size of the digital health market€¦ · innovators dilemma • the innovators dilemma is a term...
TRANSCRIPT
1 1
Research on telehealth and e-health applications to support the implementation of Personalised Medicine
Dr. Sofoklis Kyriazakos CEO - Innovation Sprint Associate Prof. - Aarhus University @skyriazakos
2 2
Overview
• Size of the digital health market • Health Trends & Personalization • Innovation in Healthcare technologies • Privacy and Regulation • New business models in healthcare • Personalization in Clinical Research • R&D Activities on personalized eHealth • Concluding remarks
3 3
Size of the digital health market
4 4
Projected CAGR for the global digital health market in
the period 2015-2020, by major segment
Source: Allied Market Research; MarketsandMarkets; Transparency Market Research; BCC Research; Roland Berger ID 387875
41%
23%
15%
4%
21%
0,0%
5,0%
10,0%
15,0%
20,0%
25,0%
30,0%
35,0%
40,0%
45,0%
Mobile health Wirelesshealth
Telehealth EHR/EMR Average
Co
mp
ou
nd
an
nu
al g
row
th r
ate
20
15
-20
20
Personalization
5 5
Most active digital health subsectors worldwide based
on invested funding in 2016 (in million U.S. dollars)
Source: StartUp Health ID 388905
2.800
1.000
765
713
593
562
436
332
280
277
0 500 1000 1500 2000 2500 3000
Patient/consumer experience
Wellness/benefits
Personalized health/quantified self
Medical device
Workflows
Big data/analytics
Population health
Clinical decision support
Research
E-commerce
Funding invested in million U.S. dollars
Personalization
6 6
Projected total global personalized mHealth devices and
services revenue from 2014 to 2020 (in billion U.S. dollars)
Source: Statista estimates; Zion Market Research ID 628190
14,5 16,5
19
21,9
25,7
30,5
35,8
0
5
10
15
20
25
30
35
40
2014 2015 2016 2017 2018 2019 2020
Rev
en
ue
in b
illio
n U
.S. d
olla
rs
7 7
Global telemedicine market size from 2015 to 2021 (in
billion U.S. dollars)*
Source: Statista estimates; MRAS ID 671374
18,1 20,2
23
26,5
30,5
35,5
41,2
0
5
10
15
20
25
30
35
40
45
2015 2016 2017 2018 2019 2020 2021
Mar
ket
size
in b
illio
n U
.S. d
olla
rs
8 8
9 9
Health Trends
10 10
Top 5 eHealth Trends
1. The Internet of Me: Your healthcare, personalized
2. Outcome Economy: Hardware producing healthy results
3. The Platform (R)evolution: Defining ecosystems, redefining healthcare
4. Intelligent Enterprise: Huge data, smarter systems, better healthcare
5. Workforce Reimagined: Collaboration at the intersection of humans and healthcare
11 11
1.The Internet of Me: Your healthcare, personalized
12 12
2.Outcome Economy: Hardware producing healthy results
13 13
3. The Platform (R)evolution: Defining ecosystems, redefining healthcare
14 14
4. Intelligent Enterprise: Huge data, smarter systems, better healthcare
15 15
5.Workforce Reimagined: Collaboration at the intersection of humans and healthcare
16 16
Innovation in Healthcare technologies
17 17
Innovators Dilemma
• The Innovators Dilemma is a term introduced by Clayton Christensen in his book (published in 1997) that describes how great companies can fail, if they do not follow innovation.
• The innovators dilemma has 2 main parameters: 1. Value to innovation is an S-curve, as product requires effort, budget, time and many iterations. The
iterations start with a basic (minimal) product and increase in terms of value to the customers. After a number of iterations, the value increase is very little, like the initial ones.
2. Incumbent players: The incumbent players have a large customer base, but require a large volume of sales per year. Startups can survive with much smaller market share and therefore can focus on innovation.
18 18
Dilemma Zone
Apple:: The Innovators Dilemma https://medium.com/@mmaybl/apple-the-innovators-dilemma-5df4e70ab109
19 19
KODAK: Failing to Jump to the Right S-Curve
Source: PMA
KODAK file for bankruptcy
KODAK file for bankruptcy
KODAK file for bankruptcy
KODAK file for bankruptcy
20 20
KODAK: Failing to Jump to the Right S-Curve
Steven Sasson went to work for Kodak in 1973, The New York Times reports. He was tasked with figuring out whether a "charged coupled device" (C.C.D.) had any practical application. This led him, through a series of steps, not only to invent the first digital camera but also to invent a device to display it on.
Patent drawing for the "Electronic still camera," filed in 1977. Google Patents
https://www.businessinsider.com/this-man-invented-the-digital-camera-in-1975-and-his-bosses-at-kodak-never-let-it-see-the-light-of-day-2015-8
21 21
data information knowledge wisdom action
The Data Journey
21
22 22
Io(M)T
23 23
Io(M)T
Source: https://image.slidesharecdn.com/ntk2015-internetofthingsiot-smarthome-150525105210-lva1-app6891/95/ntk-2015-internet-of-things-track-iot-smart-home-4-638.jpg?cb=1432551484
24 24
Big Data
25 25
Why Healthcare Organizations Must Develop a Range of Big Data Capabilities
Source: McKinsey
26 26
Big Data Adoption can Reduce US Healthcare Cost by $300 to $450 billion
Source: McKinsey
27 27
Robotics & AI
28 28
3D printing
Source: http://3d2go.com.ph/blog/breakthrough-applications-of-3d-printing-for-healthcare-and-medical-science/
29 29
Blockchain technology
Source: Petre, 2016a 2017 Source: Deloitte, 2016
30 30
Privacy & Regulation
31 31
Healthcare Privacy Concerns
32 32
Medical Device Regulation
MDR-Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, repealing Council Directives 90/385/EEC and 93/42/EEC
IVDMDR-Regulation (EU) 2017/746 of the European Parliament and of the Council of 5 April 2017 on in vitro diagnostic medical devices, repealing Directive 98/79/EC and Commission Decision 2010/227/EU
33 33
• It is necessary to clarify that software in its own right, when specifically intended by the manufacturer to be used for one or more of the medical purposes set out in the definition of a medical device, qualifies as a medical device • Software, which drives a device or influences the use of a device, shall fall within the same class as the device • Software that is independent of any other device shall be classified in its own right
• Software for general purposes, even when used in a healthcare setting, or software intended for life-style and well-being purposes is not a medical device.
• The qualification of software, either as a device or an accessory, is independent of the software's location or the type of interconnection between the software and a device
Medical Device Regulation
34 34
Medical Device Regulation
35 35 https://privacy-analytics.com/de-id-university/gdpr-and-the-future-of-clinical-trials-data-sharing-from-the-executive-roundtable-at-london-dia-2018/
General data Protection Regulation
36 36
New business models in healthcare
37 37
Pricing vs Value/Usage matrix
Source: J Elton, A. O’Riordan, Healthcare Disrupted
38 38
New Business Models
Lean Innovators
Around-the-Patient Innovators
Value Innovators
New Health Digitals
39 39
Lean Innovators
• Best practices of efficient manufacturing • Cost management & RoI • Advanced M&A expertise • Eye for niche markets • Aim to control and minimize R&D expenses
40 40
Around-the-Patient Innovators
• Focused on producing drugs • Seek to differentiate by developing of ancillary services • Create new basis for product economics • Among the most innovative companies • The are looking for key R&D partnerships and digital infrastructure
41 41
Value Innovators
• Focusing on improving patient and clinical outcomes • Make healthcare system more efficient and effective • They are offering integrated solutions to bridge divides between physicians and patients • Their customers are health providers and health insurers • Product- and analytic-centric companies
42 42
New Health Digitals
• Developing their own technologies and infrastructures • Technology-savvy that have “gone digital” • Startup organizations • 2 variations:
• “Digital Gone Healthcare”, consumer and infrastructure firms (e.g. Google, Apple) • “Healthcare Gone Digital”, healthcare-centric firms (e.g. Philips, GE Medical)
43 43
Personalization in Clinical Research
44 44
Clinical Trials – Status of Play
INVESTIGATOR PATIENT Heavy burden
Poor data collection transferred to an EDC
High rate of drop-outs
Low control over patient adherence
Phase I/II/III
SPONSOR
Data analysis by a CRO
Poor results with low quality of data
APRO
VAL
1 in 5 Drugs passes FDA
approval after Phase III
TRIA
L FIN
ISH
45 45
Numbers Behind Drug Development
AI
Can we reduce the failure rate between Phases, save cost & accelerate the process…… while getting insights on drug efficacy?
Human tested Drugs gets approval
1of 5 $40m Average cost of a Phase II or III
failure
$1-2bn Total cost of drug development
~10y Average drug development period
46 46
A Personalized Approach
PATIENT SPONSORS INVESTIGATOR
Wearable & Healthentia
AI
Treatment reminders QoL & vital signs monitoring Interaction with investigator
Trial management Customized dashboard Real time patient adherence overlook
BI & reporting Drug efficacy on patient phenotypes
Patient phenotypes
47 47
RECEIVES medication reminders scheduled questionnaires
regarding Health status
MONITORS A health diary of reported events
and questionnaires how activity and sleep can impact
her quality of life and treatment
Engaging Patients
48 48
Empowering Research
Unlock insights and customize the dashboard to visualize clinical endpoints
Overlook patients’ adherence and follow Trial Management
Send Messages and questionnaires can be scheduled or sent instantly to patients
49 49
AI Targets: Phenotypes
Model patients • Using clusters of biomarkers (sets of
vitals, PRO and lifestyle behaviors characterizing a patient)
• Model patient to the detail of a digital twin
Predict outcomes • Patient-reported outcomes from
models of past vitals, PROs and lifestyle predict new PRO
• Trial outcomes: significant correlation results to a phenotype based on our biomarkers
Digital trials
• Use data variation in derived models to generate multitude of patient avatars per model
• Apply interventions on avatars
• Validate digital trial outcomes by human trials
AI
Real World Data as “the ore” for modelling of Phenotypes • Vitals (traditional) • Patient-Reported Outcomes • Lifestyle (Physical activity, Sleep)
50 50
R&D activities on personalized eHealth
51 51
52 52
eHealth solution based on IoT and Big Data analytics to support the independent living of chronic disease patients
CloudCare2U is based on the outcome of the successful EC-funded project eWALL that has shown remarkable achievements in the treatment of patients suffering from MCI, COPD and frailty conditions. CloudCare2U is a cloud-based platform utilizing a holistic infrastructure model with “sensing” and “listening” environments with an affordable, easy to install system.
www.cloudcare2u.com
53 53
Horizontal Platform Services
Cloud Middleware
Home Sensing Midleware
Cloud Infrastructure Layer
Data ManagerCloud Gateway
Service Bricks
Remote Proxy
Profiling
Server
Notification
Manager
Intelligent
Decision
Support System
( IDSS)
Portal
Local Context Management
Device
Gateway
Se
cu
rity
an
d p
riva
cy m
an
age
me
nt
Co
mm
on
da
ta m
od
el
RESTful API (I4)
AM
QP
, R
ES
T
External Gateway
Semantic
Service
Manager
eWall Devices
Local Data
Manager
Context
ExtractionLocal IDSS
eWALL Applications
Home environment
Clo
ud
en
viro
nm
ent
https://github.com/ewallprojecteu/
Architecture
54 54
Bluetooth
ZigBee
mesh
ZigBee
mesh
ZigBee
WiFi
WiFi
Brix
Home PC
Plugwise Stick
& Circles
Kinect for Xbox One
& adapter
Arduino home sensing
USB hubKeyboard
& mouse
Elo 4201L
Home ADSL
modem
Philips Hue
gateway & lamps
Android
smartphoneFitbit
ChargeHR
Beddit
Nonin pulse
oximeter
Omron blood
pressure monitor
Arduino
explorer & Xbee
ThinkLabs
One
Wireless
USB
HDMI
Ethernet (or WiFi)
Audio
https://github.com/ewallprojecteu/
IoT Network
https://github.com/ewallprojecteu/
55 55
Family can put photos in this frame through
Flickr
Backgrounds can be changed
Weather on a window with animated
weather forecast
Activity lava lamp with present step
count
The TV shows health advertisements and
an agenda with upcoming
appointments
Calendar application
Cognitive Games
Domotics control
Advanced applications
Clock
Interaction Interface
https://github.com/ewallprojecteu/
56 56
A weekly overview shows the selected week’s
measurements along side the users behavior
pattern.
Source: www.ewallproject.eu
The Activity data is displayed in a
simplified graph.
The daily behavior data is displayed in a
simplified list with one meaningful icon and proper typography
https://github.com/ewallprojecteu/
Medical measurements
57 57
eCoaching & Intelligence
• Automatic Goal-Setting • Exercise Schedule Generator • Sleep Anomaly Classifier • Oxygen Saturation Monitor • Fall Detection • Activity Coach • Wellbeing Ads
https://github.com/ewallprojecteu/
58 58
Good morning Bob!
You have done 3,427 out of 7,500 steps.
Try to reach your step goal everyday: healthy body, healthy mind!
The weather outside is nice, why don’t you go cycling for a bit?
eCoaching
UTRECHT SDE, Utrecht, The Netherlands, 7th February 2018
https://github.com/ewallprojecteu/
59 59
Socially Assistive robots & eHealth applications
QT-CC2U integration
60 60
61 61
Virtual Agents & Behavioral Change
Council of Coaches aims to change the way we look at virtual coaching. We are creating an autonomous virtual council consisted of several coaches with different expertise and personalities that can assist people in achieving behavioral change for health and well-being. By advancing the state of the art in embodied conversational agents we are enabling fluent multi-party interaction between multiple coaches and our users.
council-of-coaches.eu
62 62
demo
https://www.youtube.com/watch?v=lWJvP3_HmXA
63 63
64 64
Virtual Agents & Personalized Rehabilitation
vCare aims to provide a smart coaching solution grounded on personalized care pathways. vCare stands for Virtual Coaching Activities for Rehabilitation in Elderly and aims at an improved rehabilitation for people as they age.
vCare-project.eu
65 65
# vCare`s Functional
Themes Market available
services Scope Existing solutions
Innovation aspects utilized by vCare
Clin
ical
P
ath
way
Per
son
aliz
ed
Co
ach
ing
Pla
tfo
rm
Ori
enta
tio
n
Sem
anti
c
Tech
no
logi
es
1 Vital Signs Monitoring telemonitoring
Heartrate monitoring
Blood pressure monitoring
O2 saturation
ECG monitoring
Glucose monitoring
Simulation-based Reinforcement
Learning X X
2 Context Awareness ambient monitoring Motion sensors
Weather sensors Supervised machine learning X X X
3 Rehabilitation Exercises serious games, tele-
rehab solution
aerobic exercises
isotonic exercises
mental training
cognitive training
Supervised machine learning X
4
Self-Management
Support (Patient
Empowerment)
dietary
recommendation
nutritional assessment
nutritional recommendations
food caloric values databases
Supervised machine learning X
5 User Experience GUI User-system
interaction
Interaction with the patient using a
simple to use graphical interface;
Humanoid avatar with multimodal
and user interface; capable of
communication
using natural language
X
6
Rehabilitation program
(based on individualised
clinical pathways)
Process adaptation Rule based adaptation
Intelligent automatic
adaptation (personalization) based
on Reinforcement Learning
X X
Personalization & Innovation
66 66
Warm up
Treadmill
Cycle
Cool down
... ...
Walk
Clinical Pathway
Markov Decision Process
translate
adapt
State
Action
Transition
Structured rules & -data
Gym
State-of-the-Art RL Benchmark
Latest algorithms
integrate & extend simulation
provenance
few-shot learning with external datasets
67 67
Concluding Remarks
• The digital era has entered the Healthcare domain already some years ago • Digital approaches that are not patient-centric will fail • Co-development is an essential element for a wider adoption of digital services • Personalization is mandatory to meet demanding needs and contribute to more efficient drug development and better
health