EVALUATION OF MACHINE LEARNING AND DEEP LEARNING APPROACHES TO CLASSIFY BREAST CANCER USING THERMOGRAPHY

Main Article Content

P. KANIMOZHI, S. SATHIYA, M. BALASUBRAMANIAN, P. SIVAGURU, P. SIVARAJ

Abstract

Emerging technologies undergo the challenge of diagnosing fatal diseases even before any symptoms could occur in human body.  One such syndrome is breast cancer which is now common in developing countries like India and increasing rapidly among women. The only remedy is diagnosing the disease earlier using latest technology like thermography, one of the best methods for cancer screening test without any harmful effect. This paper concentrates on various sectors namely segmentation where two approaches are discussed, next is feature extraction method used to extract the vital features for classification and last is classification technique employed for both machine learning and deep learning methods. Of all the models discussed Residual Network model proves highest accuracy of 96% in classification. The  models performance are related by training the thermal images and a comparative analysis is performed.

Article Details

Section
Articles