Uncertainty-Calibrated Prediction of Randomly-Timed Biomarker Trajectories with Conformal Bands
This work addresses the need for safe and reliable prediction of biomarker trajectories in healthcare, particularly for Alzheimer's disease, but it is incremental as it extends existing conformal prediction methods to a new setting.
The paper tackles the problem of uncertainty in predicting biomarker trajectories from clinical data, which poses risks for clinical deployment, by introducing a conformal method that produces prediction bands guaranteed to cover unknown trajectories with a user-prescribed probability; the method consistently achieves desired coverage, yields tighter bands than baselines, and identifies 17.5% more high-risk subjects for Alzheimer's disease progression.
Despite recent progress in predicting biomarker trajectories from real clinical data, uncertainty in the predictions poses high-stakes risks (e.g., misdiagnosis) that limit their clinical deployment. To enable safe and reliable use of such predictions in healthcare, we introduce a conformal method for uncertainty-calibrated prediction of biomarker trajectories resulting from randomly-timed clinical visits of patients. Our approach extends conformal prediction to the setting of randomly-timed trajectories via a novel nonconformity score that produces prediction bands guaranteed to cover the unknown biomarker trajectories with a user-prescribed probability. We apply our method across a wide range of standard and state-of-the-art predictors for two well-established brain biomarkers of Alzheimer's disease, using neuroimaging data from real clinical studies. We observe that our conformal prediction bands consistently achieve the desired coverage, while also being tighter than baseline prediction bands. To further account for population heterogeneity, we develop group-conditional conformal bands and test their coverage guarantees across various demographic and clinically relevant subpopulations. Moreover, we demonstrate the clinical utility of our conformal bands in identifying subjects at high risk of progression to Alzheimer's disease. Specifically, we introduce an uncertainty-calibrated risk score that enables the identification of 17.5% more high-risk subjects compared to standard risk scores, highlighting the value of uncertainty calibration in real-world clinical decision making. Our code is available at github.com/vatass/ConformalBiomarkerTrajectories.