Bone reporting and data system on CT (Bone-RADS-CT): a validation study by four readers on 328 cases from three local...

Area of Science:

• Radiology and Imaging
• Oncology
• Orthopedics

Background:

• Incidental solitary bone lesions require accurate characterization to guide management.
• Standardized reporting systems improve diagnostic consistency and patient care.
• The Bone Reporting and Data System on CT (Bone-RADS-CT) aims to standardize reporting for bone lesions.

Purpose of the Study:

• To evaluate the reproducibility and effectiveness of the Bone-RADS-CT for incidental solitary bone lesions in adults.
• To assess the diagnostic performance of Bone-RADS-CT in distinguishing benign from potentially malignant or intermediate bone lesions.
• To compare the agreement among musculoskeletal and non-musculoskeletal radiologists using the Bone-RADS-CT algorithm.

Main Methods:

• Retrospective analysis of 328 CT cases with confirmed bone lesions.
• Categorization of lesions using the Bone-RADS algorithm by radiologists with varying expertise.
• Evaluation of inter-reader agreement using weighted kappa statistics.
• Assessment of diagnostic performance (sensitivity, specificity, accuracy) against histopathology, clinical diagnosis, or follow-up.

Main Results:

• Moderate inter-reader agreement was observed for both musculoskeletal (kappa=0.553) and non-musculoskeletal radiologists (kappa=0.403).
• The Bone-RADS-CT demonstrated high sensitivities (88.8%-94.6%) for identifying lesions needing treatment.
• Specificities ranged from 42.9% to 71.1%, with overall accuracies between 78.0% and 86.6%.

Conclusions:

• Bone-RADS-CT is an effective tool for identifying adult bone lesions that warrant further investigation or treatment.
• The system shows moderate reliability among radiologists with different specialties and experience levels.
• Further refinement of the Bone-RADS-CT rating method and category definitions may enhance its performance.