Histopathological image analysis is a critical area of research with the potential to aid pathologists in faster and more accurate diagnoses. However, Whole-Slide Images (WSIs) present challenges for deep learning frameworks due to their large size and lack of pixel-level annotations. Multi-Instance Learning (MIL) is a popular approach that can be employed for handling WSIs, treating each slide as a bag composed of multiple patches or instances. In this work we propose Buffer-MIL, which aims at tackling the covariate shift and class imbalance characterizing most of the existing histopathological datasets. With this goal, a buffer containing the most representative instances of each disease-positive slide of the training set is incorporated into our model. An attention mechanism is then used to compare all the instances against the buffer, to find the most critical ones in a given slide. We evaluate Buffer-MIL on two publicly available WSI datasets, Camelyon16 and TCGA lung cancer, outperforming current state-of-the-art models by 2.2% of accuracy on Camelyon16.

Histopathological image analysis is a critical area of research with the potential to aid pathologists in faster and more accurate diagnoses. However, Whole-Slide Images (WSIs) present challenges for deep learning frameworks due to their large size and lack of pixel-level annotations. Multi-Instance Learning (MIL) is a popular approach that can be employed for handling WSIs, treating each slide as a bag composed of multiple patches or instances. In this work we propose Buffer-MIL, which aims at tackling the covariate shift and class imbalance characterizing most of the existing histopathological datasets. With this goal, a buffer containing the most representative instances of each disease-positive slide of the training set is incorporated into our model. An attention mechanism is then used to compare all the instances against the buffer, to find the most critical ones in a given slide. We evaluate Buffer-MIL on two publicly available WSI datasets, Camelyon16 and TCGA lung cancer, outperforming current state-of-the-art models by 2.2% of accuracy on Camelyon16.

Buffer-MIL: Robust Multi-instance Learning with a Buffer-Based Approach / Bontempo, G.; Lumetti, L.; Porrello, A.; Bolelli, F.; Calderara, S.; Ficarra, E.. - 14234:(2023), pp. 1-12. (Intervento presentato al convegno Image Analysis and Processing - ICIAP 2023 tenutosi a Udine, Italy nel Sep 11-15) [10.1007/978-3-031-43153-1_1].

Buffer-MIL: Robust Multi-instance Learning with a Buffer-Based Approach

Bontempo G.;Lumetti L.;Porrello A.;Bolelli F.;Calderara S.;Ficarra E.
2023

Abstract

Histopathological image analysis is a critical area of research with the potential to aid pathologists in faster and more accurate diagnoses. However, Whole-Slide Images (WSIs) present challenges for deep learning frameworks due to their large size and lack of pixel-level annotations. Multi-Instance Learning (MIL) is a popular approach that can be employed for handling WSIs, treating each slide as a bag composed of multiple patches or instances. In this work we propose Buffer-MIL, which aims at tackling the covariate shift and class imbalance characterizing most of the existing histopathological datasets. With this goal, a buffer containing the most representative instances of each disease-positive slide of the training set is incorporated into our model. An attention mechanism is then used to compare all the instances against the buffer, to find the most critical ones in a given slide. We evaluate Buffer-MIL on two publicly available WSI datasets, Camelyon16 and TCGA lung cancer, outperforming current state-of-the-art models by 2.2% of accuracy on Camelyon16.
2023
5-set-2023
Image Analysis and Processing - ICIAP 2023
Udine, Italy
Sep 11-15
14234
1
12
Bontempo, G.; Lumetti, L.; Porrello, A.; Bolelli, F.; Calderara, S.; Ficarra, E.
Buffer-MIL: Robust Multi-instance Learning with a Buffer-Based Approach / Bontempo, G.; Lumetti, L.; Porrello, A.; Bolelli, F.; Calderara, S.; Ficarra, E.. - 14234:(2023), pp. 1-12. (Intervento presentato al convegno Image Analysis and Processing - ICIAP 2023 tenutosi a Udine, Italy nel Sep 11-15) [10.1007/978-3-031-43153-1_1].
File in questo prodotto:
File Dimensione Formato  
2023ICIAP_Buffer_MIL__Robust_Multi_instance_Learning_witha_Buffer_based_Approach.pdf

Open access

Tipologia: Versione dell'autore revisionata e accettata per la pubblicazione
Dimensione 615.9 kB
Formato Adobe PDF
615.9 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1328829
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 1
social impact