Article: Hybrid local descriptor for improved detection of masses in mammographic CAD systems Journal: International Journal of Advanced Intelligence Paradigms (IJAIP) 2023 Vol.25 No.1/2 pp.185 - 199 Abstract: Early detection of breast cancer increases the chances of survival considerably. Computer aided detection (CAD) systems are assistive tools which can render cost-effective, quick and objective decision making. The proposed work addresses the development of a CAD system to automatically classify suspicious regions in mammograms as normal and abnormal, aiming to improve the detection of masses. To this end, we propose to fuse two local descriptors, namely SIFT based bag of words (BoW) and uniform local binary pattern (ULBP), using principal component analysis (PCA). The effect of issues including obscured masses, typical of dense breasts, is alleviated by the use of segmentation-devoid local descriptors. Experiments on the benchmark DDSM database with multilayer perceptron (MLP) classification, on features extracted from 4,316 suspicious regions demonstrate the efficiency of proposed system. The obtained result significantly reduces false positives and false negatives, achieving an accuracy of 92.81% with an F1 score of 0.91. Inderscience Publishers - linking academia, business and industry through research

Title: Hybrid local descriptor for improved detection of masses in mammographic CAD systems

Authors: Devi Vijayan; R. Lavanya

Addresses: Department of Electronics and Communication Engineering, Amrita School of Engineering, Amrita Vishwa Vidyapeetham, Coimbatore – 641112, India ' Department of Electronics and Communication Engineering, Amrita School of Engineering, Amrita Vishwa Vidyapeetham, Coimbatore – 641112, India

Abstract: Early detection of breast cancer increases the chances of survival considerably. Computer aided detection (CAD) systems are assistive tools which can render cost-effective, quick and objective decision making. The proposed work addresses the development of a CAD system to automatically classify suspicious regions in mammograms as normal and abnormal, aiming to improve the detection of masses. To this end, we propose to fuse two local descriptors, namely SIFT based bag of words (BoW) and uniform local binary pattern (ULBP), using principal component analysis (PCA). The effect of issues including obscured masses, typical of dense breasts, is alleviated by the use of segmentation-devoid local descriptors. Experiments on the benchmark DDSM database with multilayer perceptron (MLP) classification, on features extracted from 4,316 suspicious regions demonstrate the efficiency of proposed system. The obtained result significantly reduces false positives and false negatives, achieving an accuracy of 92.81% with an F1 score of 0.91.

Keywords: BoW; bag of words; BIRADS; breast imaging reporting and data system; CAD; computer aided diagnosis; feature fusion; hybrid descriptors; LBP; local binary descriptors; mammogram; MLP; multilayer perceptron; PCA; principal component analysis; SIFT; scale invariant feature transform.

DOI: 10.1504/IJAIP.2023.130822

International Journal of Advanced Intelligence Paradigms, 2023 Vol.25 No.1/2, pp.185 - 199

Received: 20 May 2019
Accepted: 14 Jan 2020
Published online: 11 May 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Hybrid local descriptor for improved detection of masses in mammographic CAD systems

Keep up-to-date