In medical areas, digital image processing allows for improved accuracy in detecting and diagnosing diseases, where better quality images contain large amounts of information that computationally make a difference in computer-aided diagnosis. Traditional methods from library functions in programming languages can be computationally expensive and slow to process. In the case of thresholding by Otsu’s method applied to 8 bits per pixel (bpp) images, 256 calculations are made to determine the best threshold value. In the case of 16 bpp quality, images 65,536 different calculations are required, which increases operations number. This chapter presents a Genetic Algorithm (GA) based on Otsu’s method to optimize the search for the best threshold value in image segmentation. A comparative analysis is performed using threshold value calculated by a standard function from python opencv library and threshold value calculated by GA, both applied to histopathological micrographs of 16 bpp breast cancer images. Results show that GA calculates an outstanding threshold value close to the optimal threshold calculated by the standard function with a negligible error rate of 0.376% (less than 1%) in the worst case for the image binarization process. We conclude that GA in conjunction with Otsu’s method is a viable alternative for high-quality digital image processing, especially for images with higher bpp.