Connected-UNets: a deep learning architecture for breast mass segmentation

Abstract Breast cancer analysis implies that radiologists inspect mammograms to detect suspicious breast lesions and identify mass tumors. Artificial intelligence techniques offer automatic systems for breast mass segmentation to assist radiologists in their diagnosis. With the rapid development of...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Asma Baccouche, Begonya Garcia-Zapirain, Cristian Castillo Olea, Adel S. Elmaghraby
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
Acceso en línea:https://doaj.org/article/dccc5b2e5f29459eabd921a1aed54115
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:dccc5b2e5f29459eabd921a1aed54115
record_format dspace
spelling oai:doaj.org-article:dccc5b2e5f29459eabd921a1aed541152021-12-05T12:20:20ZConnected-UNets: a deep learning architecture for breast mass segmentation10.1038/s41523-021-00358-x2374-4677https://doaj.org/article/dccc5b2e5f29459eabd921a1aed541152021-12-01T00:00:00Zhttps://doi.org/10.1038/s41523-021-00358-xhttps://doaj.org/toc/2374-4677Abstract Breast cancer analysis implies that radiologists inspect mammograms to detect suspicious breast lesions and identify mass tumors. Artificial intelligence techniques offer automatic systems for breast mass segmentation to assist radiologists in their diagnosis. With the rapid development of deep learning and its application to medical imaging challenges, UNet and its variations is one of the state-of-the-art models for medical image segmentation that showed promising performance on mammography. In this paper, we propose an architecture, called Connected-UNets, which connects two UNets using additional modified skip connections. We integrate Atrous Spatial Pyramid Pooling (ASPP) in the two standard UNets to emphasize the contextual information within the encoder–decoder network architecture. We also apply the proposed architecture on the Attention UNet (AUNet) and the Residual UNet (ResUNet). We evaluated the proposed architectures on two publically available datasets, the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) and INbreast, and additionally on a private dataset. Experiments were also conducted using additional synthetic data using the cycle-consistent Generative Adversarial Network (CycleGAN) model between two unpaired datasets to augment and enhance the images. Qualitative and quantitative results show that the proposed architecture can achieve better automatic mass segmentation with a high Dice score of 89.52%, 95.28%, and 95.88% and Intersection over Union (IoU) score of 80.02%, 91.03%, and 92.27%, respectively, on CBIS-DDSM, INbreast, and the private dataset.Asma BaccoucheBegonya Garcia-ZapirainCristian Castillo OleaAdel S. ElmaghrabyNature PortfolioarticleNeoplasms. Tumors. Oncology. Including cancer and carcinogensRC254-282ENnpj Breast Cancer, Vol 7, Iss 1, Pp 1-12 (2021)
institution DOAJ
collection DOAJ
language EN
topic Neoplasms. Tumors. Oncology. Including cancer and carcinogens
RC254-282
spellingShingle Neoplasms. Tumors. Oncology. Including cancer and carcinogens
RC254-282
Asma Baccouche
Begonya Garcia-Zapirain
Cristian Castillo Olea
Adel S. Elmaghraby
Connected-UNets: a deep learning architecture for breast mass segmentation
description Abstract Breast cancer analysis implies that radiologists inspect mammograms to detect suspicious breast lesions and identify mass tumors. Artificial intelligence techniques offer automatic systems for breast mass segmentation to assist radiologists in their diagnosis. With the rapid development of deep learning and its application to medical imaging challenges, UNet and its variations is one of the state-of-the-art models for medical image segmentation that showed promising performance on mammography. In this paper, we propose an architecture, called Connected-UNets, which connects two UNets using additional modified skip connections. We integrate Atrous Spatial Pyramid Pooling (ASPP) in the two standard UNets to emphasize the contextual information within the encoder–decoder network architecture. We also apply the proposed architecture on the Attention UNet (AUNet) and the Residual UNet (ResUNet). We evaluated the proposed architectures on two publically available datasets, the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) and INbreast, and additionally on a private dataset. Experiments were also conducted using additional synthetic data using the cycle-consistent Generative Adversarial Network (CycleGAN) model between two unpaired datasets to augment and enhance the images. Qualitative and quantitative results show that the proposed architecture can achieve better automatic mass segmentation with a high Dice score of 89.52%, 95.28%, and 95.88% and Intersection over Union (IoU) score of 80.02%, 91.03%, and 92.27%, respectively, on CBIS-DDSM, INbreast, and the private dataset.
format article
author Asma Baccouche
Begonya Garcia-Zapirain
Cristian Castillo Olea
Adel S. Elmaghraby
author_facet Asma Baccouche
Begonya Garcia-Zapirain
Cristian Castillo Olea
Adel S. Elmaghraby
author_sort Asma Baccouche
title Connected-UNets: a deep learning architecture for breast mass segmentation
title_short Connected-UNets: a deep learning architecture for breast mass segmentation
title_full Connected-UNets: a deep learning architecture for breast mass segmentation
title_fullStr Connected-UNets: a deep learning architecture for breast mass segmentation
title_full_unstemmed Connected-UNets: a deep learning architecture for breast mass segmentation
title_sort connected-unets: a deep learning architecture for breast mass segmentation
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/dccc5b2e5f29459eabd921a1aed54115
work_keys_str_mv AT asmabaccouche connectedunetsadeeplearningarchitectureforbreastmasssegmentation
AT begonyagarciazapirain connectedunetsadeeplearningarchitectureforbreastmasssegmentation
AT cristiancastilloolea connectedunetsadeeplearningarchitectureforbreastmasssegmentation
AT adelselmaghraby connectedunetsadeeplearningarchitectureforbreastmasssegmentation
_version_ 1718372045212876800