Guardado en:
Detalles Bibliográficos
Autores principales: Bajger, Adam, Obdržálek, Jan, Kůr, Vojtěch, Nenutil, Rudolf, Holub, Petr, Musil, Vít, Brázdil, Tomáš
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2511.14558
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913167730278400
author Bajger, Adam
Obdržálek, Jan
Kůr, Vojtěch
Nenutil, Rudolf
Holub, Petr
Musil, Vít
Brázdil, Tomáš
author_facet Bajger, Adam
Obdržálek, Jan
Kůr, Vojtěch
Nenutil, Rudolf
Holub, Petr
Musil, Vít
Brázdil, Tomáš
contents We present a clustering-based explainability technique for digital pathology models based on convolutional neural networks. Unlike commonly used methods based on saliency maps, such as occlusion, GradCAM, or relevance propagation, which highlight regions that contribute the most to the prediction for a single slide, our method shows the global behaviour of the model under consideration, while also providing more fine-grained information. The result clusters can be visualised not only to understand the model, but also to increase confidence in its operation, leading to faster adoption in clinical practice. We also evaluate the performance of our technique on an existing model for detecting prostate cancer, demonstrating its usefulness.
format Preprint
id arxiv_https___arxiv_org_abs_2511_14558
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Explaining Digital Pathology Models via Clustering Activations
Bajger, Adam
Obdržálek, Jan
Kůr, Vojtěch
Nenutil, Rudolf
Holub, Petr
Musil, Vít
Brázdil, Tomáš
Computer Vision and Pattern Recognition
We present a clustering-based explainability technique for digital pathology models based on convolutional neural networks. Unlike commonly used methods based on saliency maps, such as occlusion, GradCAM, or relevance propagation, which highlight regions that contribute the most to the prediction for a single slide, our method shows the global behaviour of the model under consideration, while also providing more fine-grained information. The result clusters can be visualised not only to understand the model, but also to increase confidence in its operation, leading to faster adoption in clinical practice. We also evaluate the performance of our technique on an existing model for detecting prostate cancer, demonstrating its usefulness.
title Explaining Digital Pathology Models via Clustering Activations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.14558