The AWARrD aim investigators will identify patients in the MRI databases from Assiut University and the University of Chicago. At the University of Chicago 150 patients will be selected, 50 patients each with leiomyomas, leiomyosarcomas, or atypical or smooth muscle tumor of uncertain malignant potential (STUMP). All will have had a preoperative MRI and histopathological evaluation. From the Assiut MRI database 100 patients with an equal distribution of the the same diagnoses. We will use this MRI image database with clinical and imaging features of these patients and, images will be deidentified. Out of this dataset, approximately 75 will be randomly selected to represent the derivation cohort on which the initial model will be developed and trained. Afterwards the remaining 25 samples will be used as a validation cohort on which the trained algorithm will be applied. This study will utilize computer software at the University of Chicago to define specific aspects of uterine fibroids on MR images including location, size, morphology, nodular borders, hemorrhage, enhanced and unenhanced areas, diffusion coefficients, location, size, morphology, and displacement using machine learning – both human-engineered features and deep learning features. We will use the software to determine how these image aspects relate or are unique to each category of lesion: Leiomyoma, atypical leiomyoma or leiomyosarcoma.