Finding Regions of Interest in Whole Slide Images Using Multiple Instance Learning.

Autor:	Afonso M; Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, Lisbon, 1049-001, Portugal., Bhawsar PMS; Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, 20850, Maryland, USA., Saha M; Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, 20850, Maryland, USA., Almeida JS; Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, 20850, Maryland, USA., Oliveira AL; Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, Lisbon, 1049-001, Portugal.; INESC-ID, R. Alves Redol 9, Lisbon, 1000-029, Portugal.
Jazyk:	angličtina
Zdroj:	ArXiv [ArXiv] 2024 Apr 11. Date of Electronic Publication: 2024 Apr 11.
Abstrakt:	Whole Slide Images (WSI), obtained by high-resolution digital scanning of microscope slides at multiple scales, are the cornerstone of modern Digital Pathology. However, they represent a particular challenge to AI-based/AI-mediated analysis because pathology labeling is typically done at slide-level, instead of tile-level. It is not just that medical diagnostics is recorded at the specimen level, the detection of oncogene mutation is also experimentally obtained, and recorded by initiatives like The Cancer Genome Atlas (TCGA), at the slide level. This configures a dual challenge: a) accurately predicting the overall cancer phenotype and b) finding out what cellular morphologies are associated with it at the tile level. To address these challenges, a weakly supervised Multiple Instance Learning (MIL) approach was explored for two prevalent cancer types, Invasive Breast Carcinoma (TCGA-BRCA) and Lung Squamous Cell Carcinoma (TCGA-LUSC). This approach was explored for tumor detection at low magnification levels and TP53 mutations at various levels. Our results show that a novel additive implementation of MIL matched the performance of reference implementation (AUC 0.96), and was only slightly outperformed by Attention MIL (AUC 0.97). More interestingly from the perspective of the molecular pathologist, these different AI architectures identify distinct sensitivities to morphological features (through the detection of Regions of Interest, RoI) at different amplification levels. Tellingly, TP53 mutation was most sensitive to features at the higher applications where cellular morphology is resolved. Competing Interests: 7.Author Declaration There is no conflict of interest to report.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu