Reproduce “Simulating time to event prediction with spatiotemporal echocardiography deep learning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Rohan Shad, Nicolas Quach, Robyn Fong, Patpilai Kasinpila, Cayley Bowles, Kate M. Callon, Michelle C. Li, Jeffrey Teuteberg and 3 more
The authors test whether echocardiogram video can support time-to-event prediction without training the model to predict ejection fraction (EF). They train video networks with censored survival labels simulated from expert EF readings. Across four simulated datasets, predicted individual partial log-likelihoods closely track true values, demonstrating recovery of a designed signal, not clinical survival prediction or prognostic information beyond EF.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.