Volumetric risk quantified for 3d medical image segmentation
Quantifying Volumetric Risk: Class-Aware Asymmetric Weighted Conformal Prediction for 3D Medical Image Segmentation
Computer Vision and Pattern Recognition
Summary
Medical image segmentation helps doctors identify different tissues and diseases by labeling each part of a 3D scan. The authors found that existing methods don’t give clear measures of uncertainty, especially when the images are different from what the model was trained on. They introduced a new way to calculate more trustworthy and class-specific uncertainty ranges that adapt when conditions change. Their approach ensures reliable predictions for each tissue type and helps create clearer medical reports.
What this means in practice
- •For medical imaging software teams: Produce more reliable 3D segmentations with calibrated uncertainty that adapts to changing patient data distributions.
- •For radiology reporting developers: Generate radiology reports that explicitly incorporate uncertainty information, improving communication with clinicians under distribution shifts.
Authors
Shadi Alijani, Fereshteh Aghaee Meibodi, Homayoun Najjaran
Abstract
Reliable volumetric segmentation is critical for clinical diagnostics, yet foundation models such as MedSAM remain deterministic and lack calibrated uncertainty under distribution shift. Existing conformal prediction methods offer statistical guarantees but are frequently applied in 2D and assume symmetric error distributions, so they do not capture the class-specific biases that arise in 3D multi-class segmentation. We propose Class-Aware Asymmetric Weighted Conformal Prediction (CA-WCP), which combines latent-space density-ratio weighting for covariate shift with directional quantiles for the lower and upper volume bounds, and scales each bound by a class-specific asymmetry factor derived from validation-set false-positive and false-negative rates. We prove that CA-WCP retains the weighted-exchangeability marginal coverage guarantee for every class, and we evaluate it on 3D brain tumor segmentation (BraTS 2020) and on a synthetic multi-organ CT benchmark constructed under covariate shift. On both benchmarks the 95\% Clopper--Pearson interval for the observed coverage of CA-WCP contains the nominal 90\% level for every semantic class, while interval width is reduced by 8--14\% relative to symmetric weighted conformal prediction. We further encode the calibrated intervals into structured prompts for a multimodal large language model to produce uncertainty-conditioned radiology reports, linking distribution-shift-aware uncertainty quantification to interpretable clinical communication.