1) The document discusses self-quantification systems and big data prospects and challenges from these systems. It describes the quantified self movement and tools people use to self-monitor health metrics and experiences.
2) Various types of self-monitoring devices, sensors, and services are presented. Challenges with self-quantification include privacy, security, education, and ensuring data is used for health improvement rather than risk profiling.
3) Opportunities include using self-tracking data to prevent disease, shift care from tertiary to primary settings, and generating data to further research when shared. Standards are needed for integrating self-data with electronic health records.
Big Data Prospects and Challenges of Self-Quantification Systems
1. Panel: Biomedical and Healthcare Analytics
on Big Data
Self-Quantification Systems:
Big Data Prospects and Challenges
Fernando J. Martin-Sanchez
Professor and Chair of Health Informatics
Melbourne Medical School
&
Director, Health and Biomedical Informatics Centre (HABIC)
3. Participatory health
I. Personal genome services (BYO)
II. Personal diagnostic testing (BYO)
III. Personal medical image management (DIY)
IV. Personal sensing and monitoring (DIY)
V. Personal health records (DIY)
VI. Patient reading doctor’s notes
VII. Patient initiating clinical trials
VIII. Patient reporting outcomes
IX. Patient accessing health information
X. Shared decision making
Collecting
data
Participatory
health
Exchanging
information
4. Availability of new sensors for data collection
Exposome
Genome
Phenome
Environmental risk factors
(pollution, radiation, toxic agents, …)
Biomarkers (DNA sequence,
Epigenetics)
Anatomy, Physiological, biochemical parameters
(cholesterol, temperature, glucose, heart rate…)
Social media / Integrated personal health record / Personal Health Systems
8. New market
Global annual wearable device
unit shipments crossing the 100
million milestone in 2014, and
reaching 300 million units five
years from now
Gartner hype cycle
Corporate health
plans – 13 Mill
9. The Quantified Self community
• Quantified Self is a collaboration of users and tool
makers who share an interest in self knowledge through
self-tracking.
• We exchange information about our personal projects,
the tools we use, tips we’ve gleaned, lessons we’ve
learned. We blog, meet face to face, and collaborate
online. There are three main “branches” to our work.
– The Quantified Self blog and community site.
– Show and Tell meetings (Meetup groups) - Melbourne
– Quantified Self Conferences (US and Europe)
• Groups 112, Members 17,893, Cities 89, Countries 31
10.
11. The IBES SELF-OMICS Project
• Addressing the information and communication needs of the
‘quantified individual’ for enabling participatory and
personalised medicine
• Funded by IBES (Institute for a Broadband Enabled Society)
- 2012-2013
• Resources:
http://www.broadband.unimelb.edu.au/health/monitoring/selfomics.html
http://www.scoop.it/t/selfomics
http://pinterest.com/hbir/self-omics-self-monitoring-quantified-self-omics/
12. Variety of self-monitoring devices, sensors and services
MoodPanda
Actipressure
Zeo Sleep Manager
Fitbit
23andMe
Sensaris Senspod
uBiome
iBGStar
12
16. Classification of self-quantification systems
• Capture data directly from the user
(Primary or Secondary)
• Sensor Location (Mobile or Fixed)
• Involve skin pricking (In-contact or
On-body)
• Data type (Environmental or
Touchless)
• Location of data integration
(Software-based or Hardwarebased integration)
• Location of data visualisation
16
(Standalone, etc.)
17. Data flow stages in Zeo Sleep Manager
Data Flow Stages
Data Collecting
Data Transmission
Data Saving (temporary
storage)
Data Storing (permanent
storage)
Data Analysis
Data Visualisation
Zeo Sleep Manager
Data Sharing
17
19. Second White Paper – Data integration methods
PCEHR
Individual analysis
Integrated
Analysis
20.
21.
22. Minimum Information about a
Self Monitoring Experiment (MISME)
EXPERIMENT
Who/Which part/
Where/When?
Body part
(FMA)
Sample
Device
How?
Name
Model
Manufacturer
Time
Technical
Specs
Location
Taxonomy
What
Measurement
Body structure
Body function
Around body
(based on WHO)
Method
Data
Raw
Where is
it stored
Procedures
Processed
Units
23. Self-Omics
• QS as an interface to the Human
Body
• How much information?
• People-as-sensors
• Making the personal public
• From population surveillance to
individual surveillance
Infography:
Institute for Health Technology Transformation
24. Benefits
If 10% adults USA began a
regular walking program, an
estimated $5.6 Billion in heart
disease could be saved.
25. Self-monitoring
• Project MUM-Size
– Study of very obese pregnant women – risk of complications
due to anesthesia during labor
– Using fitbit and social media support by research midwives in
the intervention group to prevent weight gain during
pregnancy
– User guide (Aria scale not suitable for pregnant women, limit
of 140 Kgs of weight)
27. Health Informatics aspects of QS
• Integration of QS data with EHR/
•
•
•
•
•
•
•
PHR
HIE of 1 - Blue Button and Blue
Button+
Meaningful use - V/D/T View/
Download/Transmit. Making sense
of data
Behaviour change
From QS early adopter to
mainstream motivated-self
Long-term or too-scary does not
work
Personalization
Designers or artists for data
presentation
28. Conclusion
Benefits
Challenges
• Privacy
• Security
• Education
of their health
• Cyberchondria
• Self-improvement
• Equity
• Risk profiling
• Regulation, accreditation
• Prevention
• Shift terciary à secondary • Role of the clinician
• Infrastructure needs
à primary à home care
• Therapeutic gap (ethics)
• Data donors for research
• Motivation
• Deepening understanding
29. References
• Almalki, M, Martin-Sanchez, F & Gray, K 2013, 'SelfQuantification: The Informatics of Personal Data Management
for Health and Fitness’, Institute for a Broadband-Enabled
Society (IBES).
• Almalki, M, Martin-Sanchez, F & Gray, K 2013. The Use of
Self-Quantification Systems: Big Data Prospects and
Challenges. Proceedings of HISA BIG DATA 2013 conference.
Accepted for publication at BMC Health Information Science
and Systems
29
30. UoM QS team
• Dr. Kathleen Gray (HaBIC)
• Manal Almalki (PhD candidate, HaBIC)
• Pilar Cantero (RA - HaBIC)
• Cecily Gibert (RA - HaBIC)
• Dr. Bernd Ploderer (Computing and Information Systems)
• Mark Whooley (MIS student)
• Matthew McGavern (MIS student)
• Prof. David Story (Chair of Anesthesia)
• Prof. Mary Wlodek (Physiology)