ABSTRACT
BACKGROUND: Knee range of motion (ROM) is an important indicator of knee function. Outside the clinical setting, patients may not be able to accurately assess knee ROM, which may impair recovery following trauma or surgery. This study aims to validate a smartphone mobile application developed to measure knee ROM compared to visual and goniometer ROM measurements. METHODS: A knee ROM Android mobile application was developed to measure knee ROM. Patients ≥ 18 years old presenting to an orthopedic clinic with native knee complaints were approached to participate. Knee ROM was measured bilaterally by an arthroplasty-trained surgeon using 1) vision, 2) goniometer, and 3) the mobile application. Measurements were compared in flexion and extension using a one-way ANOVA with post-hoc Tukey test (alpha = 0.05). RESULTS: 84 knee ROM measurements (40 left, 44 right) were obtained in 47 patients. Median Kellgren-Lawrence grade from available radiographs was grade 3. In flexion, mobile application (117.6 ± 14.7°) measurements were not significantly different from visual (116.1 ± 13.6°) or goniometer (116.2 ± 13.6°) measurements. In extension, mobile application (4.8 ± 7.3°) measurements were significantly different from visual (1.9 ± 4.1°) measurements on post-hoc analysis (p < 0.01), while no differences were present compared to goniometer (3.1 ± 5.8°) measurements. CONCLUSION: Our study found that a mobile application for evaluating knee ROM was non-inferior to goniometer-based measurements performed by an arthroplasty-trained surgeon. Future studies will investigate this application's utility in 1) remote patient care, 2) accelerating recovery during rehabilitation, 3) detecting early postoperative complications including arthrofibrosis, and 4) adding additional functionalities to the application to provide more detail-oriented descriptive analyses of patient knee function.
ABSTRACT
AI-powered segmentation of hip and knee bony anatomy has revolutionized orthopedics, transforming pre-operative planning and post-operative assessment. Despite the remarkable advancements in AI algorithms for medical imaging, the potential for biases inherent within these models remains largely unexplored. This study tackles these concerns by thoroughly re-examining AI-driven segmentation for hip and knee bony anatomy. While advanced imaging modalities like CT and MRI offer comprehensive views, plain radiographs (X-rays) predominate the standard initial clinical assessment due to their widespread availability, low cost, and rapid acquisition. Hence, we focused on plain radiographs to ensure the utilization of our contribution in diverse healthcare settings, including those with limited access to advanced imaging technologies. This work provides insights into the underlying causes of biases in AI-based knee and hip image segmentation through an extensive evaluation, presenting targeted mitigation strategies to alleviate biases related to sex, race, and age, using an automatic segmentation that is fair, impartial, and safe in the context of AI. Our contribution can enhance inclusivity, ethical practices, equity, and an unbiased healthcare environment with advanced clinical outcomes, aiding decision-making and osteoarthritis research. Furthermore, we have made all the codes and datasets publicly and freely accessible to promote open scientific research.