Back

Speaker "Sujay Kakarmath" Details Back

 

Topic

Importance of clinical context in evaluating machine learning-based solutions for healthcare

Abstract

Novel predictive algorithms developed using machine learning methods are often optimized to achieve the best area under the receiver operating characteristic curve (AUC). However, this metric is often not relevant clinically. How, then, can health professionals make conclusions about the real utility of an algorithm? The Algorithm Science team at Partners Connected Health invests a great deal of time thinking about the right questions, working out potential pitfalls and developing best practices in evaluating machine learning-based solutions.


Who is this presentation for?
Business leaders and clinical end-users of machine learning and AI solutions


Prerequisite knowledge:
None


What you'll learn?
This presentation will provide the audience with a better understanding of how to best gauge the clinical utility of an algorithm.

Profile

Dr. Kakarmath is a digital health scientist at Partners Healthcare Pivot Labs and an Instructor at Harvard Medical School. His research is focused on the evaluation of the clinical utility of digital health solutions, including machine learning and artificial intelligence-based products. Dr. Kakarmath's team works closely with technology innovators from academia, startups and industry giants to guide the ideation, design, prototyping, validation, and deployment of digital health solutions. His work has been published in prestigious journals and showcased at major academic conferences such as those of the American Academy of Neurology, the American Medical Informatics Association, the International Society for Pharmacoeconomics and Outcomes Research, the Connected Health Conference, Precision Medicine Summit and HIMSS.