Evaluation of Machine Learning Models for Breast Cancer Diagnosis Via Histogram of Oriented Gradients Method and Histopathology Images

Breast cancer is the main death rate from malignant growth worldwide and most frequently diagnosed type of in females. Machine learning systems have been developed to assist accurate detection cancer. There are numerous methods for detection. But histopathological images thought be more precise. In this study, we used HOG features extractor extract statistical histopathology invasive ductal carcinoma. We chose following at random images: 100, 200, 400, 1000, 2000. These were then train several algorithms, including decision tree, quadratic discriminant analysis, extra randomized trees, gradient boosting, gaussian process classifier, naive bayes, nearest centroid, multilayer perceptron, support vector machine, identify whether or not depict cancerous noncancerous growth. The algorithms' performance was evaluated depending on specificity, accuracy, sensitivity, precision, F1_score, AUC. algorithms worked best when number set 100. As went up, their effectiveness down.