4 papers
Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories
Yixuan Yang, Mehak Arora, Ryan Zhang +10
We present Clin-JEPA, a multi-phase co-training framework for joint-embedding predictive (JEPA) pretraining on EHR patient trajectories. JEPA architectures have enabled latent-spac…
Automated detection of pediatric congenital heart disease from phonocardiograms using deep and handcrafted feature fusion
Abdul Jabbar, Ethan Grooby, Yang Yi Poh +5
Congenital heart disease (CHD) is the most common type of birth defect, impacting about 1% of live births worldwide. Echocardiography, the gold-standard diagnostic method, is costl…
Congenital Heart Disease Classification Using Phonocardiograms: A Scalable Screening Tool for Diverse Environments
Abdul Jabbar, Ethan Grooby, Jack Crozier +7
Congenital heart disease (CHD) is a critical condition that demands early detection, particularly in infancy and childhood. This study presents a deep learning model designed to de…
Skin Cancer Machine Learning Model Tone Bias
James Pope, Md Hassanuzzaman, William Chapman +5
Background: Many open-source skin cancer image datasets are the result of clinical trials conducted in countries with lighter skin tones. Due to this tone imbalance, machine learni…