Extracting Gems from Digital Health Data (Technion and Teva)

We will have two lectures (approx. 35-40 minutes each) in Hebrew. Light refreshments will be served before the first lecture.

Estimated schedule:
1745 refreshments at the J&J conference hall (3rd floor)
1800 First lecture
1850 Second lecture

First lecture:

Lecturer: Dr. Kira Radinsky, chief scientist and director of data science at eBay.
Title: Mining Electronic Health Records and the Web for Drug Repurposing

We jointly harness large-scale electronic health records and feasible conceptual links among concepts drawn from Wikipedia to provide guidance about drug repurposing -- the process of applying known drugs in new ways to treat diseases. We claim that researchers decide on exploratory targets for repurposing based on trends in research and observations on small numbers of cases, leading to potentially costly biases of focus and neglect. We propose a methodology for identifying potential drugs to be repurposed via mining 10 years of nation-wide medical records of more than 1.5 million people and extracting medical knowledge from Wikipedia to help reduce spurious correlations.The resulting system seeks to identify potential biological processes to justify potential influences between medications and target diseases via links on a graph constructed from Wikipedia data. We show results of the system on two studies of drug repurposing for hypertension and diabetes. In both cases, we present drug families identified by the algorithm which were previously unknown. Clinical opinion by experts in the field and clinical trials on those drug families provide evidence that the drugs discovered by the system show promise for repurpose for those diseases.

Kira Radinsky is the chief scientist and director of data science at eBay, where she is building the next-generation predictive data mining, deep learning, and natural language processing solutions that will transform ecommerce. She also serves as a visiting professor at the Technion, Israel’s leading science and technology institute, where she focuses on the application of predictive data mining in medicine. Kira cofounded SalesPredict (acquired by eBay in 2016), a leader in the field of predictive marketing—the company’s solutions that leveraged large-scale data mining to predict sales conversions. One of the up-and-coming voices in the data science community, Kira is pioneering the field of web dynamics and temporal information retrieval. She gained international recognition for her work at Microsoft Research, where she developed predictive algorithms that recognized the early warning signs of globally impactful events, including political riots and disease epidemics. She was named one of MIT Technology Review’s 35 young innovators under 35 for 2013 and one of Forbes’s 30 under 30 rising stars in enterprise technology for 2015; in 2016, she was recognized as woman of the year by Globes. Kira is a frequent presenter at global tech events, including TEDx and the World Wide Web Conference, and she has published in Harvard Business Review.

Second lecture:

Lecturer: Dr. Shai Fine, Sr. Director, Analytics & Big Data at Teva Pharmaceutical Industries.
Title: Digital Health in the age of Big Data; from Wearable Sensors to Deep Learning

This talk will take you on a journey to the Digital Health revolution and practical use-cases of Personalized Medicine in Teva. We will start with exploring a use case of algorithms integrated into digital sub-study (in Central Nervous System therapeutic area). We will demonstrate how we perform analysis of a movement disorder by methods of signal processing on data from sensors and data collected by eDiary. Then, an analysis of Time Series data by Deep Learning methods will be shown, towards solving a state detection problem and detecting sleep/awake patterns. Finally, we will show how we leverage Machine Learning to assess the risk of an asthmatic patient to experience asthma exacerbation(s) in the future; Advanced ML methods are being used to build several predictive models, using data from smart medical devices, data from wearables, as well as patient-specific data from Electronic Medical Records (EMRs).

Shai has over 23 years of experience at the top Technologies companies in the world (IBM & Intel). Listed 10 Patents and 33 articles published with over 2,500 citations. Shai leads the Analytics & Big Data team within the Personalized & Predictive Medicine @Teva - responsible for harnessing advanced machine learning methods in order to develop algorithms associated with health solutions, response prediction and diagnostics. Prior to joining Teva, Shai was the Principal Engineer at Intel Labs, Senior Machine Learning Technical Lead and head of the Deep Learning Capstone activities at Intel Collaboration Research Institute for Computation Intelligence. Before that, Shai was Senior Manager, Analytics Dep. at IBM. Shai holds a PhD in Computer Science from the Hebrew University of Jerusalem.