CANDOR：基于冻结编码器的机会校准邻域不一致性医学影像评估方法

View PDF HTML (experimental)

Abstract:In medical imaging, one frozen foundation encoder is often reused for many tasks. The encoder is usually chosen by the area under the receiver operating characteristic curve (AUROC) of a head trained on its features. The AUROC does not measure whether the encoder separates positive from negative images in its feature space. Chance-calibrated neighborhood discordance (CANDOR) measures this separation without training a head. Among images acquired the same way, a positive image is discordant when its k nearest negatives are more similar to it on average than its k nearest positives are. The discordance rate D is the share of discordant positives. When the neighbors come from positive and negative candidate sets of equal size, D has the chance level 1/2 for features without information about the label. With unequal candidate sets, all 11 chest encoders have rates above 1/2 (0.732 to 0.927) on the 12 chest findings of MIMIC-CXR and CheXpert. These rates increase as the finding becomes rarer. With equal sets, the same encoders have rates below 1/2 (0.277 to 0.463). On the chest test split, linear heads miss 0.301 of the discordant positives and 0.165 of the other positives. Across 22 encoders, 605,443 images, and 8 tasks in 7 domains, the best encoder of every task has a rate below 1/2. Because computing D takes one forward pass per image, a hospital can compare candidate encoders before it trains any head.

Subjects:	Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
Cite as:	arXiv:2607.18451 [cs.LG]
	(or arXiv:2607.18451v4 [cs.LG] for this version)
	https://doi.org/10.48550/arXiv.2607.18451 arXiv-issued DOI via DataCite

Submission history

From: Soroosh Tayebi Arasteh [view email]
[v1] Mon, 20 Jul 2026 19:01:12 UTC (134 KB)
[v2] Mon, 24 Aug 2026 14:48:56 UTC (134 KB)
[v3] Sun, 20 Sep 2026 14:17:42 UTC (155 KB)
[v4] Sun, 25 Oct 2026 16:42:49 UTC (76 KB)