Tumor detection and classification in breast mammography based on fine-tuned convolutional neural networks
Ahmed, A., Keshk, A., M. Abo-Seida, O., Sakr, M. (2022). Tumor detection and classification in breast mammography based on fine-tuned convolutional neural networks. EKB Journal Management System, 9(1), 74-84. doi: 10.21608/ijci.2021.103605.1063
Abeer saber Ahmed; Arabi Elsayed Keshk; Osama M. Abo-Seida; Mohamed Sakr. "Tumor detection and classification in breast mammography based on fine-tuned convolutional neural networks". EKB Journal Management System, 9, 1, 2022, 74-84. doi: 10.21608/ijci.2021.103605.1063
Ahmed, A., Keshk, A., M. Abo-Seida, O., Sakr, M. (2022). 'Tumor detection and classification in breast mammography based on fine-tuned convolutional neural networks', EKB Journal Management System, 9(1), pp. 74-84. doi: 10.21608/ijci.2021.103605.1063
Ahmed, A., Keshk, A., M. Abo-Seida, O., Sakr, M. Tumor detection and classification in breast mammography based on fine-tuned convolutional neural networks. EKB Journal Management System, 2022; 9(1): 74-84. doi: 10.21608/ijci.2021.103605.1063

Tumor detection and classification in breast mammography based on fine-tuned convolutional neural networks

IJCI. International Journal of Computers and Information
Article 7, Volume 9, Issue 1, January 2022, Page 74-84  XML PDF (1.17 MB)
Document Type: Original Article
DOI: 10.21608/ijci.2021.103605.1063
View on SCiNiTO View on SCiNiTO
Authors
Abeer saber Ahmed email orcid 1; Arabi Elsayed Keshk2; Osama M. Abo-Seida1; Mohamed Sakr3
1Department of Computer Science, Faculty of Computers and Information, Kafr El-Sheikh University, Kafr El-Sheikh 33511, Egypt
2Computer Science, Faculty of Computers and Information, Menoufia University
3Department of Computer Science, Faculty of Computers and Information, Menoufia University, Menoufia 32511, Egypt
Abstract
Breast cancer (BC) is one of the most dangerous diseases for women. Breast screening is a technique performed to discover BC at an early stage and reduce the mortality rate. Mammography, which allows patients to identify changes in their breasts before they feel them, is the primary screening tool for BC diagnosis. In this study, pretrained convolutional neural networks (CNNs) like visual geometry group (VGG) VGG-16 and VGG-19 are implemented to detect and classify breast tumors on the INbreast dataset. In the proposed model, breast images are initially preprocessed to improve image quality and reduce computation time. Then, the parameters learned in the networks are transferred to learn with the breast parameters to improve the classification results. Therefore, this work utilized to make an efficient manipulation for the obtained information from the large volume of data generated so that that correct classification may enhance the treatment options. Furthermore, in the evaluation stage, four metrics accuracy, sensitivity, specificity, and area under the ROC curve (AUC) were considered to measure the performance of the proposed model. It was found that the proposed model obtained accuracy, sensitivity, specificity, and AUC values of 97.1%, 96.3%, 97.9%, and 0.988%, respectively.
Keywords
breast cancer; machine learning; segmentation; transfer learning; deep learning
Statistics
Article View: 340
PDF Download: 219