Autor según el artículo: Singh, Vivek Kumar; Romani, Santiago; Rashwan, Hatem A; Akram, Farhan; Pandey, Nidhi; Kamal Sarker, Md Mostafa; Abdulwahab, Saddam; Torrents-Barrena, Jordina; Saleh, Adel; Arquez, Miguel; Arenas, Meritxell; Puig, Domenec
Departamento: Ciències Mèdiques Bàsiques Enginyeria Informàtica i Matemàtiques
Autor/es de la URV: Abdellatif Fatahallah Ibrahim Mahmoud, Hatem / Abdulwahab, Saddam Abdulrhman Hamed / AKRAM, FARHAN / Arenas Prat, Meritxell / Puig Valls, Domènec Savi / Romaní Also, Santiago
Palabras clave: Mass shape classification Mass segmentation Mammography Cnn Cgan Cancer molecular subtypes
Resumen: © Springer Nature Switzerland AG 2018. This paper proposes a novel approach based on conditional Generative Adversarial Networks (cGAN) for breast mass segmentation in mammography. We hypothesized that the cGAN structure is well-suited to accurately outline the mass area, especially when the training data is limited. The generative network learns intrinsic features of tumors while the adversarial network enforces segmentations to be similar to the ground truth. Experiments performed on dozens of malignant tumors extracted from the public DDSM dataset and from our in-house private dataset confirm our hypothesis with very high Dice coefficient and Jaccard index (>94% and >89%, respectively) outperforming the scores obtained by other state-of-the-art approaches. Furthermore, in order to detect portray significant morphological features of the segmented tumor, a specific Convolutional Neural Network (CNN) have also been designed for classifying the segmented tumor areas into four types (irregular, lobular, oval and round), which provides an overall accuracy about 72% with the DDSM dataset.
Áreas temáticas: Linguística e literatura Interdisciplinar Geociências Engenharias iv Educação Ciências ambientais Ciência da computação Arquitetura, urbanismo e design Administração pública e de empresas, ciências contábeis e turismo
Acceso a la licencia de uso: https://creativecommons.org/licenses/by/3.0/es/
ISSN: 16113349
Direcció de correo del autor: saddam.abdulwahab@urv.cat hatem.abdellatif@urv.cat saddam.abdulwahab@urv.cat meritxell.arenas@urv.cat santiago.romani@urv.cat domenec.puig@urv.cat
Identificador del autor: 0000-0001-5421-1637 0000-0003-0815-2570 0000-0001-6673-9615 0000-0002-0562-4205
Fecha de alta del registro: 2024-10-12
Versión del articulo depositado: info:eu-repo/semantics/submittedVersion
Enlace a la fuente original: https://link.springer.com/chapter/10.1007/978-3-030-00934-2_92#citeas
URL Documento de licencia: https://repositori.urv.cat/ca/proteccio-de-dades/
Referencia al articulo segun fuente origial: Automatic Evaluation Of Disclosure Risks Of Text Anonymization Methods. 11071 LNCS 833-840
Referencia de l'ítem segons les normes APA: Singh, Vivek Kumar; Romani, Santiago; Rashwan, Hatem A; Akram, Farhan; Pandey, Nidhi; Kamal Sarker, Md Mostafa; Abdulwahab, Saddam; Torrents-Barrena, (2018). Conditional generative adversarial and convolutional networks for X-ray breast mass segmentation and shape classification. Heidelberg: Springer Berlin Heidelberg
DOI del artículo: 10.1007/978-3-030-00934-2_92
Entidad: Universitat Rovira i Virgili
Año de publicación de la revista: 2018
Tipo de publicación: Proceedings Paper