Latent neural differential equations predict blood clot growth from sparse data

Researchers have developed a novel computational framework utilizing latent neural differential equations to predict blood clot growth from limited patient data. This method effectively infers unknown model parameters and forecasts thrombosis progression, showing improved accuracy with more available observations. The study compared seven probabilistic methods, with stochastic neural ordinary differential equations (SNODE) and stochastic neural functional differential equations (SNFDE) demonstrating the best performance in parameter inference and future clot-growth forecasting. AI

IMPACT This framework could enable more personalized and accurate medical predictions for conditions like thrombosis.

RANK_REASON Academic paper detailing a new computational framework for medical prediction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• latent neural differential equations
• SNFDE
• stochastic neural functional differential equations
• stochastic neural ordinary differential equations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →