Prediction of breast cancer based on computer vision and artificial intelligence techniques.

[Display omitted] • The suggested approach BC-AI is utilized to determine the prediction of breast cancer using computer vision. • A data-driven computer-aided diagnostic (CAD) method for identifying patients as malignant, non-cancerous, or neither. • Feature extraction using the GLCM and HOG-based approach. • Self-constructing ensemble learning fuzzy algorithm (S-ELFA) blends a fuzzy methodology with an advanced neural network for optimum breast cancer illness detection and diagnosis. Breast cancer is a leading cause of mortality among women. Early detection will increase the chances of successful treatment and minimize the death rate. Even though many studies have been conducted to detect breast cancer, medical experts still face difficulty distinguishing between malignant and benign tumors. Hence, a technique enabling medical practitioners to effectively identify breast cancer was developed in this study. A computer-aided diagnostic (CAD) tool is used for classifying and diagnosing patients. The input images are pre-processed at the initial stage and a algorithm based on histogram of oriented gradients (HOG) and gray level co-occurrences matrix (GLCM) is applied to extract key features from pre-processed images. Then, shuffle shepherded optimization (SSO) selects the best features from the extracted features. Finally, the proposed self-constructing ensemble learning fuzzy algorithm (S-ELFA) identifies benign and malignant tumors. ROC, sensitivity, specificity, precision, and accuracy metrics were used to evaluate the developed method. [ABSTRACT FROM AUTHOR]