Benchmarking Offline Multi-Objective Reinforcement Learning in Critical Care

A. Bansal, and D. Sharma

arXiv:2512.08012, Dec 2025

EIT for ARDS lung injury prediction: comparable to CT and enhanced by randomized ventilation data

This report investigated the effectiveness of electrical impedance tomography (EIT) versus computed tomography (CT) in predicting acute respiratory distress syndrome (ARDS) lung injury. It also explored how randomized ventilation strategies improved model robustness.

Outperforming Classical Machine Learning Models With Physics-Informed Machine Learning for Mechanical Ventilation

This report built a physics-informed variational autoencoder to generate synthetic data for mechanical ventilation.

Integration of Machine Learning with bioimaging-based feature ranking for predicting ventilator-induced lung injury

Research Opportunity Program Poster Fair, University of Toronto (March 12, 2025).