arXiv:2607.22727v1 Announce Type: new Abstract: Medical image segmentation models often report high benchmark accuracy under ideal imaging conditions, yet their failures under clinical degradation can be quiet: sensor noise, patient motion, low- resolution acquisition, and contrast variability may

中文全文翻译正在补齐，可先查看下方英文原文。

[Submitted on 22 Jul 2026]

Title:Trustworthy Medical Segmentation: Uncertainty-Aware U-Net Evaluation Under Clinical Image Degradation

View a PDF of the paper titled Trustworthy Medical Segmentation: Uncertainty-Aware U-Net Evaluation Under Clinical Image Degradation, by Pranav Kaliaperumal and 1 other authors

View PDF HTML (experimental)

Abstract:Medical image segmentation models often report high benchmark accuracy under ideal imaging conditions, yet their failures under clinical degradation can be quiet: sensor noise, patient motion, low- resolution acquisition, and contrast variability may all alter model behavior without producing an obvious warning. We present a reproducible framework for evaluating uncertainty-aware segmentation under con- trolled clinical degradation. Our experiments use a synthetic multimodal brain tumor MRI cohort generated with a biophysical phantom simulator that follows the BraTS protocol. We train U-Net and Attention U-Net baselines for multi-class tumor sub-region segmentation and augment both models with Monte Carlo dropout to estimate per-voxel uncertainty. Across eight clinically motivated corruption types at five severity levels, we measure segmentation accuracy, calibration, failure detection, and selective prediction coverage. On clean data, Attention U-Net achieves a whole-tumor Dice of 0.990; under severe Gaussian noise, its performance falls to 0.089. Predictive uncertainty rises with degradation and tracks segmentation error (Pearson r = 0.53 under severity-3 Gaussian noise), allowing us to flag failures with an AUROC of 0.843. These results argue for uncertainty-aware inference as a practical safety layer in physician-in-the-loop radiology workflows. We release the code, trained models, and evaluation protocol to support direct reproduction.

Submission history

From: Pranav Kumar Kaliaperumal [

view email

]

[v1]

Wed, 22 Jul 2026 06:35:33 UTC (216 KB)