22 March 2022 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Cerebellar ataxia (CA) is concerned with the incoordination of movement caused by cerebellar dysfunction. Movements of the eyes, speech, trunk, and limbs are affected. Conventional machine learning approaches utilizing centralised databases have been used to objectively diagnose and quantify the severity of CA. Although these approaches achieved high accuracy, large scale deployment will require large clinics and raises privacy concerns. In this study, we propose an image transformation-based approach to leverage the advantages of state-of-the-art deep learning with federated learning in diagnosing CA. We use motion capture sensors during the performance of a standard neurological balance test obtained from four geographically separated clinics. The recurrence plot, melspectrogram, and poincaré plot are three transformation techniques explored. Experimental results indicate that the recurrence plot yields the highest validation accuracy (86.69%) with MobileNetV2 model in diagnosing CA. The proposed scheme provides a practical solution with high diagnosis accuracy, removing the need for feature engineering and preserving data privacy for a large-scale deployment.
12 October 2020 in IEEE Transactions on Biomedical Engineering
Cerebellar ataxia (CA) refers to the disordered movement that occurs when the cerebellum is injured or affected by disease. It manifests as uncoordinated movement of the limbs, speech, and balance. This study is aimed at the formation of a simple, objective framework for the quantitative assessment of CA based on motion data. We adopted the Recurrence Quantification Analysis concept in identifying features of significance for the diagnosis. Eighty-six subjects were observed undertaking three standard neurological tests (Romberg's, Heel-shin and Truncal ataxia) to capture 213 time series inertial measurements each. The feature selection was based on engaging six different common techniques to distinguish feature subset for diagnosis and severity assessment separately. The Gaussian Naive Bayes classifier performed best in diagnosing CA with an average double cross-validation accuracy, sensitivity, and specificity of 88.24%, 85.89, and 92.31%, respectively. Regarding severity assessment, the voting regression model exhibited a significant correlation (0.72 Pearson) with the clinical scores in the case of the Romberg's test. The Heel-shin and Truncal tests were considered for diagnosis and assessment of severity concerning subjects who were unable to stand. The underlying approach proposes a reliable, comprehensive framework for the assessment of postural stability due to cerebellar dysfunction using a single inertial measurement unit.
8 February 2023 in IEEE Access
Cerebellar ataxia is the poorly coordinated movement that results from injury or disease affecting the cerebellum. The diagnosis and assessment of ataxia are significantly challenging due to dependency on clinicians’ experience and the attendant subjectivity of such a process. In recent years, neuroimaging and sensor-based approaches, supported by effective machine learning techniques have made advances in the pursuit of addressing these clinical challenges. In this work, we present an outline of approaches to applying machine learning to this clinical challenge. We first provide a fundamental clinical overview with practical problems and then from a machine learning perspective, outline possible approaches with which to address these clinical challenges. Also discussed are the limitations in existing methods, the provision of cross disciplinary approaches and the current state-of-the-art as a potential basis for future research.