Salvato in:
Dettagli Bibliografici
Autori principali: Mirabadi, Ali Khajegili, Archibald, Graham, Darbandsari, Amirali, Contreras-Sanz, Alberto, Nakhli, Ramin Ebrahim, Asadi, Maryam, Zhang, Allen, Gilks, C. Blake, Black, Peter, Wang, Gang, Farahani, Hossein, Bashashati, Ali
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2402.03592
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912333226311680
author Mirabadi, Ali Khajegili
Archibald, Graham
Darbandsari, Amirali
Contreras-Sanz, Alberto
Nakhli, Ramin Ebrahim
Asadi, Maryam
Zhang, Allen
Gilks, C. Blake
Black, Peter
Wang, Gang
Farahani, Hossein
Bashashati, Ali
author_facet Mirabadi, Ali Khajegili
Archibald, Graham
Darbandsari, Amirali
Contreras-Sanz, Alberto
Nakhli, Ramin Ebrahim
Asadi, Maryam
Zhang, Allen
Gilks, C. Blake
Black, Peter
Wang, Gang
Farahani, Hossein
Bashashati, Ali
contents Cancer subtyping is one of the most challenging tasks in digital pathology, where Multiple Instance Learning (MIL) by processing gigapixel whole slide images (WSIs) has been in the spotlight of recent research. However, MIL approaches do not take advantage of inter- and intra-magnification information contained in WSIs. In this work, we present GRASP, a novel lightweight graph-structured multi-magnification framework for processing WSIs in digital pathology. Our approach is designed to dynamically emulate the pathologist's behavior in handling WSIs and benefits from the hierarchical structure of WSIs. GRASP, which introduces a convergence-based node aggregation mechanism replacing traditional pooling mechanisms, outperforms state-of-the-art methods by a high margin in terms of balanced accuracy, while being significantly smaller than the closest-performing state-of-the-art models in terms of the number of parameters. Our results show that GRASP is dynamic in finding and consulting with different magnifications for subtyping cancers, is reliable and stable across different hyperparameters, and can generalize when using features from different backbones. The model's behavior has been evaluated by two expert pathologists confirming the interpretability of the model's dynamic. We also provide a theoretical foundation, along with empirical evidence, for our work, explaining how GRASP interacts with different magnifications and nodes in the graph to make predictions. We believe that the strong characteristics yet simple structure of GRASP will encourage the development of interpretable, structure-based designs for WSI representation in digital pathology. Data and code can be found in https://github.com/AIMLab-UBC/GRASP
format Preprint
id arxiv_https___arxiv_org_abs_2402_03592
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GRASP: GRAph-Structured Pyramidal Whole Slide Image Representation
Mirabadi, Ali Khajegili
Archibald, Graham
Darbandsari, Amirali
Contreras-Sanz, Alberto
Nakhli, Ramin Ebrahim
Asadi, Maryam
Zhang, Allen
Gilks, C. Blake
Black, Peter
Wang, Gang
Farahani, Hossein
Bashashati, Ali
Computer Vision and Pattern Recognition
Cancer subtyping is one of the most challenging tasks in digital pathology, where Multiple Instance Learning (MIL) by processing gigapixel whole slide images (WSIs) has been in the spotlight of recent research. However, MIL approaches do not take advantage of inter- and intra-magnification information contained in WSIs. In this work, we present GRASP, a novel lightweight graph-structured multi-magnification framework for processing WSIs in digital pathology. Our approach is designed to dynamically emulate the pathologist's behavior in handling WSIs and benefits from the hierarchical structure of WSIs. GRASP, which introduces a convergence-based node aggregation mechanism replacing traditional pooling mechanisms, outperforms state-of-the-art methods by a high margin in terms of balanced accuracy, while being significantly smaller than the closest-performing state-of-the-art models in terms of the number of parameters. Our results show that GRASP is dynamic in finding and consulting with different magnifications for subtyping cancers, is reliable and stable across different hyperparameters, and can generalize when using features from different backbones. The model's behavior has been evaluated by two expert pathologists confirming the interpretability of the model's dynamic. We also provide a theoretical foundation, along with empirical evidence, for our work, explaining how GRASP interacts with different magnifications and nodes in the graph to make predictions. We believe that the strong characteristics yet simple structure of GRASP will encourage the development of interpretable, structure-based designs for WSI representation in digital pathology. Data and code can be found in https://github.com/AIMLab-UBC/GRASP
title GRASP: GRAph-Structured Pyramidal Whole Slide Image Representation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.03592