Certain investigations on thermography images for early detection of breast cancer using optimization techniques

Abstract

Breast cancer is one of the most likely cancers that cause death for omen all over the world. The early detection of breast cancer helps in minimizing death rates. The automatic detection of cancer is significant for determining abnormal breast tissue at an early stage. Due to the presence of noise and distortion, the inaccurate identification of disease introduces severe problems in human health. Thus, improving the breast cancer detection rate is difficult. Breast cancer images have a huge number of features for identifying the accurate location of the disease. In spite of that the important features are not easily detected for cancer diagnosis. This leads to increased complexities while detecting breast cancers. Recently, several classifications and optimization techniques were employed to detect breast cancer at an early stage. Yet, the accurate detection was not performed with minimal time complexity. Recently, a lot of research works have been designed to detect cancer-affected breast images. The performances of existing techniques were not sufficient due to the presence of noise. Several screening processes and classification methods are presented to detect the tumor at an early stage. However, early detection needs an accurate and reliable diagnosis. The time consumed for detecting breast cancer remained unaddressed. To overcome such issues, three different methods namely the Multimodal Firefly Optimization (MFFO) technique, Statistical Maximum Likelihood Optimization and Curvilinear Support Vector Machine method (SMLOCSVM), and Population Rescaled Differential Evolution newline