This summer, I worked with Saurav Bose and Dr. Aaron Masino in the Masino Lab at the Children’s Hospital of Philadelphia. The Masino Lab is computationally focused with an overarching goal of improving child health and healthcare through the development and application of artificial intelligence (AI) methods.

Access to labeled data is a challenge in healthcare data research that often limits the development and deployment of intelligent ML-based predictive analytics in the clinic. The contrastive learning framework is a novel methodology proposed by Chen et al. (SimCLR) that performs model training in a fully unsupervised, architecture-agnostic way. The objective of this research project was to test its utility in predicting Acute Respiratory Distress Syndrome (ARDS) using frontal chest X-ray images.