Detection of Breast Cancer Through Mammogram Using Wavelet-Based LBP Features and IWD Feature Selection Technique

• Published in: SN computer science Vol. 3; no. 2; p. 175
• Main Authors: Kalita, Dhruba Jyoti, Singh, Vibhav Prakash, Kumar, Vinay
• Format: Journal Article
• Language: English
• Published: Singapore Springer Singapore 01.03.2022; Springer Nature B.V
• Subjects: Ant colony optimization; Breast cancer; CAD; Classification; Computer aided design; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Data Structures and Information Theory; Datasets; Feature selection; Genetic algorithms; Heuristic; Heuristic methods; Information Systems and Communication Service; Mammography; Medical research; Mortality; Optimization algorithms; Optimization techniques; Original Research; Particle swarm optimization; Pattern Recognition and Graphics; Simulated annealing; Software Engineering/Programming and Operating Systems; Support vector machines; Vision; Water drops; Mammography; Discrete wavelet transform; Local binary patterns; Meta-heuristic optimization; Computer-aided diagnosis
• ISSN: 2662-995X; 2661-8907
• DOI: 10.1007/s42979-022-01071-7

Breast cancer is as one of the common reasons of deaths in women. To detect this cancer in early stage, a computer-aided diagnosis (CAD) system can be designed using X-ray mammograms, which further can assist the doctors and radiologists in their clinical decisions. This paper work proposes a CAD system which is designed by extracting the local binary patterns (LBP) in wavelet domain from mammograms and integrating intelligent water drop (IWD) algorithm to extract most important set of features from extracted features set. With this selected set of features, we then train a support vector machine (SVM) to classify new mammograms. The accuracy, precision and recall rates for our proposed CAD system are found to be 97.9%, 95.499% and 97.3%, respectively. Further, it has been found that the proposed model outperforms many of the existing CAD systems. We have also compared IWD with other Meta-heuristic features selection techniques such as ant colony optimization (ACO), particle swarm optimization (PSO), simulated annealing (SA) and genetic algorithm (GA) and found that it outperforms the others.