A new study published on arXiv explores the optimal historical data window for predicting hospital readmissions. Researchers found that for unstructured clinical notes, a shorter window of three to six months prior to surgery yielded the best predictive performance. In contrast, structured data showed improved performance with longer time windows, plateauing after twelve months. These findings suggest that simply using more historical data does not always lead to better machine learning predictions, offering guidelines for optimizing readmission prediction models. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT Provides data-driven guidelines for optimizing predictive models in healthcare, potentially improving patient outcomes and reducing costs.
RANK_REASON Academic paper on machine learning for healthcare applications.