- Analytics use case(s): Patient-Level Prediction
- Study type: Methods Research
- Tags: Prediction models, PatientLevelPrediction, Federated learning
- Study lead: Jenna Reps
- Study lead forums tag: Jenna Reps
- Study start date: 01-09-2026
- Study end date: -
- Protocol: -
- Publications: -
- Results explorer: -
What is the best strategy for model development when an outcome is rare and data are spread out across different sites? Should you do local model vs federated learning vs transfer learning? Is there a cetain data size/outcome count where one approach is better than others?