International Conference on Computer and Knowledge Engineering

Home / 13th International Conference on Computer and Knowledge Engineering

An Efficient Approach for Breast Abnormality Detection through High-Level Features of Thermography Images

Authors :

Farhad Abedinzadeh Torghabeh¹ Yeganeh Modaresnia² Seyyed Abed Hosseini³

1- Department of Biomedical Engineering, Mashhad Branch, Islamic Azad University, Mashhad, Iran 2- Department of Biomedical Engineering, Mashhad Branch, Islamic Azad University, Mashhad, Iran 3- Department of Electrical Engineering, Mashhad Branch, Islamic Azad University, Mashhad, Iran

Keywords :

breast abnormalities،thermal imaging،deep transfer learning

Abstract :

Breast cancer is a significant global health concern affecting millions of women worldwide. Timely detection is paramount in improving prognosis and survival rates. In this context, infrared thermography has emerged as a promising noninvasive imaging modality for breast cancer diagnosis. This study used raw and pre-processed images using contrast-limited adaptive histogram equalization (CLAHE) and contrast enhancement techniques. One of the key challenges in analyzing breast infrared thermogram images is extracting meaningful features that can aid in accurate diagnosis. To address this, three well-known pre-trained convolutional neural networks, such as AlexNet, GoogLeNet, and SqueezeNet, were used to extract high-level features automatically. Subsequently, the resulting features were subjected to the principal component analysis and the top 100 features were selected, which were then utilized as input for supervised learning classifiers. The proposed method was validated using a publicly available DMR dataset of 100 healthy and 100 abnormal breast thermal images. Notably, the proposed methodology achieves an outstanding accuracy of 99.60% and sensitivity of 100% by employing AlexNet features derived from CLAHE pre-processed images in conjunction with a decision tree classifier. These results underscore the efficacy of the proposed approach in accurate breast cancer detection using infrared thermography.