Can bone tumor radiology reporting systems be used in children?: a validation study for Bone-RADS and OT-RADS in MRI in the evaluation of solitary bone tumors.

Abstract

Bone tumor radiology reporting systems have been developed, including Bone-RADS and OT-RADS, aimed at standardizing imaging studies' reports and guiding the management of bone lesions. This study aims to evaluate and compare the use of Bone Tumor Radiology Reporting Systems, Bone-RADS and OT-RADS, in MRI of children with solitary bone lesions. Patients younger than 18 years old who had an MRI for the evaluation of a solitary bone lesion between January 2015 and October 2025 were included in the study. Five radiologists with different expertise levels evaluated the MRI examinations. Bone-RADS and OT-RADS were assessed using both their original categorical systems and binary classification models. We assessed inter-rater reliability using Cohen's kappa and Fleiss' kappa. For both scoring systems, a binary classification was used to assess diagnostic performance. Eighty patients (mean age - standard deviation/minimum-maximum: 11.95 - 3.65/4-17 years old; 66 male) were included in the study. For both Bone-RADS and OT-RADS, the overall agreement between all raters, as assessed using Fleiss Kappa, showed moderate agreement (k = 0.470, 95% CI: 0.410-0.530) and (k = 0.432, 95% CI: 0.389-0.475), respectively. The accuracy of Bone-RADS ranged from moderate to good, and OT-RADS from moderate to excellent. Both Bone-RADS and OT-RADS were reliable, and demonstrated high accuracy in evaluating solitary bone tumors in children. Bone-RADS has a higher level of interrater agreement, which we attributed to the guidance provided by its algorithm. OT-RADS requires knowledge of each disease due to the lack of a comprehensive lexicon and an algorithm. Question Can bone tumor radiology reporting systems, Bone-RADS and OT-RADS in MRI, be used in the evaluation of solitary bone tumors? Findings Both Bone-RADS and OT-RADS demonstrated moderate interrater agreement in the evaluation of solitary bone tumors in children. Bone-RADS revealed moderate to good accuracy, while OT-RADS revealed moderate to excellent accuracy. Clinical relevance Bone-RADS and OT-RADS were reliable and demonstrated high accuracy in evaluating solitary bone tumors in children. These bone reporting systems can be used for solitary bone tumors in children on MRI.