Benefits

• Detecting life threatening conditions 24hours sooner than symptoms exhibited

• Lower morbidity and improved patient care

173173

EuResist project uses BigData to predict response ofHIV patients to treatmentbased on viral genomics




g

174

Need

• Predicting best treatment strategy for HIVsufferers from a cocktail of drugs, comparingnew cases with a database of over 61,000previous cases, over 150,000 therapy optionsand nearly 700,000 viral loads

Benefits

• Using decision support delivers 78% accuracyin treatment plans

• EuResist prediction engine outperformed 9 outof 10 human experts in predicting outcome

A new set of digital transformation drivers for healthcare isemerging

Mobile Revolution




Speed / Immediacy

Always On Location Aware

Social Media Explosion

Sharedknowledge

Connectedness

Hyper Digitization

Streaming video

Sensor networks

The Power of Analytics

Intelligence

Optimalresponse

Connected healthcare devices and mobile applicationsWe see users in the health device marketsplit into three main segments




split into three main segments

Current product offerings have focused onthe needs of the Motivated Healthy andChronically Monitored segments

Sources: IBM Institute for Business Value Analysis

MotivatedHealthy

ChronicallyMonitored

Informationseekers

Increasing sophistication of wireless devices for health andfitness – improved design & integration with systems to trackpersonal goals

Motivated

Healthy




personal goals

Apple and Nike

Combines running and music for thefitness market. Smartphoneapplications collect data from

sensors in fitness equipment,enabling users to monitor theirperformance

Fitbit

Monitors activity with motion andprovides online data storage andanalytical tools. Does not provide

ability share data outside of itswalled garden

Nike+ FuelBand

3-Axis sports-testedaccelerometer, monitors activityas steps, calories and Nike Fuel,

which can be uploaded withNike+ Connect to set goals,collect rewards

Telehealth and care coordination to improve management ofchronic conditions

Chronically

Monitored

Patient – daily health surveyand vital signs recording

Care providermonitors results




Shared Carecoordination

PrimaryCare (GP)

CommunityCare

HospitalCare

Patient

Remote Monitoring

All data sentto a secure

server

Wider care teamarrange any actionrequired

Service deskprovides first linesupport

and vital signs recording monitors results

Video Conference

Determinants of poor health (60% of disease burden)*

Information

seekers




Determinants of poor health (60% of disease burden) Hypertension

Tobacco use

Harmful use of alcohol

High serum cholesterol

Overweight

Unhealthy diet

Insufficient physical activity

Can technology help improve

health through behaviouralchange?

* WHO European Health Report 2009

Interest in health care devices is shifting towards preventivecare and communicating changes in health status

Information

seekers




Measure and managea known health

problem

Encourage physicalactivity

Inform others of someone‟schanging health

condition

20% less

30% more

100% more

51%

41%

13%

17%

5%

10%

While half of currentdevices in usetoday are for knownhealth problems…

…interest in usingapps and devicesfor maintaininghealth is expectedto rise significantly

Source: IBM Institute for Business Value, 2009 Connected Health Devices Survey (Current Devices: N = 1256, Total mentions = 1507; Future Devices: N=604, Total mentions = 503)

Current

Future



IBM Watson – cognitive computing to improve healthcare outcomes

http://localhost/var/www/apps/conversion/1308/Digital%20Health%20Days/Materials/IBM%20Watson%20and%20WellPoint%20%20a%20progress%20report%20-%20YouTube.flv




Understands natural languageand physician / patientcommunication

Adapts and learns from userselections and responses

Generates and evaluates evidence-based hypothesis toimprove quality of patient care

1

2

3

IBM Watson – Your future health advisor?

http://localhost/var/www/apps/conversion/1308/Digital%20Health%20Days/Materials/IBM%20Watson%20and%20WellPoint%20%20a%20progress%20report%20-%20YouTube.flv

http://localhost/var/www/apps/conversion/1308/Digital%20Health%20Days/Materials/IBM%20Watson2.flv

http://localhost/var/www/apps/conversion/1308/Digital%20Health%20Days/Materials/IBM%20Watson2.flv




http://localhost/var/www/apps/conversion/1308/Digital%20Health%20Days/Materials/IBM%20Watson%20works%20to%20transform%20health%20care%20-%20YouTube.flv



Watson and the Jeopardy! Challenge

Michael Sanchez




IBM DeepQA

What is DeepQA?




QA – Question Answering – Systems designed to answer questions posed in natural language

Goal – create a system capable of playing Jeopardy! at human championship level – In real time

IBM‘s follow up project to DeepBlue

p

Initial Performance




DeepQA Architecture




p

Content Acquisition




Take an initial corpus of documents – Unstructured data – For Jeopardy! – roughly ~400 TB of data – Including all of Wikipedia

Parsed into ―Syntactic Frames‖ – Subject-Verb-Object

Generalized into ―Semantic Frames‖ – Probability associated

Forms a ―Semantic Net‖

q

Question Analysis




Attempt to understand what the question is asking

Many different approaches are taken – Attempting to come up with all possible interpretation of the question

Hypothesis Generation




Primary Search – Attempt to come up with as much answer content as possible fromsources

– Various search techniques – 85% of time correct answer within top 250 at this stage

Candidate Answer Generation – Use appropriate techniques to extract answer from content

Filter – Lightweight scoring of candidate answers – ~100 answers let through

Hypothesis Scoring




Retrieve additional evidence supporting each candidate answer that passed filtering

Score the candidate answers based on supporting evidence – More than 50 different types of scoring methods – Ex. Temporal, Geospatial, Popularity, Source Reliability

Result Merging




Identify related answers and combine their scores

Generate confidence estimation – Indicates how confident in the answer the system is – System training is important here

• Different question types might weigh scores differently

– Probabilistic

Results are then ranked on confidence – Highest confidence = best answer

DeepQA Performance




DeepQA on Watson




With a single CPU - ~ 2 hours to get an answer – Not fast enough for Jeopardy! – Questions take ~ 3 seconds on average to read

Take advantage of the parallel capabilities of DeepQA

– 90 Power 750 servers = 2880 CPUs• 80 TFLOPS

– Able to answer in 3-5 seconds

Jeopardy! Challenge




In January 2011 Watson competed against two of the best Jeopardy! Champions – Ken Jennings – $3,172,700 in winnings – Brad Rutter - $3,470,102 in winnings

Two matches played

– Questions chosen from unaired episodes

Outcome




First Game – Watson wins - $35,734 – Rutter - $10,400, Jennings - $4,800

Second Game

– Watson - $77,147 – Jennings - $24,000, Rutter - $21,600

Future Applications




Take advantage of DeepQA‘s ability to process large amounts of unstructured Data

Medicine – Amount of data increasing doubling every 5 years – Almost entirely unstructured

Finance – 5 documents from Wall Street every minute – Millions of transactions



References




The AI Behind Watson: http://www.aaai.org/Magazine/Watson/watson.php

What is Watson?: http://static.usenix.org/event/lisa11/tech/slides/perrone.pdf

Building Watson: http://www.youtube.com/watch?v=3G2H3DZ8rNc

IBM Watson: The Science Behind an Answer:http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=6177717

Watson: http://en.wikipedia.org/wiki/Watson_%28computer%29

Question Answering: http://en.wikipedia.org/wiki/Question_answering

DeepQA Research Team:http://researcher.watson.ibm.com/researcher/view_project_subpage.php?id=2159

http://www.aaai.org/Magazine/Watson/watson.php

http://static.usenix.org/event/lisa11/tech/slides/perrone.pdf

http://www.youtube.com/watch?v=3G2H3DZ8rNc

http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=6177717

http://en.wikipedia.org/wiki/Watson_(computer)

http://en.wikipedia.org/wiki/Question_answering

http://researcher.watson.ibm.com/researcher/view_project_subpage.php?id=2159




http://en.wikipedia.org/wiki/Question_answering

http://en.wikipedia.org/wiki/Watson_(computer)

http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=6177717

http://www.youtube.com/watch?v=3G2H3DZ8rNc

http://static.usenix.org/event/lisa11/tech/slides/perrone.pdf

http://www.aaai.org/Magazine/Watson/watson.php




Be Ready for Watsonwith IBM Content and PredictiveAnalytics

201

Michelle Blackmer

Title: Be Ready for Watson with IBM Content and Predictive Analytics




IBM Content and Predictive Analytics (ICPA), IBM's first Ready for Watsonsolution offering, enables subject matter experts and other knowledgeworkers to explore information in order to analyze the past, understand thepresent and predict the future. This solution pairs IBM Content Analyticsand IBM SPSS Predictive Analytics to reveal new and actionable insights instructured and unstructured content; unstructured content accounts for80% of an organization‘s data. As a result, organizations can find more

effective ways to improve customer satisfaction, inform decision making

and increase operational efficiencies. Attendees will learn how SetonHealthcare is leveraging ICPA to reduce inconvenient and costly congestiveheart failure readmissions.

Session Agenda




Introduction to Watson and IBM Content and PredictiveAnalytics for Healthcare (ICPA)

Using ICPA to Prevent CHF Readmissions –Seton Healthcare Family Case Study

203

Disclaimer




4

Disclaimer

In form at ion regarding potent ia l future products is intended to out l ine our general

produ ct direct ion and i t should n ot be rel ied on in making a purchasing d ecis ion. The

informat ion ment ioned regarding potent ia l future produ cts is not a commitm ent,

prom ise, or legal obl igat ion to del iver any m ater ia l, code or fun ct ional i ty . Inform at ion

about potent ia l future produc ts may not be incorpo rated into any c ontract . The

development, re lease, and t iming of any futu re features or fu nct ional i ty d escr ibed for

our pro ducts remains at our sole discret ion.THE INFORMATION CONTA INED IN THIS PRESENTATION IS PROVIDED FOR

INFORMATIONAL PURPOSES ONLY. WHILE EFFORTS WERE MADE TO VERIFY THE

COMPLETENESS AND ACCURACY OF THE INFORMATION CONTAINED IN THIS

PRESENTATION, IT IS PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND,EXPRESS OR IMPLIED. IN ADDITION, THIS INFORMATION IS BASED ON IBM’SCURRENT PRODUCT PLANS AND STRATEGY, WHICH ARE SUBJECT TO CHANGE BY

IBM W ITHOUT NOTICE. IBM SHALL NOT BE RESPONSIBLE FOR ANY DAMAGESARISING OUT OF THE USE OF, OR OTHERWISE RELATED TO, THIS PRESENTATION

OR ANY OTHER DOCUMENTATION. NOTHING CONTA INED IN THIS PRESENTATION IS

INTENDED TO, NOR SHALL HAVE THE EFFECT OF, CREATING ANY WARRANTIES OR

REPRESENTATIONS FROM IBM (OR ITS SUPPLIERS OR L ICENSORS), OR AL TERING

THE TERMS AND CONDITIONS OF ANY AGREEMENT OR L ICENSE GOVERNING THE

USE OF IBM PRODUCTS AND/OR SOFTWARE.204

Inconsistent quality and increasing costsrequire healthcare transformation in key areas




* New England Journal of Medicine – Rehospitalization Among Patients in the Medicare Fee-for-Service Program, April 2009** http://www.healthleadersmedia.com/content/COM-263665/3-Readmissions-to-Reduce-Now

The needfor betterclinical

outcomes

•One in fivepatients suffer from

preventable readmissions … represents$17.4 billion of the current $102.6 billionMedicare budget*

•1.5 million patients in the

U.S. harmed annually by errors in theway medications are prescribed,delivered and taken

The need

for betteroperationaloutcomes

In 2012, Hospitals will

be penalized for high

readmission rates - Medicare dischargepayments starting will be reduced in keyareas**

•$475 billion: Estimated

annual US healthcare spending onadministrative and clinical waste, fraud,

abuse and other waste



In May, Gary arrived in India afterhe celebrated his anniversary in

In May 1898 Portugal celebrated the400th anniversary of this explorer’s

Why is Jeopardy! so difficult?

Answering complex natural language questionsrequires more than keyword evidence




explorer

India

In May

1898

India

In May

celebrated

anniversary

in Portugal

he celebrated his anniversary inPortugal

Portugal

400thanniversary

celebrated

Gary

208

400th anniversary of this explorer sarrival in India

This evidence suggests“ Gary ” is the answerBUT the system must

learn that keywordmatching may be weakrelative to other types of

evidence

arrived in

arrival in

Legend

Keyword “Hit”

Reference Text

Answer

Weak evidenceRed Text

requires more than keyword evidence

On the 27th of May 1498, Vasco daGama landed in Kappad Beach

In May 1898 Portugal celebrated the400th anniversary of this explorer ’s

i l i I di

Watson Leverages Multiple Algorithms to Gather DeeperEvidence




27th May 1498

Vasco daGama

landed in

arrival in

explorer

India

Para-phrases

Geo-KB

DateMatch

209

Stronger evidence canbe much harder to find

and score …

… and the evidence is stillnot 100% certain

Search far and wide

Explore many hypotheses

Find judge evidence

Many inference algorithms

Gama landed in Kappad Beach

400th anniversary

Portugal

May 1898

celebrated

arrival in India.

Kappad Beach

LegendTemporal Reasoning

Reference Text

Answer

Statistical Paraphrasing

GeoSpatial Reasoning

• How are you measuring and

... but the biggest blind spot still remainsDoes unlocking the unstructured datahelp accelerate your transformation?




y greducing preventativereadmissions?

• How are you providingclinicians with targeteddiagnostic assistance?

• Which patients are followingdischarge instructions?

• How are you leveraging

unstructured data to preventand detect fraud?

• How are you using data topredict intervention programcandidates?

• Would revealing insightstrapped in unstructuredinformation facilitate more

informed decision making?

Physician notes and discharge summaries

Patient history, symptoms and non-symptoms

Pathology reports

Tweets, text messages and online forums

Satisfaction surveys Claims and case management data

Forms based data and comments

Emails and correspondence

Trusted reference journals including portals

Paper based records and documents

210 * AIIM website, accepted industry

percentage

Over 80% ofstored health

i f ti i

Medical Transcription Discharge Summary Sample # 2:DATE OF ADMISSION: MM/DD/YYYYDATE OF DISCHARGE: MM/DD/YYYY

ADMITTING DIAGNOSIS: Syncope.CHIEF COMPLAINT: Vertigo or dizziness.HISTORY OF PRESENT ILLNESS: This is an (XX)-year-old male with a past medical history of coronary artery disease, CABG done a fewyears ago, atrial fibrillation, peripheral arterial disease, peripheral neuropathy, recently retired one year ago secondary to leg pain. Thepatient came to the ER for an episode of vertigo while reaching for some books. The patient was able to reach the books, to supportself, but did not have any syncope. No nausea or vomiting. No chest pain. No shortness of breath. Came to ER and had a CT head,which was within normal limits. The impression was atrophy with old ischemic changes but no acute intracranial findings. No focalweakness, headache, vision changes or speech changes. The patient has had similar episodes since one year. Peripheral neuropathysince one year and not relieved with multiple medications. The patient also complains of weight loss of 25 pounds in the last 6months. No colonoscopy done. Recent history of hematochezia but believes it was secondary to proctitis and secondary to decreasedappetite. No nausea, vomiting, no abdominal pain.PROCEDURES PERFORMED: The patient had a chest x-ray, which showed cardiomegaly with atherosclerotic heart disease, pleuralthickening and small pleural effusion, a left costophrenic angle which has not changed when compared to prior examination, COPDpattern. The patient also had a head CT which showed atrophy with old ischemic changes. No acute intracranial findings.CONSULTS OBTAINED: A rehab consult was done.

Cardiology Consultation Transcribed Medical Transcription Sample Reports REFERRING PHYSICIAN: John Doe, MD CONSULTING PHYSICIAN: Jane Doe, MD HISTORY OF PRESENT ILLNESS: This (XX)-year-old lady is seen in consultation for Dr. JohnDoe. She has been under consideration for ventral hernia repair and has a background of aorticvalve replacement and known coronary artery disease. The patient was admitted with complaintsof abdominal pain, anorexia, and vomiting. She underwent a CT scan of the abdomen and pelvisand this showed the ventral hernia involving the transverse colon, but without strangulation. Therewas an atrophic right kidney. She had bilateral renal cysts. The hepatic flexure wall was thickened.There was sigmoid diverticulosis without diverticulitis. It has been recommended to her that sheundergo repair of the ventral hernia. For this reason, cardiology consult is obtained to assesswhether she can be cared from the cardiac standpoint. PAST CARDIAC HISTORY: Bypass surgery. She underwent echocardiography and cardiaccatheterization prior to the operation. Echocardiography showed an ejection fraction of 50%. Therewas marked left ventricular hypertrophy with septal wall 1.60 cm and posterior wall 1.55 cm.Coronary arteriography showed 90% stenosis in the anterior descending artery, situated distally just before the apex of the left ventricle. Only mild to moderate narrowing was seen elsewhere inthe coronary circulation. CORONARY RISK FACTORS: Her father had an i rregular heartbeat and her brother had a fatalheart attack. She herself has had high blood pressure for 20 years. She has elevated cholesteroland takes Lipitor She has had diabetes for 20 years She is not a cigarette smoker She does little

Unstructured Data is Messy butFilled with Key Medical Facts




PAST MEDICAL/SURGICAL HISTORY: Positive for atrial fibrillation. The patient had AVR 6 years ago. Peripheral arterial disease withhypertension, peripheral neuropathy, atherosclerosis, hemorrhoids, proctitis, CABG, and cholecystectomy.

FAMILY HISTORY: Positive for atherosclerosis, hypertension, autoimmune diseases in the family.SOCIAL HISTORY: Never smoked. Alcohol socially. No drugs.ALLERGIES: NO KNOWN DRUG ALLERGIES.REVIEW OF SYMPTOMS: Weight loss of 25 pounds within the last 6 months, shortness of breath, constipation, bleeding fromhemorrhoids, increased frequency of urination, muscle aches, dizziness and faintness, focal weakness and numbness in both leg s, kneesand feet.PHYSICAL EXAMINATION: VITAL SIGNS: Blood pressure 188/74, pulse 62, respirations 18 and saturatio n of 98% on room air. GeneralAppearance: The patient is a pleasant man, comfortable. HEENT: Conjunctivae are normal. PERRLA. EOMI. NECK: Nomasses. Trachea is central. No thyromegaly. LUNGS: Clear to auscultation and percussion bilaterally. HEART: Irregularrhythm. ABDOMEN: Soft, nontender, and nondistended. Bowel sounds are positive. GENITOURINARY: Prostate is hypertrophic withsmooth margin. EXTREMITIES: Upper and lower limbs bilaterally normal. SKIN: Normal. NEUROLOGIC: Cranial nerves are grosslywithin normal limits. No nystagmus. DTRs are normal. Good sensation. The patient is alert, awake, and oriented x3. Mild confusion.LABORATORY DATA AND RADIOLOGICA L RESULTS: WBC 8.6, hemoglobin 13.4, hematocrit 39.8, platelets 207,000, MCV 91.6, neutrophilpercentage of 72.6%. Sodium 133, potassium 4.7, chloride 104. Blood urea nitrogen of 18 and creatinine of 1.1. PT 17.4, INR 1.6, PTT33.The patient had a chest x-ray, which showed cardiomegaly with atherosclerotic heart disease, pleural thickening and small pleuraleffusion, a left costophrenic angle which has not changed when compared to prior examination, COPD pattern. The patient also had ahead CT, which showed atrophy with old ischemic changes. No acute intracranial findings.HOSPITAL COURSE AND TREATMENT: This is an (XX)-year-old male with syncope.1. Syncope. This may be secondary to questionable cerebral ischemia/atrial fibrillation/hypotension, so Neurology was kept on b oardand the patient was scheduled for a carotid Doppler and a 2-D echo. Orthostatics were ordered. Vitamin B12, TSH, free T4 and T3 wereordered along with cortisol level in the morning. FOBT x3 were done and cardiology followup as outpatient. The patient had a carotidDoppler done on the next day and it showed mild irregular plaque disease, right and left internal carotid arteries, approximately 20-59%. The patient's vitamin B12 level came the next morning and the level was 1180. His folate was 18.7 and his TSH was 1.98, free T4 of1.38 and T4 level of 7.4, cortisol level of 15.4, which are within normal limits. Dr. Doe, who is the patient's cardiologist, wasinformed. Dr. Doe was kind enough to see the patient the very next day, and his impression was that the patient has atrial fibrillation,rate controlled, status post AVR, St. Jude, and peripheral neuropathy. Subtherapeutic INR, the patient's relative target INR is 2-3. Hesuggested PT evaluation and suggested a low dose of SSRI and Dr. Doe was of the opinion that the patient does not need any fu rthercardiac recommendation. CT chest, abdomen, and pelvis were done. CT chest had an impression of coronary artery calcification, aorticvalve replacement, cardiomegaly, suspect a very s mall left pleural effusion, no acute active pulmonary disease. CT abdomen and pelvisshowed prior cholecystectomy, diverticulosis of sigmoid colon, two benign-appearing simple cysts involving the right kidney, calcifiedarteriosclerotic plaque disease of the abdominal aorta and iliac vessels bilaterally. The patient was ruled out of any malignancywhatsoever.

2. Hypertension. The patient at home was on Cardizem ER 90 mg thrice daily, and it was changed initially to Cardizem 90 mg thrice daily,and then with Dr. Doe's request, we changed the Cardizem to 240 mg t.i.d.3. Atrial fibrillation with subtherapeutic INR. The patient at home was on Digitalis. That was continued. Dr. Doe was of the opinion thatthe patient himself takes care of the Coumadin, and Dr. Doe was of the opinion that probably that is why the patient is not able tomaintain therapeutic INR. In the hospital, the patient's warfarin was increased to 5 mg q.h.s., and at the time of the discharge, he wasrequested to follow his appointments so that his INR can be maintained.4. Gout. The patient was on allopurinol. There were no acute issues regarding the gout.5. Prophylaxis. The patient was on Protonix and TEDs.6. Social. The patient is FULL CODE.

DISCHARGE DIAGNOSIS: Syncope.

DISCHARGE DISPOSITION: The patient is discharged to home.

DISCHARGEMEDICATIONS: The atientwasdischar edon the followin medications Cardizem90 m o thrice dail di oxin0 125m

Echocardiogram Sample Report:DATE OF STUDY: MM/DD/YYYYDATE OF INTERPRETATION OF STUDY:Echocardiogram was obtained for assessment of left ventricular

function. The patient has been admitted with diagnosis ofsyncope. Overall, the study was suboptimal due to poor sonic window.FINDINGS:1. Aortic root appears normal.2. Left atrium is mildly dilated. No gross intraluminal pathology isrecognized, although subtle abnormalities could not be excluded. Rightatrium is of normal dimension.3. There is echo dropout of the interatrial septum. Atrial septal defectscould not be excluded.4. Right and left ventricles are normal in internal dimension. Overall leftventricular systolic function appears to be normal. Eyeball ejectionfraction is around 55%. Again, due to poor sonic window, wall motionabnormalities in the distribution of lateral and apical wall could not beexcluded.5. Aortic valve is sclerotic with normal excursion. Color flow imaging and

Doppler study demonstrates trace aortic regurgitation.6. Mitral valve leaflets are also sclerotic with normal excursion. Color flowimaging and Doppler study demonstrates trace to mild degree of mitralregurgitation.7. Tricuspid valve is delicate and opens normally. Pulmonic valve is notclearly seen. No evidence of pericardial effusion.CONCLUSIONS:1. Poor quality study.2. Eyeball ejection fraction is 55%.3. Trace to mild degree of mitral regurgitation.4. Trace aortic regurgitation.

and takes Lipitor. She has had diabetes for 20 years. She is not a cigarette smoker. She does littlephysical exercise. REVIEW OF SYMPTOMS: CARDIOVASCULAR AND RESPIRATORY: She has no chest pain. She

sometimes becomes short of breath if she walks too far. No cough. She has occasional swelling ofher feet. Occasionally, she gets mildly lightheaded. Has not lost consciousness. She tends to beaware of her heartbeat when she is tired. She has no history of heart murmur or rheumatic fever.GASTROINTESTINAL: Recent GI symptoms as noted above, but she does not usually have suchproblems. She has had no hematemesis. She has no history of ulcer or jaundice. She sometimeshas loose stools. No constipation and no blood in the stool. GENITOURINARY: She tends to haveurinary frequency. She gets up once at night to pass urine. No dysuria, incontinence. She has hadprevious urinary infections. No stones noted. NEUROLOGIC: She has occasional headaches. Noseizures. No trouble with vision, hearing, or speech. No limb weakness. MUSCULOSKELETAL: Shetends to have joint and muscle pains and has a history of gout. HEMATOLOGIC: No anemia,abnormal bleeding, or previous blood transfusion. GYNECOLOGIC: No gynecologic or breastproblems. PAST MEDICAL HISTORY: She has had shoulder and hand injuries and has had carpal tunnelsurgery. She has been diabetic and has been on insulin. She has chronic renal insufficiency withcreatinine around 2.2. She has had hypothyroidism. She has had morbid obesity. She has chronicobstructive sleep apnea and uses BiPAP. She has had hysterectomy and oophorectomy in the past.Otherwise as noted above. MEDICATIONS: Prior to hospital, she was taking glipizide XL 2.5 mg daily, metoprolol 50 mgb.i.d., Cipro 250 mg b.i.d., atorvastatin 40 mg dail y, Synthroid 75 mcg daily, aspirin 81 mg dail y,

and Lantus 36 units dail y. Currently, she is taking Lipitor 40 mg dail y, Lantus 10 units at bedtime,Synthroid 75 mcg daily, metoprolol 50 mg b.i.d., and Zosyn 2.25 grams q.6h. SOCIAL HISTORY: She does not drink alcohol. PHYSICAL EXAMINATION: GENERAL APPEARANCE: She is not currently dyspneic, in no distress. She is alert, oriented, andpleasant. HEENT: Pupils are normal and react normally. No icterus. Mucous membranes well colored. NECK: Supple. No lymphadenopathy. Jugular venous pressure not elevated. Carotids equal. HEART: The heart rate is 82 per minute and regular and the blood pressure 132/78. The cardiacimpulse has a normal quality. There i s a grade 3/6 ejection systolic murmur heard medial to theapex and at the aortic area, wi th well heard radiation to the neck vessels. CHEST: Chest is clear to percussion and auscultation. Normal respiratory effort. ABDOMEN: Soft and nontender. The presence of a large ventral hernia is noted. EXTREMITIES: There is no edema. Posterior tibial pulses were felt bilaterally, but I did not feel thedorsalis pedis. SKIN: No rash or significant lesions are noted. LABORATORY AND DIAGNOSTIC DATA: Electrolytes are normal. BUN and creatinine 18/2.2.Blood sugar 150. White count is 7.6, hemoglobin 11.7 with hematocrit 34.9, platelets 187,000.LFTs were normal. Hemoglobin A1c 7.7. TSH 1.82. Troponin I was normal on three occasions. Chest x-ray showed an enlarged heart with postoperative changes, but no evidence of acute

pathology. EKG shows probable left atrial enlargement. Low voltage QRS, probable inferior wallmyocardial infarction and anterior wall infarction, age undetermined. ASSESSMENT: 1. Aortic valve replacement with bioprosthetic valve. Residual systolic murmur. 2. Arteriosclerotic heart disease with severe stenosis in anterior descending artery, but this issituated distally and subtends only a small mass of myocardium. 3. Well preserved left ventricular systolic function. The EKG appearance of previous myocardial

infarction is probably serious, indicating multiple other medical problems as listed aboveand also documented in the chart. RECOMMENDATIONS: It appears that she does not wish to proceed with thesurgery at this time, and if such surgery is not

Cardiology Consultation Transcribed Medical Transcription Sample ReportsDATE OF CONSULTATION: MM/DD/YYYY REFERRING PHYSICIAN: John Doe, MD CONSULTING PHYSICIAN: Jane Doe, MD REASON FOR CONSULTATION: Surgical evaluation for coronary artery disease. HISTORY OF PRESENT ILLNESS: The patient is a (XX)-year-old female who has a known history of coronary arterydisease. She underwent previous PTCA and stenting procedures in December and most recently in August. Since that time,she has been relatively stable with medical management. However, in the past several weeks, she started to notice someexertional dyspnea with chest pain. For the most part, the pain subsides with rest. For this reason, she was re-evaluatedwith a cardiac catheterization. This demonstrated 3-vessel coronary artery disease with a 70% lesion to the right coronaryartery; this was a proximal lesion. The left main had a 70 % stenosis. The circumflex also had a 99% stenosis. Overall leftventricular function was mildly reduced with an ejection f raction of about 45%. The left ventriculogram did note some apicalhypokinesis. In view of these findings, surgical consultation was requested and the patient was seen and evaluated by Dr.Doe.PAST MEDICAL HISTORY: 1. Coronary artery disease as described above with previous PTCA and stenting procedures.

2. Dyslipidemia.3. Hypertension.4. Status post breast lumpectomy for cancer with followup radiation therapy to the chest.ALLERGIES: None.MEDICATIONS: Aspirin 81 mg daily, Plavix 75 mg daily, Altace 2.5 mg daily, metoprolol 50 mg b.i.d. and Lipitor 10 mgq.h.s. SOCIAL HISTORY: She quit smoking approximately 8 months ago. Prior to that time, she had about a 35 - to 40-pack-yearhistory. She does not abuse alcohol.FAMILY MEDICAL HISTORY: Mother died prematurely of breast cancer. Her father died prematurely of gastric carcinoma. REVIEW OF SYMPTOMS: There is no history of any CVAs, TIAs or seizures. No chronic headaches. No asthma, TB,hemoptysis or productive cough. There is no congenital heart abnormality or rheumatic fever history. She has nopalpitations. She notes no nausea, vomiting, constipation, diarrhea, but immediately prior to admission, she did developsome diffuse abdominal discomfort. She says that since then, this has resolved. No diabetes or thyroid problem. There isno depression or psychiatric problems. There is no musculoskeletal disorders or history of gout. There are no hematologicproblems or blood dyscrasias. No bleeding tendencies. Again, she had a history of breast cancer and underwentlumpectomy procedures for this with followup radiation therapy. She has been followed in the past 10 years andmammography shows no evidence of any recurrent problems. There is no recent f evers, malaise, changes in appetite orchanges in weight.PHYSICAL EXAMINATION: Her blood pressure is 120/70, pulse is 80. She is in a sinus rhythm on the EKGmonitor. Respirations are 18 and unlabored. Temperature is 98.2 degrees Fahrenheit. She weighs 160 pounds, she is 5 feet4 inches. In general, this was an elderly-appearing, pleasant female who currently is not in acute distress. Skin color andturgor are good. Pupils were equal and reactive to light. Conjunctivae clear. Throat is benign. Mucosa was moist andnoncyanotic. Neck veins not distended at 90 degrees. Carotids had 2+ upstrokes bilaterally without bruits. Nolymphadenopathy was appreciated. Chest had a normal AP diameter. The lungs were clear in the apices and bases, no

wheezing or egophony appreciated. The heart had a normal S1, S2. No murmurs, clicks or gallops. The abdomen was soft,nontender, nondistended. Good bowel sounds present. No hepatosplenomegaly was appreciated. No pulsatile masses werefelt. No abdominal bruits were heard. Her pulses are 2+ and equal bilaterally in the upper and lower extremities. Noclubbing is appreciated. She is oriented x3. Demonstrated a good amount of strength in the upper and lowerextremities. Face was symmetrical. She had a normal gait. IMPRESSION: This is a (XX)-year-old female with significant multivessel coronary artery disease. The patient also has a leftmain lesion. She has undergone several PTCA and stenting procedures within the last year to year and a half. At this point,in order to reduce the risk of any possible ischemia in the future, surgical myocardial revascularization is recommended. PLAN: We will plan to proceed with surgical myocardial revascularization. The risks and benefits of this procedure wereexplained to the patient. All questions pertaining to this procedure were answered.

Medications, Diseases, Symptoms, Non-

Symptoms, Lab Measurements, Social

History, Family History and Much More

Unstructured Information Use LandscapeNatural Language Process ing is Needed to




Claims Analysis

CommerceSearch

Search Analyze / Visualize(Trends, Patterns, Relationships) Deep QA

Safety,Defects,

Maintenanc

e

EnterpriseSearch

eDiscovery, Legal

Risk,Fraud,

Security

mantic Understanding, Ontology Mgt and Big Dat

VoC,Churn,

Cust Svc

SocialMedia,

Marketing

ProductQuality

ExpertiseLocator

KnowledgeMgt

CallCenter,

Help, SelfService

Research(Biz, Edu,

Legal,Scientific)

TradingSupport

HealthcareOperationa

l

HealthcareClinical

Publishing(Tag, Locate)

IterativeQ&A

IncidentManagement

BJC Healthcare and Washington UniversityPartnershipImproving care and increasing revenue




"We anticipate this solution to be a game changerin biomedical research and patient care …accelerate the

pace of clinical and translational research …"

Dr. Rakesh Nagarajan, MD, PhD, AssociateProfessor, Department of Pathology and

Immunology,Washington University

213

Business ChallengeExisting Biomedical Informatics (BMI) resourceswere disjointed, siloed, redundant and onlyavailable to a few researchers - key insights notaccessible, trapped in unstructured clinicalnotes, diagnostic reports, etc.

What‟s Smart? Leverage unstructured information along withstructured data by using IBM Content Analyticswith IBM InfoSphere Warehouse

Smarter Business OutcomesResearchers now able to see new trends,patterns and find answers in days instead of

weeks or months eliminating manual methodsalso enables new grant revenue

while lowering research costs

IBM is helping to transform healthcareRevealing c l in ical and operat ional insigh ts in the high impact overlap

between

cl in ical and o perat ional – enabl ing low cost accountable care




cl in ical and o perat ional enabl ing low cost accountable care

214

Diagnostic assistance

Clinical treatment

effectivenessCritical care intervention

Research for improveddisease management

Readmission prevention

Claims management

Fraud detection and prevention

Voice of the patient

Patient discharge andfollow-up care

IBM Content and Predictive Analytics for Healthcare

Improved patient satisfaction at lower costs Enhanced patient care with optimized

outcomes

Clinical

Outcomes

Operational

Outcomes


How it works




Unstructured Data(Nurses notes, discharge notes,etc.)

Structured Data(Billing data, EMR, etc.)

RawInformation

Search and Visually

Explore (Mine)

Monitor, Dashboard

and Report

Question and Answer*

Custom Solutions

DynamicMultimodeInteraction

IBM Contentand Predictive

Analytics

Content Analytics• Natural Language Processing• Medical Fact and Relationship

Extraction (Annotation)• Trend, Pattern, Anomaly,

Deviation Analysis

Predictive Analytics

• Predictive Scoring andProbability Analysis

Analyzed andVisualized

Information

HealthIntegrationFramework

Data Warehouse and Model

Master Data Management

Advanced Case Management

Business AnalyticsPartners (HLI) Specialized Research

IBM Watson forHealthcare

Confirm hypotheses or seek alternative

ideas with confidence based responsesfrom learned knowledge*

* Future optional capability




What‟s so innovative? Patient Age: 42




A 42-year old white male presents for a physical. Herecently had a right hemicolectomy invasivegrade 2 (of 4) adenocarcinomain the ilocecal valve was found

and excised. At the same timehe had an appendectomy.The appendix showed nodiagnostic abnormality.

Gender: Male

Race: White

Procedure hemicolectomy

diagnosis: invasive

adenocarcinoma

anatomical site:

ileocecal valve

grade: 2 (of 4)

Procedure appendectomy

diagnosis: normal anatomical site:

appendix

* Future capability

Accurately extractburied medical factsand relationships with

medical annotators

Analyze compiled information for trends,patterns, deviations, anomalies and relationshipsin aggregate to reveal new insights with content

analyticsModel, score and predict the probability ofoutcomes with predictive analytics

Make insights accessible andactionable for all clinical andoperational knowledge workers

(and systems)

Physicians

Other CliniciansCare CoordinatorsResearchers

Executives Business AnalystsClaimsFraud

Other Systems and Applications

KnowledgeWorkers

Confirm hypotheses or seek alternative ideas from learned knowledgevia Watson for Healthcare from the same user interfaces*

217

Seton Healthcare Family

Reducing CHF readmission to improve care“ IBM Content and Predictive Analytics for Healthcare uses the same type of natural language processing as IBM

Featured in




Business ChallengeSeton Healthcare strives to reduce the occurrence ofhigh cost Congestive Heart Failure (CHF)readmissions by proactively identifying patients likely tobe readmitted on an emergent basis.

What‘s Smart? IBM Content and Predictive Analytics for Healthcaresolution will help to better target and understand high-risk CHF patients for care management programs by:

Smarter Business Outcomes • Seton will be able to proactively target care

management and reduce re-admission of CHFpatients.

• Teaming unstructured content with predictive

analytics, Seton will be able to identify patients likelyfor re-admission and introduce early interventions toreduce cost, mortality rates, and improved patientquality of life.

IBM solution• IBM Content and

Predictive Analytics

for Healthcare• IBM Cognos

BusinessIntelligence

• IBM BAO solutionservices

• Utilizing natural language processing to extract keyelements from unstructured History and Physical,

Discharge Summaries, Echocardiogram Reports, andConsult Notes

• Leveraging predictive models that have demonstratedhigh positive predictive value against extracted elements

of structured and unstructured data

• Providing an interface through which providers can

intuitively navigate, interpret and take action

Watson, enabling us to leverage information in new ways not possible before. We can access an integrated view

of relevant clinical and operational information to drive more informed decision making and optimize patient and

operational outcomes.”

Charles J. Barnett, FACHE, President/Chief Executive Officer, Seton Healthcare Family

The Data We Thought Would Be Useful Wasn‘t

What Really Causes Readmissions at Seton

Key Findings




The Data We Thought Would Be Useful … Wasn t

• 113 candidate predictors from structured and unstructured data sources

• Structured data was less reliable then unstructured data – increased the reliance onunstructured data

New Unexpected Indicators Emerged … Highly Predictive Model

• 18 accurate indicators or predictors (see next slide)

Predictor Analysis % EncountersStructured Data

% EncountersUnstructured

Data

Ejection Fraction(LVEF)

2% 74%

Smoking Indicator 35%

(65% Accurate)

81%

(95% Accurate)

Living Arrangements <1% 73%(100% Accurate)

Drug and Alcohol Abuse

16% 81%

Assisted Living 0% 13% 219

97% at 80th percentile

49% at 20th percentile

What Really Causes Readmissions at Seton

Top 18 Indicators

New Insights Uncovered by Combining Content and Predictive




18. Jugular Venous Distention Indicator

17. Paid by Medicaid Indicator

16. Immunity Disorder Disease Indicator15. Cardiac Rehab Admit Diagnosis with CHF Indicator

14. Lack of Emotion Support Indicator

13. Self COPD Moderate Limit Health History Indicator

12. With Genitourinary System and Endocrine Disorders

11. Heart Failure History

10. High BNP Indicator

9. Low Hemoglobin Indicator

8. Low Sodium Level Indicator

7. Assisted Living (from ICA Extract)

6. High Cholesterol History

5. Presence of Blood Diseases in Diagnosis History

4. High Blood Pressure Health History

3. Self Alcohol / Drug Use Indicator (Cerner + ICA)

2. Heart Attack History

1. Heart Disease History

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

0 1 2 3 4 5 6

R a n k i n

g o f S t r e n g t h o f M o d e l V a r i a b l e

P r oj e ct e d O d ds R a o

1 8 17 16 15 14 13 12 11 10

9 8 7 6 5 4 3 2 1

g y g

Analytics• LVEF and Smoking are significant indicators of CHF but not

readmissions

• Assisted Living and Drug and Alcohol Abuse emerged as key predictors(only found in unstructured data)

• Many predictors are found in “History” notations and observations

Patient X was hospitalized 6 times over an 8 month period. The same

The Impact of Readmissions at Seton

CHF Patient X – What Happened? Admit / Readmission

30-Day Readmission




basic information was available at each encounter and Patient X‟sreadmission prediction score never dropped below 95 (out of possible100)


Patient X (DSS & Cerner)High Model Score (100)Gender: Male

Age: 73Insurance: MedicaidLack of Emotional Support: YesSodium Level: Low

Cholesterol Level: HighCOPD History: YesHeart Disease & Heart Failure History:YesHBP History: Yes

NLP Clinical DocumentationLiving Arrangement:Permanent

Assisted Living: NoSmoking History: YesSmoking Amount: N/A

Alcohol Abuse History: Yes

Drug Abuse History: N/AEjection Fraction: N/A

Apr-18-2009 May-12-2009 May-20-2009 Oct-11-2009 Nov-24-2009 Dec-20-2009

8 days24 days 144 days 44 days 26 days

Individual Patient Data at Each Encounter (Patient X @ Dec 20, 2009)

98% 98% 96% 95% 96% 100%

Patient PopulationMonitoring Clinical and

Operational Data



The Impact of Readmissions at Seton

CHF Patient X – What Happened? Admit / Readmission

30-Day Readmission

17% of Out-83% of Out-of-PocketCosts Were Avoidable



© 2010 IBM Corporation© 2011 IBM Corporation

Patient PopulationMonitoring Clinical and

Operational Data

Summary of KeyReadmission Risk Factors for Patient X

• Possible InterventionFactors: High Cholesterol,Low Sodium, EmotionalSupport, High Blood

Apr-18-2009 May-12-2009 May-20-2009 Oct-11-2009 Nov-24-2009 Dec-20-2009

8 days24 days 144 days 44 days 26 days

17% of Out-

of-PocketCosts at 1st Encounter

Costs Were Avoidable

(5 UnnecessaryEncounters)

IBM Content and Predictive Analytics … Ready for Watson

Complements IBM Watson to analyze and visualizepast, present and future scenarios in context




Question What isKnown

Complement ICPA with Watson for Healthcareto get real time, confidence based answers

with evidence based learning

224

Analyze and

Visualize thePastUnderstand trends,

patterns, deviations, anomalies,context and more in large corpuses of

historical clinical and operational

Predict the

FutureUse predictive models andscoring to make more

informed decisions throughpredictive and futurescenario modeling

See the

Present Analyze and extract text fromin-process documents or otherinformation to find structureddata errors … feed the resultsto other cases and systems



How to get started

NO FUTNOExpand and integrateICPA b d l ti

Complement with IBMW t f H lth

Start with IBM Contentd P di ti A l ti




IBM BAO Solution Services – Center of Excellence

Al il bl W kl d

Expand solution value byintegrating other systems andcapabilities

Maximize solution valueby extending with IBMWatson for Healthcare forreal-time confidencebased answers

W UREW Address pressing clinical andoperational issues today

IBM BAO Enterprise Services

Advanced Case Management

Business AnalyticsData Warehouse and

Healthcare Data Model

Master Data Management

Partner Solutions

U n i q u e V a l u e

D e l i v e r e d

ICPA-based solutions Watson for Healthcareand Predictive Analytics

1 2 3

KEY

ClinicalOutcomes

OperationalOutcomes

ClinicalOutcomes

OperationalOutcomes

pn healthcare 100

Documents