Breast cancer detection remains a critical global health concern, necessitating timely and accurate diagnostic methodologies. Thermography has emerged as a promising non-invasive imaging modality for breast cancer detection, leveraging temperature differentials associated with tumor presence. However, interpreting thermo graphic images accurately poses significant challenges, leading to delayed diagnoses and treatment initiation. This paper presents a novel computational methodology tailored for breast cancer detection from thermographic images, focusing on early diagnosis and treatment planning. The proposed system integrates a hybrid model combining Genetic Algorithm (GA) and Ant Colony Optimization (ACO) for feature selection, alongside the MobileNetV3 architecture for feature extraction. Comparative analysis with an existing model utilizing Genetic Algorithm followed by Grey Wolf Optimization Algorithm (GA-GWO) provides insights into the efficacy of different feature selection approaches. Experimental evaluation demonstrates the superiority of the proposed model, achieving an accuracy of 99.6% compared to 97% for the existing approach. The research findings underscore the potential of advanced computational techniques in improving breast cancer diagnosis and highlight avenues for future research in this domain.
Manaswini et al. (Sun,) studied this question.