Health and Biomedical Big Data
for Translational Research
Yu-Chuan (Jack) Li, M.D., Ph.D., FACMI Professor in Biomedical Informatics,
Dean, College of Medical Science and Technology
Taipei Medical University
Taipei Medical University
6000 students, 620 faculty members, 7 colleges
3 teaching hospitals; 3150 total beds
All JCI Accredited, in the Taipei metropolitan area
Closest to the world’s 2nd highest building
– Taipei 101
School of Medical Technology and Biotechnology
Graduate Institute of Neural Regenerative Medicine
Graduate Institute of Biomedical Informatics
Graduate Institute of Cancer Biology and Drug Discovery
Graduate Institute of Translational Medicine
College of Medical Science and
Technology
Total: 300 undergrad and 150 M.S. and
Ph.D. students
Computer Methods and
Programs in Biomedicine
International Journal for
Quality in Health Care
Editor-in-Chief
ISQua / OUP Elsevier
2000 submissions, 340 paper published / year
Defining Big Data
Big Data is a collection of data sets so
large and complex that it becomes
difficult to process using on-hand
database management tools or
traditional data processing applications.
http://en.wikipedia.org/wiki/Big_data
Elements of “Big Data”
The degree of complexity within the
data set
The amount of value that can be
derived from innovative vs. traditional
analysis techniques
The use of longitudinal (time-series)
information supplements the analysis
http://mike2.openmethodology.org/wiki/Big_Data_Definition
Challenges Biomedical BD
Locating/accessing data and software tools
Standardizing data and metadata
Extending policies for sharing BD
Organizing, managing, and processing
Developing new methods for analyzing &
integrating BD
Training researchers who can use BD
effectively
From http://bd2k.nih.gov/about_bd2k.html#sthash.jNfoKrca.dpbs
Defining Translational Research
T1: translation to humans
T2: translation to patients
T3: translation to practice
T4: translation to population
Big Data Sources for Translational
Bioinformatics Research
T1: genomics, SNPs, expression…
T2: clinical trials, registries, EHR…
T3: EBM, guidelines, protocols…
T4: Insurance claims, OTC drugs
consumptions, social networks…
SNP – Single Nucleotide Polymorphism
EHR – Electronic Health Record
EBM – Evidence-base medicine
OTC – Over the counter
GM Weber; KD Mandl; IS Kohane. JAMA. Published online May 22, 2014. doi:10.1001/jama.2014.4228
TBI for T1
Cancer new drug design and discovery
Target therapy like EGFRi – gefitinib,
afatinib
Druggable genome
Drug-Gene-Interaction database (DGIdb)
TBI for T2
Clinicaltrial.gov
SEER (Surveillance, Epidemiology, and
End Results) from NCI – outcome
research
Association between complex and
Mendelian diseases – Blair et. al. Cell 2013
NHIRDB Taiwan
DzMap (The Disease Map)
CAMA (Cancer Association Map Animation)
44,033 studies found for: cancer
7.3 million cancer patients
Data set
+ Tools
DR Blair et al. Cell. 155(1), 26 Sep 2013, P.70–80
The Disease Map
http://disease-map.net
3 years, 1 billion outpatient visits
A window size of 12 months
25
26
Hands-on session
• SEER - demo
• NHIRDB - files
• CAMA - demo
• DzMap - demo
http://seer.cancer.gov/manuals/
2014/appendixc.html
Site-Specific Collaborative Staging
Codes
http://seer.cancer.gov/tools/codi
ngmanuals/
TBI for T3
EBM paradigm
Guideline databases
Decision support systems
Behavioral change for clinicians
TBI for T4
Patient support groups –
patientslikeme.com
Social network – FB, Twitter, Weibo…
Behavioral guidance for large groups of
patients – PCEHR in Australia
Research Paradigm from Data Perspective
What does it mean
Translational research can be achieved
through bio-medical informatics
With the Big Data approach,
translational research can be sped-up,
value-added and more innovative
Collect, share and utilize – bigger is
better – major breakthroughs rely on it
Exercise
Try to find on CAMA -
http://203.71.86.98/web/runq16.html#
Digestive diseases associated with
colon cancer in female across all ages
Download the and unzip the files
What are the diagnoses of patient with
pid of 9f6766287dd1bbb1bbf1ee767bbc738f
What are the drugs given? Cancer_2011_OPD_Data-mapped-new
Cancer_2011_Drug-mapped-new
Answers-Dx
Malignant bronchus and lung neoplasm,
unspecified site
Secondary bone malignant neoplasm
Acute upper respiratory infections of
unspecified site
Death occurring in less than 24 hours
from onset of symptoms, not otherwise
explained
Answers-Rx
SODIUM BICARBONATE
SODIUM CHLORIDE
SODIUM CHLORIDE
DOPAMINE HCL
SODIUM CHLORIDE
CEFTRIAXONE (DISODIUM 3.5 H2O)
Thank you for your attention