Vibhor Wadhwa1, Vidur Malhotra2, Yin Xi2, Shaun Nordeck2, Katherine Coyner3, Avneesh Chhabra4. 1. Department of Radiology, University of Arkansas for Medical Sciences, Little Rock, AR; Musculoskeletal Radiology, University of Texas Southwestern Medical Center, Dallas, TX. 2. Musculoskeletal Radiology, University of Texas Southwestern Medical Center, Dallas, TX. 3. Orthopaedic Surgery, University of Texas Southwestern Medical Center, Dallas, TX. 4. Musculoskeletal Radiology, University of Texas Southwestern Medical Center, Dallas, TX. Electronic address: avneesh.chhabra@utsouthwestern.edu.
Abstract
PURPOSE: The purpose was to evaluate feasibility of bone and joint segmentations from three-dimensional magnetic resonance imaging (3D MRI). METHODS: Segmented joint models from 3D MRI data set were obtained for 42 patients. Blinded angular and joint space measurements were performed on 3D MRI model, two-dimensional (2D) MRI, and radiography (XR). RESULTS: Medial joint space was similar on both XR and 3D MRI (P=.3). The XR measurements were statistically different but closer to 3D MRI for lateral patellar tilt angle, patellar tendon length, and lateral knee joint space, whereas 2D MRI measurements were closer to XR in terms of trochlear depth, sulcal angle, and patellar length. CONCLUSION: 3D bone and joint segmentations are feasible from isotropic MRI data sets.
PURPOSE: The purpose was to evaluate feasibility of bone and joint segmentations from three-dimensional magnetic resonance imaging (3D MRI). METHODS: Segmented joint models from 3D MRI data set were obtained for 42 patients. Blinded angular and joint space measurements were performed on 3D MRI model, two-dimensional (2D) MRI, and radiography (XR). RESULTS: Medial joint space was similar on both XR and 3D MRI (P=.3). The XR measurements were statistically different but closer to 3D MRI for lateral patellar tilt angle, patellar tendon length, and lateral knee joint space, whereas 2D MRI measurements were closer to XR in terms of trochlear depth, sulcal angle, and patellar length. CONCLUSION: 3D bone and joint segmentations are feasible from isotropic MRI data sets.