arXiv cs.AI· Sol Lee, Hyunji Kim, Sungrae Hong, Donghee Han, Mun Yi·· 9 天前AI 评分18
面向可信脑肿瘤分割的缺失模态感知校准方法 MMA-LTS
Missing Modality-Aware Calibration for Trustworthy Brain Tumor Segmentation
AI 导读
针对多模态脑肿瘤分割中模态缺失导致置信度校准失准的问题，研究者提出缺失模态感知局部温度缩放（MMA-LTS），一种基于体素的事后置信度校准方法，通过模态可用性可学习 token 与体素级难度分数估计空间自适应温度场。在 BraTS 2020 和 FeTS 2024 上，MMA-LTS 在多种模态缺失场景下改善了校准效果，同时保持 SOTA 模型的分割精度。
正文 · 原文
Abstract:Multimodal brain tumor segmentation typically leverages multiple MRI modalities, yet incomplete modality acquisition is common in clinical practice due to protocol heterogeneity and scan failures. Although recent methods maintain segmentation accuracy under missing modality conditions, they frequently overlook prediction reliability, leading to miscalibrated confidence estimates that hinder clinical adoption. Existing calibration techniques are largely modality-agnostic or assume that prediction difficulty decreases monotonically as additional modalities become available. However, in brain tumor segmentation, prediction difficulty depends primarily on which modalities are absent rather than how many, leading to combination-specific and spatially heterogeneous calibration errors. To address this, we propose Missing Modality-Aware Local Temperature Scaling (MMA-LTS), a post-hoc voxel-wise confidence calibration method. It estimates a spatially adaptive temperature field conditioned on a modality-availability learnable token and a voxel-wise difficulty score. Experiments on BraTS 2020 and FeTS 2024 show that MMA-LTS improves calibration while preserving the segmentation accuracy of state-of-the-art models across diverse missing-modality scenarios, thereby enhancing trustworthiness toward clinical deployment.
| Comments: | MICCAI2026 poster |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11419 [cs.CV] |
| (or arXiv:2610.11419v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11419 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sol Lee [view email]
[v1]
Thu, 8 Oct 2026 07:46:21 UTC (4,087 KB)
来源：arXiv cs.AI · arxiv.org