Cargando…

Vision-Transformer-Based Transfer Learning for Mammogram Classification

Breast mass identification is a crucial procedure during mammogram-based early breast cancer diagnosis. However, it is difficult to determine whether a breast lump is benign or cancerous at early stages. Convolutional neural networks (CNNs) have been used to solve this problem and have provided usef...

Descripción completa

Detalles Bibliográficos
Autores principales: Ayana, Gelan, Dese, Kokeb, Dereje, Yisak, Kebede, Yonas, Barki, Hika, Amdissa, Dechassa, Husen, Nahimiya, Mulugeta, Fikadu, Habtamu, Bontu, Choe, Se-Woon
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9857963/
https://www.ncbi.nlm.nih.gov/pubmed/36672988
http://dx.doi.org/10.3390/diagnostics13020178
_version_ 1784873979700314112
author Ayana, Gelan
Dese, Kokeb
Dereje, Yisak
Kebede, Yonas
Barki, Hika
Amdissa, Dechassa
Husen, Nahimiya
Mulugeta, Fikadu
Habtamu, Bontu
Choe, Se-Woon
author_facet Ayana, Gelan
Dese, Kokeb
Dereje, Yisak
Kebede, Yonas
Barki, Hika
Amdissa, Dechassa
Husen, Nahimiya
Mulugeta, Fikadu
Habtamu, Bontu
Choe, Se-Woon
author_sort Ayana, Gelan
collection PubMed
description Breast mass identification is a crucial procedure during mammogram-based early breast cancer diagnosis. However, it is difficult to determine whether a breast lump is benign or cancerous at early stages. Convolutional neural networks (CNNs) have been used to solve this problem and have provided useful advancements. However, CNNs focus only on a certain portion of the mammogram while ignoring the remaining and present computational complexity because of multiple convolutions. Recently, vision transformers have been developed as a technique to overcome such limitations of CNNs, ensuring better or comparable performance in natural image classification. However, the utility of this technique has not been thoroughly investigated in the medical image domain. In this study, we developed a transfer learning technique based on vision transformers to classify breast mass mammograms. The area under the receiver operating curve of the new model was estimated as 1 ± 0, thus outperforming the CNN-based transfer-learning models and vision transformer models trained from scratch. The technique can, hence, be applied in a clinical setting, to improve the early diagnosis of breast cancer.
format Online
Article
Text
id pubmed-9857963
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-98579632023-01-21 Vision-Transformer-Based Transfer Learning for Mammogram Classification Ayana, Gelan Dese, Kokeb Dereje, Yisak Kebede, Yonas Barki, Hika Amdissa, Dechassa Husen, Nahimiya Mulugeta, Fikadu Habtamu, Bontu Choe, Se-Woon Diagnostics (Basel) Article Breast mass identification is a crucial procedure during mammogram-based early breast cancer diagnosis. However, it is difficult to determine whether a breast lump is benign or cancerous at early stages. Convolutional neural networks (CNNs) have been used to solve this problem and have provided useful advancements. However, CNNs focus only on a certain portion of the mammogram while ignoring the remaining and present computational complexity because of multiple convolutions. Recently, vision transformers have been developed as a technique to overcome such limitations of CNNs, ensuring better or comparable performance in natural image classification. However, the utility of this technique has not been thoroughly investigated in the medical image domain. In this study, we developed a transfer learning technique based on vision transformers to classify breast mass mammograms. The area under the receiver operating curve of the new model was estimated as 1 ± 0, thus outperforming the CNN-based transfer-learning models and vision transformer models trained from scratch. The technique can, hence, be applied in a clinical setting, to improve the early diagnosis of breast cancer. MDPI 2023-01-04 /pmc/articles/PMC9857963/ /pubmed/36672988 http://dx.doi.org/10.3390/diagnostics13020178 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Ayana, Gelan
Dese, Kokeb
Dereje, Yisak
Kebede, Yonas
Barki, Hika
Amdissa, Dechassa
Husen, Nahimiya
Mulugeta, Fikadu
Habtamu, Bontu
Choe, Se-Woon
Vision-Transformer-Based Transfer Learning for Mammogram Classification
title Vision-Transformer-Based Transfer Learning for Mammogram Classification
title_full Vision-Transformer-Based Transfer Learning for Mammogram Classification
title_fullStr Vision-Transformer-Based Transfer Learning for Mammogram Classification
title_full_unstemmed Vision-Transformer-Based Transfer Learning for Mammogram Classification
title_short Vision-Transformer-Based Transfer Learning for Mammogram Classification
title_sort vision-transformer-based transfer learning for mammogram classification
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9857963/
https://www.ncbi.nlm.nih.gov/pubmed/36672988
http://dx.doi.org/10.3390/diagnostics13020178
work_keys_str_mv AT ayanagelan visiontransformerbasedtransferlearningformammogramclassification
AT desekokeb visiontransformerbasedtransferlearningformammogramclassification
AT derejeyisak visiontransformerbasedtransferlearningformammogramclassification
AT kebedeyonas visiontransformerbasedtransferlearningformammogramclassification
AT barkihika visiontransformerbasedtransferlearningformammogramclassification
AT amdissadechassa visiontransformerbasedtransferlearningformammogramclassification
AT husennahimiya visiontransformerbasedtransferlearningformammogramclassification
AT mulugetafikadu visiontransformerbasedtransferlearningformammogramclassification
AT habtamubontu visiontransformerbasedtransferlearningformammogramclassification
AT choesewoon visiontransformerbasedtransferlearningformammogramclassification