A Lightweight Multi-Cancer Tumor Localization Framework for Deployable Digital Pathology

TL;DR

This paper presents MuCTaL, a scalable, multi-cancer tumor localization framework using deep learning that generalizes across different tumor types and provides spatial tumor probability heatmaps for digital pathology.

Contribution

The study introduces MuCTaL, a novel multi-cancer tumor localization model trained on diverse cancer types, demonstrating high accuracy and cross-cancer generalization with an accessible inference workflow.

Findings

Achieved 0.97 ROC-AUC on training cancers

Achieved 0.71 ROC-AUC on unseen pancreatic cancer

Provided an open-source scalable digital pathology tool

Abstract

Accurate localization of tumor regions from hematoxylin and eosin-stained whole-slide images is fundamental for translational research including spatial analysis, molecular profiling, and tissue architecture investigation. However, deep learning-based tumor detection trained within specific cancers may exhibit reduced robustness when applied across different tumor types. We investigated whether balanced training across cancers at modest scale can achieve high performance and generalize to unseen tumor types. A multi-cancer tumor localization model (MuCTaL) was trained on 79,984 non-overlapping tiles from four cancers (melanoma, hepatocellular carcinoma, colorectal cancer, and non-small cell lung cancer) using transfer learning with DenseNet169. The model achieved a tile-level ROC-AUC of 0.97 in validation data from the four training cancers, and 0.71 on an independent pancreatic ductal adenocarcinoma cohort. A scalable inference workflow was built to generate spatial tumor probability heatmaps compatible with existing digital pathology tools. Code and models are publicly available at https://github.com/AivaraX-AI/MuCTaL.