Table 2.

Review studies on radiomics in breast imaging published in peer reviewed journals from 01/2018 to 01/2021, ordered by newest first.

Reference	Modality/Techique	Purpose	Radiomics Features Category and Purpose	Population	Results	Conclusion
Reig et al., 2020 [50]	MRI	Review focused on machine learning techniques in breast MRI	Pre-processing, neural networks, deep learning, machine learning, segmentation, texture analysis	Breast malignant and benign pathology.		The Author discuss the possible future directions of machine learning in the current workflow of breast lesions assessed with MRI.
Granzier et al., 2019 [51]	MRI	Systematic review, response prediction of neoadjuvant therapy	Various radiomic feature models, evaluated with the Radiomics Quality Score (RQS)	Studies ranging between 35-414 BC	AUC values ranged from 0.83 to 0.85. The best performing multivariate prediction model, based on logistic regression analysis, showed AUC of 0.94.	The systematic review revealed large heterogeneity for each step of the MRI-based radiomics workflow. Consequently, the results are difficult to compare.