Matthew McDermott

Matthew McDermott, PhD

Berkowitz Family Living Laboratory Postdoctoral Research Fellow

Dr. Matthew McDermott received his PhD in Computer Science from MIT, working with Professor Peter Szolovits on clinical and biomedical representation learning; in particular injecting prior domain knowledge and leveraging unsupervised data to build clinically meaningful representations of medical and biological data modalities. His research has investigated topics ranging from semi-supervised biomedical regression problems leveraging partially labeled data; domain-specific pre-training methods for clinical language, intensive care unit numerical timeseries, and protein sequences; and novel frameworks for inducing inductive bias into pre-training algorithms. Prior to his PhD, Matthew studied mathematics at Harvey Mudd College for his undergraduate degree, and worked as a software engineer in data engineering at Google.

Using machine learning to develop smart reflex testing protocols.
Authors: McDermott M, Dighe A, Szolovits P, Luo Y, Baron J.
J Am Med Inform Assoc
View full abstract on Pubmed
Clinical Artificial Intelligence: Design Principles and Fallacies.
Authors: McDermott MBA, Nestor B, Szolovits P.
Clin Lab Med
View full abstract on Pubmed
Using Machine Learning to Develop Smart Reflex Testing Protocols.
Authors: McDermott M, Dighe A, Szolovits P, Luo Y, Baron J.
ArXiv
View full abstract on Pubmed