Contrastive learning improves CT grading of kidney cancer tumors
Cross-Modal Contrastive Learning from Histopathology and CT for Automated Renal Cell Carcinoma Grading
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Kidney cancer tumors come in different grades that affect treatment decisions, but determining the grade usually needs an invasive tissue biopsy. The authors created a new computer method called RCC-Align that learns from both microscope images of tumor tissue and CT scans, so it can better guess tumor grades just from CT scans alone. Their method performed better than other CT-based methods on a cancer dataset, especially in identifying less aggressive tumors. This could help doctors when biopsies are risky or not possible, but the method needs more testing before use in clinics.
What this means in practice
- •For radiology teams: Improve noninvasive grading of kidney tumors using CT scans informed by tissue image data during training, enabling risk assessment without biopsy.
- •For medical image software developers: Integrate cross-modal contrastive learning models that align pathology and CT data for more accurate automated tumor grading tools.
Authors
Amit Das, Tanmay Shukla, Naofumi Tomita, Faraz Farhadi, Jessica Sin, Ari Hakimi, Chad Vanderbilt, Jie-Fu Chen, Ritesh Kotecha, Weijie Ma, Bing Ren, Saeed Hassanpour
Abstract
Background: Clear cell renal cell carcinoma (ccRCC) exhibits substantial clinical heterogeneity, and accurate grade assessment is essential for risk stratification and treatment planning. However, conventional grading requires invasive tissue sampling. We developed RCC-Align, a cross-modal contrastive learning framework that leverages paired histopathology and computed tomography (CT) data during training to improve noninvasive CT-based ccRCC grade prediction. Methods: RCC-Align aligns paired whole-slide histopathology images (WSIs) and CT scans through contrastive cross-modal objectives, transferring grade-discriminative information from microscopic tissue morphology to macroscopic radiologic representations. The framework was trained and evaluated on paired TCGA and CPTAC cohorts using patient-level five-fold cross-validation. Performance for low- versus high-grade ccRCC classification was compared against CT-only baselines (DINOv2-Base and DINOv2-Finetuned) and a WSI-based reference model (GigaPath-Finetuned). Cross-modal alignment was assessed using cosine similarity analysis. Results: RCC-Align achieved an AUC of 0.601 (95% CI, 0.524-0.673) and AUPRC of 0.599 (95% CI, 0.541-0.676), outperforming DINOv2-Finetuned (AUC 0.545; AUPRC 0.543) with significantly improved low-grade prediction (p = 0.004). RCC-Align also demonstrated stronger paired WSI-CT embedding alignment compared with baselines. The WSI-based GigaPath reference achieved an AUC of 0.719. Conclusion: Pathology-guided contrastive learning improves CT-based ccRCC grading while requiring only CT at inference. This approach may complement tissue diagnosis when biopsy is unsafe, infeasible, or limited by intratumoral heterogeneity. Validation in larger, multi-institutional cohorts with external testing is needed before clinical translation.