Interactive visualization of hospital readmission data and model insights
When a patient is incorrectly predicted to be readmitted, this may lead to:
When a high-risk patient is incorrectly predicted as low-risk, this may lead to:
We've carefully balanced our model to minimize both false positives and negatives, with a slight preference toward minimizing false negatives due to the higher potential harm. Additionally, we emphasize that this tool should support, not replace, clinical judgment.