Our results demonstrated that the ACR Bone-RADS has a high reproducibility, with a substantial interobserver agreement on almost all the assessment categories of the score, even among radiologists with different experience in bone tumor imaging.

The score revealed a high accuracy in the identification of benign lesions; however, it can underestimate the risk in cases of cartilaginous tumors. Therefore, we recommend incorporating lesion size into the decision-making process and considering additional imaging instead of follow-up for a chondroid lesion with Bone-RADS score=2.   We believe the ACR Bone-RADS system is a promising approach for improving the accuracy of bone lesions first assessment, especially for non-musculoskeletal radiologists. It offers a standardized framework for reporting, facilitating effective communication between radiologists and clinicians while aiding in appropriate patient care and management.