Luca Foschini Data Founder & Health Data Scientist

Luca Foschini, Ph.D. is a researcher, data scientist, and executive who has spent the last decade building intelligent systems to measure human health and behavior. He is the co-founder and Chief Data Scientist of Evidation Health. Evidation Health is a new kind of health and measurement company that provides the world’s most innovative biopharma and health care companies the technology and guidance they need to understand how everyday behavior and health interact. Previously he has held R&D positions as Google,, and the CERN. 

Luca’s research interests have been at the intersection of computer science and life sciences and include the development of efficient algorithms for collecting and analyzing massive time series data, machine learning methods for mining human metrics and bio-signals, and research on privacy and security. Luca has co-authored more than 30 academic papers and conference abstracts, and several patents.  More recently, Luca had an active role in working with regulatory bodies to facilitate the use of commercial wearable devices in clinical research.

Research Interests

Wearable and Sensors in Medicine

Wearable and Sensors in Medicine

There is a growing opportunity to use remote sensing technologies to better measure outcomes in clinical research settings. I’ve served as expert with the Clinical Trial Transformation Initiative (CTTI) to help release Recommendations for Advancing the Use of Mobile Technologies for Data Capture & Improved Clinical Trials (see: Press releasepanel) and the Duke-Margolis Center for Health Policy to frame applications to Real World Evidence Generation (See my talk at National Academy of Science, Engineering, and Medicine).

Digital Biomarkers

Digital Biomarker Discovery

Continuously collected Patient Generated Health Data (PGHD) can enable a low-burden way for a patient to keep track of their health status and disease progression. This is made possible by extracting digital biomarkers from PGHD, which need to be validated against patient reported outcomes (PRO) or other clinical assessments. See my interviews with Rock Health and Bio-IT World on how we think about digital biomarker discovery, and episode on Italian journalistic program Report (in Italian).

Machine Learning on Time Series and Bio-signals

Machine Learning on Time Series and Bio-signals

Machine learning has had a tremendous impact on many branches of science and technology. Machine learning research on time series data is just at the beginning, but it has the potential to automate learnings from multi-channel, high-frequency sensor streams for large population. See our work at SIGKDD 2017 for an overview of challenges and opportunities, and our research on computing individualized cardiovascular response signatures presented at NeurIPS2018/ML4H

Contact Me

I like to speak.