Cargando…

BUViTNet: Breast Ultrasound Detection via Vision Transformers

Convolutional neural networks (CNNs) have enhanced ultrasound image-based early breast cancer detection. Vision transformers (ViTs) have recently surpassed CNNs as the most effective method for natural image analysis. ViTs have proven their capability of incorporating more global information than CN...

Descripción completa

Detalles Bibliográficos
Autores principales: Ayana, Gelan, Choe, Se-woon
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9689470/
https://www.ncbi.nlm.nih.gov/pubmed/36359497
http://dx.doi.org/10.3390/diagnostics12112654
_version_ 1784836543050940416
author Ayana, Gelan
Choe, Se-woon
author_facet Ayana, Gelan
Choe, Se-woon
author_sort Ayana, Gelan
collection PubMed
description Convolutional neural networks (CNNs) have enhanced ultrasound image-based early breast cancer detection. Vision transformers (ViTs) have recently surpassed CNNs as the most effective method for natural image analysis. ViTs have proven their capability of incorporating more global information than CNNs at lower layers, and their skip connections are more powerful than those of CNNs, which endows ViTs with superior performance. However, the effectiveness of ViTs in breast ultrasound imaging has not yet been investigated. Here, we present BUViTNet breast ultrasound detection via ViTs, where ViT-based multistage transfer learning is performed using ImageNet and cancer cell image datasets prior to transfer learning for classifying breast ultrasound images. We utilized two publicly available ultrasound breast image datasets, Mendeley and breast ultrasound images (BUSI), to train and evaluate our algorithm. The proposed method achieved the highest area under the receiver operating characteristics curve (AUC) of 1 ± 0, Matthew’s correlation coefficient (MCC) of 1 ± 0, and kappa score of 1 ± 0 on the Mendeley dataset. Furthermore, BUViTNet achieved the highest AUC of 0.968 ± 0.02, MCC of 0.961 ± 0.01, and kappa score of 0.959 ± 0.02 on the BUSI dataset. BUViTNet outperformed ViT trained from scratch, ViT-based conventional transfer learning, and CNN-based transfer learning in classifying breast ultrasound images (p < 0.01 in all cases). Our findings indicate that improved transformers are effective in analyzing breast images and can provide an improved diagnosis if used in clinical settings. Future work will consider the use of a wide range of datasets and parameters for optimized performance.
format Online
Article
Text
id pubmed-9689470
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-96894702022-11-25 BUViTNet: Breast Ultrasound Detection via Vision Transformers Ayana, Gelan Choe, Se-woon Diagnostics (Basel) Article Convolutional neural networks (CNNs) have enhanced ultrasound image-based early breast cancer detection. Vision transformers (ViTs) have recently surpassed CNNs as the most effective method for natural image analysis. ViTs have proven their capability of incorporating more global information than CNNs at lower layers, and their skip connections are more powerful than those of CNNs, which endows ViTs with superior performance. However, the effectiveness of ViTs in breast ultrasound imaging has not yet been investigated. Here, we present BUViTNet breast ultrasound detection via ViTs, where ViT-based multistage transfer learning is performed using ImageNet and cancer cell image datasets prior to transfer learning for classifying breast ultrasound images. We utilized two publicly available ultrasound breast image datasets, Mendeley and breast ultrasound images (BUSI), to train and evaluate our algorithm. The proposed method achieved the highest area under the receiver operating characteristics curve (AUC) of 1 ± 0, Matthew’s correlation coefficient (MCC) of 1 ± 0, and kappa score of 1 ± 0 on the Mendeley dataset. Furthermore, BUViTNet achieved the highest AUC of 0.968 ± 0.02, MCC of 0.961 ± 0.01, and kappa score of 0.959 ± 0.02 on the BUSI dataset. BUViTNet outperformed ViT trained from scratch, ViT-based conventional transfer learning, and CNN-based transfer learning in classifying breast ultrasound images (p < 0.01 in all cases). Our findings indicate that improved transformers are effective in analyzing breast images and can provide an improved diagnosis if used in clinical settings. Future work will consider the use of a wide range of datasets and parameters for optimized performance. MDPI 2022-11-01 /pmc/articles/PMC9689470/ /pubmed/36359497 http://dx.doi.org/10.3390/diagnostics12112654 Text en © 2022 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
Choe, Se-woon
BUViTNet: Breast Ultrasound Detection via Vision Transformers
title BUViTNet: Breast Ultrasound Detection via Vision Transformers
title_full BUViTNet: Breast Ultrasound Detection via Vision Transformers
title_fullStr BUViTNet: Breast Ultrasound Detection via Vision Transformers
title_full_unstemmed BUViTNet: Breast Ultrasound Detection via Vision Transformers
title_short BUViTNet: Breast Ultrasound Detection via Vision Transformers
title_sort buvitnet: breast ultrasound detection via vision transformers
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9689470/
https://www.ncbi.nlm.nih.gov/pubmed/36359497
http://dx.doi.org/10.3390/diagnostics12112654
work_keys_str_mv AT ayanagelan buvitnetbreastultrasounddetectionviavisiontransformers
AT choesewoon buvitnetbreastultrasounddetectionviavisiontransformers