New Digital Twin Model Predicts Patient Trajectories with Irregular Data

Researchers have developed a new model called CHDT, designed to generate counterfactual human digital twins for predicting long-term patient trajectories under various interventions. This model explicitly accounts for irregular visit intervals and cross-mediator dependencies, which are often overlooked in existing prediction models. CHDT demonstrated improved fidelity in long-horizon patient-state prediction compared to baseline models when tested on real-world Alzheimer's Disease Neuroimaging Initiative (ADNI) data and simulated ALPACA cohorts. AI

IMPACT This research could advance in silico clinical trial design and personalized medicine by enabling more accurate long-term patient trajectory predictions.

RANK_REASON The cluster contains a research paper detailing a new model and its evaluation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alpaca
• Alzheimer's Disease Neuroimaging Initiative

• paper
• other

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