Relaxing the definition of equivalent mathematical expressions to get simpler and more interpretable models

The Cava laboratory is a research group interested in improving the interpretability and fairness of predictive models deployed in healthcare settings. We create and study algorithms that can embed these notions when working with health data.
The lab is part of the Computational Health Informatics Program at Boston Children’s Hospital, affiliated with Harvard Medical School.
Join the Team! If you’re looking to do research in the areas of machine learning, interpretability, fairness, and health informatics, see the opportunities here.
Recent Papers
Investigation of a Novel Noninvasive Risk Analytics Algorithm With Laboratory Central Venous Oxygen Saturation Measurements in Critically Ill Pediatric Patients
Critical Care Explorations
Intersectional consequences for marginal fairness in prediction models of emergency admissions
Preprint
Electrocardiogram-based deep learning to predict mortality in paediatric and adult congenital heart disease
European Heart Journal
Recent Posts
About our recent HUMIES award-winning algorithm for clinical prediction models
A new perspective on how this social theory relates to fair machine learning.