Breast cancer diagnosis based on hybrid SqueezeNet and improved chef-based optimizer

The most frequent disease in women and the one that accounts for the majority of cancer-related fatalities in females is breast cancer. An important milestone in breast cancer CAD systems is the automatic recognition and delineation of masses in mammograms. We have developed a novel technique in thi...

Full description

Saved in:
Bibliographic Details
Published in:Expert systems with applications Vol. 237; p. 121470
Main Authors: Huang, Qirui, Ding, Huan, Effatparvar, Mehdi
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.03.2024
Subjects:
ISSN:0957-4174, 1873-6793
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The most frequent disease in women and the one that accounts for the majority of cancer-related fatalities in females is breast cancer. An important milestone in breast cancer CAD systems is the automatic recognition and delineation of masses in mammograms. We have developed a novel technique in this study to identify potential mass candidates in mammograms automatically. The suggested method is a pipeline method with three stages. First, a noise reduction method based on median filtering is performed to eliminate the noises. Then, the region of interest is threshold out of the pre-processed images using the Kapur technique. Afterward, to eliminate useless features, optimum feature extraction and selection are designed. It is established by an Improved version of the Chef-Based Optimization (ICBO) algorithm. Lastly, a classification is implemented based on an optimal SqueezeNet model for the final diagnosis. This classifier is also optimized based on the ICBO algorithm to maximize efficiency. Simulations of the method are applied to the MIAS database, and its results are compared with different state-of-the-art methods to show the method's advantage.
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2023.121470