Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning

Invasive ductal carcinoma (IDC) is the most common form of breast cancer. For the non-operative diagnosis of breast carcinoma, core needle biopsy has been widely used in recent years for the evaluation of histopathological features, as it can provide a definitive diagnosis between IDC and benign les...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Fahdi Kanavati, Masayuki Tsuneki
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
Acceso en línea:https://doaj.org/article/bedf7aee21894d6381efdee2e9c6cae4
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:bedf7aee21894d6381efdee2e9c6cae4
record_format dspace
spelling oai:doaj.org-article:bedf7aee21894d6381efdee2e9c6cae42021-11-11T15:29:15ZBreast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning10.3390/cancers132153682072-6694https://doaj.org/article/bedf7aee21894d6381efdee2e9c6cae42021-10-01T00:00:00Zhttps://www.mdpi.com/2072-6694/13/21/5368https://doaj.org/toc/2072-6694Invasive ductal carcinoma (IDC) is the most common form of breast cancer. For the non-operative diagnosis of breast carcinoma, core needle biopsy has been widely used in recent years for the evaluation of histopathological features, as it can provide a definitive diagnosis between IDC and benign lesion (e.g., fibroadenoma), and it is cost effective. Due to its widespread use, it could potentially benefit from the use of AI-based tools to aid pathologists in their pathological diagnosis workflows. In this paper, we trained invasive ductal carcinoma (IDC) whole slide image (WSI) classification models using transfer learning and weakly-supervised learning. We evaluated the models on a core needle biopsy (<i>n</i> = 522) test set as well as three surgical test sets (<i>n</i> = 1129) obtaining ROC AUCs in the range of 0.95–0.98. The promising results demonstrate the potential of applying such models as diagnostic aid tools for pathologists in clinical practice.Fahdi KanavatiMasayuki TsunekiMDPI AGarticlebreastinvasive ductal carcinomadeep learningweakly-supervised learningtransfer learningwhole slide imageNeoplasms. Tumors. Oncology. Including cancer and carcinogensRC254-282ENCancers, Vol 13, Iss 5368, p 5368 (2021)
institution DOAJ
collection DOAJ
language EN
topic breast
invasive ductal carcinoma
deep learning
weakly-supervised learning
transfer learning
whole slide image
Neoplasms. Tumors. Oncology. Including cancer and carcinogens
RC254-282
spellingShingle breast
invasive ductal carcinoma
deep learning
weakly-supervised learning
transfer learning
whole slide image
Neoplasms. Tumors. Oncology. Including cancer and carcinogens
RC254-282
Fahdi Kanavati
Masayuki Tsuneki
Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning
description Invasive ductal carcinoma (IDC) is the most common form of breast cancer. For the non-operative diagnosis of breast carcinoma, core needle biopsy has been widely used in recent years for the evaluation of histopathological features, as it can provide a definitive diagnosis between IDC and benign lesion (e.g., fibroadenoma), and it is cost effective. Due to its widespread use, it could potentially benefit from the use of AI-based tools to aid pathologists in their pathological diagnosis workflows. In this paper, we trained invasive ductal carcinoma (IDC) whole slide image (WSI) classification models using transfer learning and weakly-supervised learning. We evaluated the models on a core needle biopsy (<i>n</i> = 522) test set as well as three surgical test sets (<i>n</i> = 1129) obtaining ROC AUCs in the range of 0.95–0.98. The promising results demonstrate the potential of applying such models as diagnostic aid tools for pathologists in clinical practice.
format article
author Fahdi Kanavati
Masayuki Tsuneki
author_facet Fahdi Kanavati
Masayuki Tsuneki
author_sort Fahdi Kanavati
title Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning
title_short Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning
title_full Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning
title_fullStr Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning
title_full_unstemmed Breast Invasive Ductal Carcinoma Classification on Whole Slide Images with Weakly-Supervised and Transfer Learning
title_sort breast invasive ductal carcinoma classification on whole slide images with weakly-supervised and transfer learning
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/bedf7aee21894d6381efdee2e9c6cae4
work_keys_str_mv AT fahdikanavati breastinvasiveductalcarcinomaclassificationonwholeslideimageswithweaklysupervisedandtransferlearning
AT masayukitsuneki breastinvasiveductalcarcinomaclassificationonwholeslideimageswithweaklysupervisedandtransferlearning
_version_ 1718435266468773888