The most popular buzz word nowadays in the technology world is “Machine Learning (ML).” Most economists and business experts foresee Machine Learning changing every aspect of our lives in the next 10 years through automating and optimizing processes. This is leading many organizations including drug companies to implement Machine Learning into their businesses. The presentation will start with the introduction of basic concept of Machine Learning, the computer science technology that provides systems with the ability to learn without being explicitly programmed, and it will discuss what it means by “without being explicitly programmed”. The presentation will also introduce basic ML algorithm -SVM, Decision Tress, Regression, Artificial Neural Network (ANN), and DNN. The presentation will also discuss the impact and potential of Machine Learning in our daily lives and pharmaceutical industry. The presentation will show how CDISC data can be a perfect match on Machine Learning implementation. In this Machine Learning/AI driven process, data is considered as the most important component. 80 to 90 % of works in Machine Learning is preparing data. Since FDA mandated CDISC standards submission as of Dec 17th, 2016, all the clinical trial data are prepared in CDISC SDTM and ADaM data format. The presentation will show how CDISC data is better choice than Real World Evidence (RWE) data for ML model. The presentation will also show how pharmaceutical industry use CDISC data to build ML model and apply ML model for Real World evidence. Finally, the presentation will show how Pharma industry can use their own in-house data and Machine Learning to build innovative, data-driven business models.

1. 1
Perfect Partnership – Machine Learning & CDISC
Kevin Lee
Director of Data Science

2. 2
The views and opinions expressed in the following PowerPoint
slides are those of the individual presenter and should not be
attributed to Drug Information Association, Inc. (“DIA”), its
directors, officers, employees, volunteers, members, chapters,
councils, Communities or affiliates, or any organization with
which the presenter is employed or affiliated.
These PowerPoint slides are the intellectual property of the
individual presenter and are protected under the copyright laws
of the United States of America and other countries. Used by
permission. All rights reserved. Drug Information Association,
Drug Information Association Inc., DIA and DIA logo are
registered trademarks. All other trademarks are the property of
their respective owners.
Disclaimer – Content Slide

3. 3
Q1 : World’s Largest Transportation Company?
In January 2018, there is about
7,000,000 “drivers”.
The uber is operating in 600
cities in 78 countries.
There has been 5 billion rides.
Uber – about 85% of US
hailing market.

4. 4
Q2 : World’s Largest Accommodation Provider?
There are more than 4.5 million Airbnb listings in 81,000 cities.
Airbnb hosts have earned $41 billion in 10 years.
As of June’17, Airbnb’s total valuation is about $31 billion.

5. 5
Common Characteristics of Exponential
Organization
Data
Algorithm (Machine Learning/AI)
Exponential &
Scalable Growth

6. 6
How Data and Algorithms help the exponential
growth
More Data
Better
Algorithms
(ML/AI)
Better
Products
More
Users

7. 7
What is Machine Learning?
An application of artificial
intelligence (AI) that
provides systems the
ability to automatically
learn and improve from
experience without being
explicitly programmed.

8. 8
Explicit Programming vs Machine Learning
Explicit Programming Machine Learning

9. 9
How does Human Learn? - Experience

10. 10
How does Machine Learn?
Algorithm
Test Data (cats)
Real Data
cat

11. 11
More data,
better model
X
Y
X
Y Different data,
different model

12. 12
Data Quality in Machine Learning
Garbage in Garbage out

13. 13
The economist – The world’s most valuable
resource is no longer oil, but DATA.

14. 14
Can Pharmaceutical be the exponential organization?
Pharm =
exponential organization
????

15. 15
Gold mines - Clinical trial data in
Pharmaceutical industries
Clean – Pharma companies spent a
lot of hours to clean the data.
Unbiased - Prospective study,
randomized
Blinded – double-blinded
Structured
Metadata
Standards – CDISC
Already pharmaceutical companies
already owned the data.
Data in Pharmaceutical industry

16. 16
Main purpose - for the submission
No or limited analysis after submission
Do not know exactly where clinical trial data is
Limited/No access
No CDR (Central Data Repository)
What is the reality of clinical trial data

17. 17
Clinical Trial Data with ML/AI for Patients/Doctor
Machine Learning/ AI
CDISC
Clinical
Trial
Data
Patients/
Doctors/
Healthcare
Providers/
Drug
Companies

18. 18
CDISC with Machine Learning
Algorithm
CDISC
(Test
Data)
Real
World
Data
Prediction

19. 19
CDISC + Machine
Learning
More than drug
Drugs &
Experience
Conclusion
What Apple gives us?

20. 20
Kevin Lee
Director of Data Science
kevin.kyosun.lee@gmail.com
Twitter @HelloKevinLee
LinkedIn https://www.linkedin.com/in/HelloKevinLee/
Join the conversation #DIA2018
Thank You