Laleh Rahmanian - Department of Radiologic Technology, School of Allied Medical Sciences, AhvazJundishapur University of Medical Sciences, Ahvaz, Iran.
Marziyeh Tahmasbi - Department of Radiologic Technology, School of Allied Medical Sciences, AhvazJundishapur University of Medical Sciences, Ahvaz, Iran.

Breast cancer is one of the primary causes of cancer death among women. Early detection allows patients to receive appropriate treatment, thus increasing the possibility of survival. Concerning the potential of artificial intelligence (AI) approaches as quantitative tools to improve the diagnosis and classification of breast lesions, this review aimed to report different radiomic features and classifiers used for the detection and classification of breast lesions using AI techniques and the accuracy and sensitivity of these methods. PubMed, Science Direct, Web of Science, and Google Scholar databases were explored up to July ۲۰۲۲, using different combinations of the keywords: “radiomics”, “breast cancer”, “deep learning”, “machine learning”, “texture features”, and “artificial intelligence”. The results were screened. Seven more recent and relevant papers were included in the study. Based on the results of reviewed articles, various machine learning and deep learning techniques have been proposed for automatic breast cancer detection. Among different applied radiomics, morphologic features, as well as, statistical features extracted from breast images, were the most used training features by the AI methods for the detection and classification of breast lesions. The most frequent deep learning models were convolutional neural networks (CNN models) with the maximum classification accuracy (about ۹۹ %) and sensitivity (about ۹۸%). Supported Vector Machine (SVM) was the most popular machine learning technique with maximum accuracy and sensitivity of about ۹۹.۷% and ۹۸%, respectively. In conclusion, AI and radiomic features can offer a powerful tool for the quantitative assessment and early diagnosis of breast cancer.

Breast Cancer, Classification, Machine Learning, Deep Learning, Artificial Intelligence.