Maximizing the value of data, computing, data science in an academic medical center, or 'towards a molecularly informed Learning Health System. Given in October at the University of Florida in Gainesville

1. Data Harmonization for a Molecularly
Driven Health System
Warren A. Kibbe, Ph.D.
Professor, Biostats & Bioinformatics
Chief Data Officer, Duke Cancer Institute
warren.kibbe@duke.edu
@wakibbe
#DataSharing
#LearningHealthSystem
#DataHarmonization

2. Sections
• Learning Health Systems
• Data Commons
• Data Harmonization

3. The World is Changing
• Pace of Commercialization
• Reach of Markets
• Role of Data
• Change in Healthcare
• Change in Computing
• Societal Changes

4. Is the US able to keep up?

5. R&D By Country

6. US R&D Funding as share of GDP

7. R&D spending / STEM

8. How do we continue to innovate?

9. Data Science

10. Twitter impacts science

11. Eric Topol

12. Changes in Computing
• Converged devices
• Converged IT
• Ubiquity of devices, data, mHealth

13. 2017200220072012
10/23/2001
(~5yrsold)
1/9/2007
(~10yrsold)
iPod(10GBmax)
iPhone(EDGE,16GBmax)
9/16/1999
(~3yrsold)
802.11bWiFi
4/3/2010
(~13yrsold)
iPad(EDGE,64GBmax)
4/23/2005
(~8yrsold)
9/26/2006
(~9yrsold)
7/15/2006
2/7/2007
Google
Drive
4/24/2012
(~15yrsold)7/11/2008
(~11yrsold)
iPhone3G
(16GBmax)
9/12/2012
(~15yrsold)
iPhone5(LTE,128GBmax)
Google
Baseline
3/9/2015
(~18yrsold)
Apple
ResearchKit
HTCVRHeadset
4/5/2016
(~19yrsold)
7/14/2014
(~17yrsold)
NextGen
Courtesy of Jerry Lee, NCI
Changes in Technology

14. Pace of Technology Adoption

15. Changes in Commercialization

16. Changes in Oncology
• Cancer is a grand challenge
• Anatomic vs molecular classification
• Health vs Disease

17. Understanding Cancer
• Precision medicine will lead to fundamental
understanding of the complex interplay between
genetics, epigenetics, nutrition, environment and
clinical presentation and direct effective,
evidence-based prevention and treatment.
Ramifications across many aspects of health care

18. IOM
(Now NAM)
Report
2006-11

19. NAM
Workshops

20. “Science, informatics, incentives, and
culture are aligned for continuous
improvement and innovation, with best
practices seamlessly embedded in the
delivery process and new knowledge
captured as an integral by-product of the
delivery experience.”
—Institute of Medicine
LEARNING HEALTH SYSTEMS

21. Another imperative is that such systems
do their work:
• Transparently (how does one learn
without well documented processes?)
• Reproducibly (good practices must
always be repeatable at scale and
scientifically reproducible)
• Only with the above can the science in
“data science” be done with sufficient
rigor
LEARNING HEALTH SYSTEMS

22. ASSEMBLE
ANALYZE
INTERPRET
FEEDBACK
CHANGE
LEARNING HEALTH SYSTEMS

23. Learning Health Systems in NEJM

24. Goals
• Contain rising cost of healthcare
• Maximize the value of care
• Increase public discourse and
marketplace for healthcare

25. Drivers
• Decision Making is too complex
• Clinical decisions are based on
practice, not evidence
• Inefficiency and waste in healthcare

26. Human cognitive capacity is constant

27. Lack of Evidence

28. EHRs and the Learning Health System

29. LHS definition

30. Problems for LHS to solve

31. Inefficient Healthcare

32. Poor Health in spite of high expenditures

33. Curve hasn’t improved

34. 2015

35. View from 2006

36. EHRs are now ubiquitous
But evidence-driven decision support
remains a future vision

37. Hope
Cloud computing, data commons,
service-based computing provide some
powerful tools for solving data access,
data analysis, data analytics, and data
visualization problems at scale,
securely.

38. Sebastian Thrun

39. So what is a Data Commons

40. Commons Topology
Compute Platform: Cloud or HPC
Services: APIs, Containers, Indexing,
Software: Services & Tools
scientific analysis tools/workflows
Data
“Reference” Data Sets
User defined data
DigitalObjectCompliance
App store/User Interface
PaaS
SaaS
IaaS
https://datascience.nih.gov/commons

41. Commons Compliance
• Treat products of research – data,
methods, papers etc. as digital
objects
• These digital objects exist in a
shared virtual space
• Digital object compliance through
FAIR principles:
– Findable
– Accessible (and usable)
– Interoperable
– Reusable

42. Data Sharing and the FAIR Principles
FAIR –
Making data
Findable,
Accessible,
Attributable,
Interoperable,
Reusable,
and provide Recognition
Force11 white paper
https://www.force11.org/group/fairgroup/fairprinciples

43. “The Commons is an effort at
creating a sharing economy and
for building community. We hope
for a more cost effective and
productive research
environment while bringing
people together in a unique
way.“
Phil Bourne

44. 44
Blue Ribbon Panel Report
Cancer Moonshot℠ Blue Ribbon Panel
“The Cancer Moonshot Task Force was
directed to consult with external experts
from relevant scientific sectors, including
the presidentially appointed National
Cancer Advisory Board(NCAB).
A Blue Ribbon Panel of scientific experts
was created to advise the NCAB.”

45. Vision:
Enable the creation of a Learning Healthcare System for
Cancer, where as a nation we learn from the contributed
knowledge and experience of every cancer patient. As
part of the Cancer Moonshot, we want to unleash the
power of data to enhance, improve, and inform the journey
of every cancer patient from the point of diagnosis
through survivorship.

46. A National Cancer Data Ecosystem
Cancer Research
Data Commons
SBG CGC
Broad FireCloud ISB CGC
Courtesy NCI-CBIIT

47. Data Commons Framework – What Is It?
47
Modular Components
Secure user authentication and authorization
Metadata validation and tools
Domain-specific, extensible data models and dictionaries
API and container environment for tools and pipelines
Access to computational workspaces for storing data, tools, and
results
Reusable, expandable
framework for a Data
Commons
Core principles and
structures for a Data
Commons
Set of modular
components that can be
leveraged across Data
Commons

48. Narrow Middle Architecture (End-to-End Design)
1. AuthN / AuthZ
2. Metadata validation
3. Extensible data model
4. APIs for containers, workflows & tools
5. Workspaces
science outdata in
Courtesy Bob
Grossman, U. Chicago

49. 49
NCI Cancer Research Data Commons (CRDC) - Concept
NCI Scope: “Create a data
science infrastructure necessary
to connect repositories, analytical
tools, and knowledge bases”
Data commons co-locate data,
storage and computing
infrastructure with commonly
used services, tools & apps for
analyzing and sharing data to
create an interoperable resource
for the research community.*
*Robert L. Grossman, Allison Heath, Mark Murphy, Maria Patterson and Walt Wells, A
Case for Data Commons Towards Data Science as a Service, IEEE Computing in Science
and Engineer, 2016. Source of image: The CDIS, GDC, & OCC data commons
infrastructure at the University of Chicago Kenwood Data Center.

50. 50
Data Commons Framework
Clinical Proteomics ImagingGenomics Immuno-
oncology
Animal Models Cancer Biomarkers
NCI Cancer Research
Data Commons
SBG CGC
Broad FireCloud ISB CGC
Elastic
Compute
Query
Visualization
Clinical Proteomics Tumor
Analysis Consortium*
Tool
Deployment
The Cancer Imaging Archive*
TCIA
Web
Interface
APIs Data
Submission
Authentication
& Authorization
Authentication
& Authorization
Data Models &
Dictionaries
Computational
Workspaces
Data Contributors and Consumers
Tool
Repositories
Metadata
Validation
& Tools
Analysis
Courtesy NCI-CBIIT

51. Gen3 Data Commons

52. Gen3 Data Commons

53. Gen3 Data Commons

54. NCI Genomic Data Commons

55. NCI Genomic Data Commons

56. NCI Genomic Data Commons

57. Data Harmonization
• The process of semantic and
syntactic mapping of data to a set of
definitions, predefined data
elements, data model.
• Validation and Harmonization of
primary and secondary data is crucial
to enable analysis and reuse

58. Spanning the Semantic Chasm of Despair
Building a Translational Bridge
CD2H
Thanks to Melissa Haendel

59. Project Highlight: Harmonizing clinical data models
Sentinel
I2b2/ACT
OMOP
PCORNET
▪ Different countries use different “outlets”.
▪ There is a need for travel adapters.
The Solution:
▪ Use a converter between various adapters.
▪ Allow researchers to ask a question once and
receive results from many different sources

60. Project Highlight: LOINC2HPO
◆ Develop a software tool to map
LOINC codes to HPO terms
◆ Develop software to convert
EHR observations into HPO
terms for use in clinical
research
Steps
Develop a tool for converting LOINC laboratory codes and values into more
phenotypically meaningful language (Human Phenotype Ontology) to allow for
translational interoperability and new analytics
2657-5 “Nitrite [Mass/volume] in Urine” Numeric
20407-3 “Nitrite [Mass/volume] in Urine by Test
strip”
Numeric
32710-6 “Nitrite [Presence] in Urine” Positive/Negati
ve
5802-4 “Nitrite [Presence] in Urine by Test strip” Positive/Negati
ve
50558-6 “Nitrite [Presence] in Urine by
Automated test strip
Positive/Negati
ve
LOINC Outcome
HPO: Nitrituria

61. INSERT CDE Browser Screenshot?
CIBMTR Center for Cancer
Research
Over 35 NCI Programs, Plus
Cancer Centers and Consortia
GDC

62. Data Sharing Index
• We need metrics for data, software,
algorithm use, usability, conformance
• Data sharing stimulates science,
innovation, commercialization
• Providing recognition and attribution
to data providers and software &
algorithm builders is critical for a
robust data sharing ecosystem
• Support and measure FAIRness!

63. Questions?
Warren Kibbe, Ph.D.
warren.kibbe@duke.edu
@wakibbe