Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Shakarami, Ashkan, Nicole, Lorenzo, Cappellesso, Rocco, Tos, Angelo Paolo Dei, Ghidoni, Stefano
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2507.10250
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909688783699968
author Shakarami, Ashkan
Nicole, Lorenzo
Cappellesso, Rocco
Tos, Angelo Paolo Dei
Ghidoni, Stefano
author_facet Shakarami, Ashkan
Nicole, Lorenzo
Cappellesso, Rocco
Tos, Angelo Paolo Dei
Ghidoni, Stefano
contents Accurate and timely cancer diagnosis from histopathological slides is vital for effective clinical decision-making. This paper introduces DepViT-CAD, a deployable AI system for multi-class cancer diagnosis in histopathology. At its core is MAViT, a novel Multi-Attention Vision Transformer designed to capture fine-grained morphological patterns across diverse tumor types. MAViT was trained on expert-annotated patches from 1008 whole-slide images, covering 11 diagnostic categories, including 10 major cancers and non-tumor tissue. DepViT-CAD was validated on two independent cohorts: 275 WSIs from The Cancer Genome Atlas and 50 routine clinical cases from pathology labs, achieving diagnostic sensitivities of 94.11% and 92%, respectively. By combining state-of-the-art transformer architecture with large-scale real-world validation, DepViT-CAD offers a robust and scalable approach for AI-assisted cancer diagnostics. To support transparency and reproducibility, software and code will be made publicly available at GitHub.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10250
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DepViT-CAD: Deployable Vision Transformer-Based Cancer Diagnosis in Histopathology
Shakarami, Ashkan
Nicole, Lorenzo
Cappellesso, Rocco
Tos, Angelo Paolo Dei
Ghidoni, Stefano
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Accurate and timely cancer diagnosis from histopathological slides is vital for effective clinical decision-making. This paper introduces DepViT-CAD, a deployable AI system for multi-class cancer diagnosis in histopathology. At its core is MAViT, a novel Multi-Attention Vision Transformer designed to capture fine-grained morphological patterns across diverse tumor types. MAViT was trained on expert-annotated patches from 1008 whole-slide images, covering 11 diagnostic categories, including 10 major cancers and non-tumor tissue. DepViT-CAD was validated on two independent cohorts: 275 WSIs from The Cancer Genome Atlas and 50 routine clinical cases from pathology labs, achieving diagnostic sensitivities of 94.11% and 92%, respectively. By combining state-of-the-art transformer architecture with large-scale real-world validation, DepViT-CAD offers a robust and scalable approach for AI-assisted cancer diagnostics. To support transparency and reproducibility, software and code will be made publicly available at GitHub.
title DepViT-CAD: Deployable Vision Transformer-Based Cancer Diagnosis in Histopathology
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2507.10250