arXiv：cs.AI· Hikmat Khan, Metin N. Gurcan, Anil Parwani, Wendy L. Frankel, Wei Chen, Muhammad Khalid Khan Niazi·· 6 小时前
MorphContrast：形态感知深度学习从全切片图像预测结直肠癌生存
Morphology-aware deep learning predicts colorectal cancer survival from whole-slide images
AI 导读
研究提出形态感知对比多实例学习框架 MorphContrast，以组织形态标签作为结构监督，从 H&E 全切片图像预测结直肠癌五年生存。在 Alliance/CALGB 89803 上单模型 AUC 为 0.642，事后集成达 0.70，风险组 hazard ratio 3.79；外部 SurGen 队列（n=218）单模型与集成 AUC 分别为 0.669 和 0.678。
正文
Abstract:Background: Five-year survival prediction from hematoxylin-and-eosin whole-slide images could support risk stratification, but the task is weakly supervised, and the tissue morphologies carrying prognostic information remain incompletely characterized.
Methods: We conducted a retrospective secondary analysis using TCGA-COAD/READ (n=562) for morphology-specific analysis, Alliance/ CALGB 89803 (n=424) for model development, and SurGen (n=218) for external evaluation. We propose MorphContrast, a morphology-aware contrastive multiple-instance learning framework that uses tissue labels as structural supervision for five-year survival prediction. MorphContrast combines gated-attention pooling with three auxiliary objectives: patch-level morphology labels define within-patient contrasts, outcome concordance or discordance defines cross-patient contrasts within matched morphologies, and diversity regularization discourages representational collapse. Morphology-matched patches from patients with discordant outcomes serve as hard negatives, encouraging prognostic discrimination within comparable histologic contexts.
Results: Morphology-specific models show weak but distributed prognostic discrimination: adipose tissue achieves an area under the receiver operating characteristic curve (AUC) of 0.622, stroma achieves 0.604, and both perform comparably to tumor epithelium (0.601-0.605). On Alliance/CALGB 89803, the single model achieves an AUC of 0.642, while the post hoc ensemble reaches a pooled out-of-fold AUC of 0.70 and a risk-group hazard ratio of 3.79 (95% confidence interval: 2.55-5.63). On the external SurGen colon cancer cohort (n=218), the single model and ensemble achieve AUCs of 0.669 and 0.678, respectively.
Conclusions: These findings support tissue morphology as structural supervision for outcome-aware representation learning in colorectal cancer prognosis.
| Comments: | Supplementary material included: one figure and two tables |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.17603 [cs.CV] |
| (or arXiv:2610.17603v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.17603 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hikmat Khan Ph.D [view email]
[v1]
Tue, 13 Oct 2026 02:22:42 UTC (4,643 KB)
来源：arXiv：cs.AI · arxiv.org