发表机构
College of Engineering and Computer Science, VinUniversity（VinUniversity工程与计算机科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对手术分割网络在采集退化下的高置信度静默失效问题，提出TCSR-Monitor框架，结合多维度可观测线索监测失效，在EndoVis 2017上泛化能力优于置信度基线，但存在误报及迁移局限。
AI 中文摘要
手术分割网络在采集退化下可能发生静默失效：即便模型置信度仍保持较高水平，预测的掩码也可能出错。现有的部署时监测器主要依赖不确定性估计，因此可能遗漏高置信度失效。我们提出TCSR-Monitor（时序保形手术风险监测器，Temporal Conformal Surgical Risk Monitor），这是一种事后失效监测框架，将置信度与可观测的形状、时序一致性及图像质量线索相结合。TCSR-Monitor包裹一个冻结的分割模型，无需模型内部信息，且在部署时无需真值。我们还引入一种验证协议，用于评估在分布偏移下警报是否仍保持可信。在EndoVis 2017数据集上，留一损坏评估显示，TCSR-Monitor可泛化至未见过的采集退化，且显著优于基于置信度的基线。循环控制证实它预测的是分割失效，而非仅检测损坏的图像。Mondrian保形校准平衡了不同退化严重程度下的漏检率，但单一全局阈值在中等退化时仍会在多达40%的正确分割帧上产生误报。零样本迁移至SAM2展示了特征可移植性，不过在两个评估阈值下，熵均优于迁移后的监测器。总体而言，采集退化下的可靠监测需依赖置信度之外的互补可观测信号，但仍存在显著的误报和迁移局限性。
英文摘要
Surgical segmentation networks can fail silently under acquisition degradation: predicted masks may be wrong even when model confidence remains high. Existing deployment-time monitors rely primarily on uncertainty estimates and can therefore miss confident failures. We present TCSR-Monitor (Temporal Conformal Surgical Risk Monitor), a post-hoc failure-monitoring framework that combines confidence with observable shape, temporal-consistency, and image-quality cues. TCSR-Monitor wraps a frozen segmentation model, requires no model internals, and operates without ground truth at deployment. We also introduce a validation protocol to assess whether alarms remain credible under distribution shift. On EndoVis 2017, leave-one-corruption-out evaluation shows that TCSR-Monitor generalizes to unseen acquisition degradations and substantially outperforms confidence-based baselines. A circularity control confirms that it predicts segmentation failure rather than simply detecting corrupted images. Mondrian conformal calibration balances miss-rates across degradation severities, but a single global threshold still produces false alarms on up to 40% of correctly segmented frames at moderate corruption. Zero-shot transfer to SAM2 demonstrates feature portability, although entropy outperforms the transferred monitor at both evaluated thresholds. Overall, reliable monitoring under acquisition degradation benefits from complementary observable signals beyond confidence alone, but substantial false-alarm and transfer limitations remain.
CommentsAccepted at MICCAI'2026 @UNSURE Workshop
