Mammogram content-based image retrieval based on malignancy classification

Article type: Research Article

Authors: Chikamai, Keith^{a; *} | Viriri, Serestina^a | Tapamo, Jules-Raymond^b

Affiliations: [a] School of Mathematics, Statistics and Computer Science, University of KwaZulu-Natal, Durban 4000, South Africa | [b] School of Computer Engineering, University of KwaZulu-Natal, Durban 4001, South Africa

Correspondence: [*] Corresponding author: Keith Chikamai, School of Mathematics, Statistics and Computer Science, University of KwaZulu-Natal, WestVille campus, Durban, 4000, South Africa. Tel. : +27 0 31 260 3230; Fax: +27 0 31 260 7001; E-mail: [email protected].

Abstract: Content-based image retrieval (CBIR) technique is increasingly gaining research attention as a Computer Aided Diagnosis (CAD) approach for breast cancer diagnosis. This work discusses a novel feature modeling technique for CBIR systems based on classifier scores and standard statistical calculations on the same. Established textural and geometric features are initially used to represent medical characteristics, before being used to generate secondary features through classifier scoring using the Support Vector Machine and Quadratic Discriminant Analysis classifiers. The model is validated through a range of benchmarks, and is shown to perform competitively in comparison to similar works.

Keywords: Mammography, microcalcifications, image processing, CBIR, machine learning, computer aided diagnosis

DOI: 10.3233/IDA-163101

Journal: Intelligent Data Analysis, vol. 21, no. 5, pp. 1193-1212, 2017