The recent proliferation of novel machine learning techniques in quantitative marker-based 3D gait analysis (3DGA) has shown promise for improving interpretations of clinical gait analysis. The objective of this study was to characterize the state of the literature on using machine learning in the analysis of marker-based 3D gait analysis to provide clinical insights that may be used to improve clinical analysis and care. Methods: A scoping review of the literature was conducted using the PubMed and Web of Science databases. Search terms from eight relevant articles were identified by the authors and added to by experts in clinical gait analysis and machine learning. Inclusion was decided by the adjudication of three reviewers. Results: The review identified 4324 articles matching the search terms. Adjudication identified 105 relevant papers. The most commonly applied techniques were the following: support vector machines, neural networks (NNs), and logistic regression. The most common clinical conditions evaluated were cerebral palsy, Parkinson’s disease, and post-stroke. Conclusions: ML has been used broadly in the literature and recent advances in deep learning have been more successful in larger datasets while traditional techniques are robust in small datasets and can outperform NNs in accuracy and explainability. XAI techniques can improve model interpretability but have not been broadly used.
Background: Osteogenesis imperfecta (OI), a heritable connective tissue disorder with wide clinical variability, predisposes to recurrent fractures and bone deformity. Management requires a multidisciplinary approach in which intramedullary rodding plays an important role, especially for moderate and severe forms. We investigated the patterns of surgical procedures in OI in order to establish the benefits of rodding. The main hypothesis that guided this study was that rodded participants with moderate and severe OI would have lower fracture rates and better mobility. Methods: With data from the Linked Clinical Research Centers, we analyzed rodding status in 558 individuals. Mobility and fracture data in OI Types III and IV were compared between rodded and non-rodded groups. Univariate regression analyses were used to test the association of mobility outcomes with various covariates pertinent to rodding. Results: Of the individuals with OI, 42.1% had undergone rodding (10.7% of those with Type I, 66.4% with Type III, and 67.3% with Type IV). Rodding was performed more frequently and at a younger age in femora compared with tibiae. Expanding intramedullary rods were used more frequently in femora. In Type III, the rate of fractures per year was significantly lower (p ≤ 0.05) for rodded bones. In Type III, the mean scores on the Gillette Functional Assessment Questionnaire (GFAQ) and Brief Assessment of Motor Function (BAMF) were higher in the rodded group. However, Type-IV non-rodded subjects had higher mean scores in nearly all mobility outcomes. OI type, the use of expanding rods in tibiae, and anthropometric measurements were associated with mobility outcomes scores. Conclusions: Current practice in 5 orthopaedic centers with extensive experience treating OI demonstrates that most individuals with moderate and severe types of OI undergo rodding procedures. Individuals with severe OI have improved mobility outcomes and lower fracture rates compared with their non-rodded peers, which suggests that early bilateral rodding benefits OI Type III. Our analysis showed a change in practice patterns in the final years of the study in the severe forms, with earlier and more simultaneous rodding procedures performed. Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.