Detection and classification of breast cancer availing deep canid optimization based deep CNN

Breast cancer is one of the substantial diseases that affect millions of females each year also the velocity of affected individuals is rising every year. Timely recognition of the illness is the only feasible solution to reduce its influence of the disease. Numerous techniques are invented by researchers in support of the determination of breast cancer and the usage of histopathology descriptions provided the auspicious solution. As an enhancement, in this research, a Deer-Canid based deep CNN is implemented by means of the histopathology images used for the detection of breast cancer through the taxonomy of benign, malignant, and normal regions. The segmentation of the histopathology images is performed using the V-net architecture that segments the image without losing its originality. The primary involvement of the research relies on the Deer-Canid optimization that helps in attaining the global best solution and effectively minimizes the time taken for the classification. The superiority of the research is proved by measuring the values of accuracy, precision, recall, and f1 measure, and the proposed Deer Canid optimization-based deep CNN attained the values of 92.967%, 94.342%, 93.454%, 92.896%, which is more efficient.