🤖 AI Summary
This study addresses continuous, non-invasive monitoring of motor symptoms in Parkinson’s disease (PD), specifically detecting the presence and quantifying the severity of tremor, rigidity, and dyskinesia from wrist-worn accelerometer time-series data. We systematically compare two state-of-the-art time-series models—ROCKET and InceptionTime—on small-sample PD datasets for the first time. ROCKET achieves superior performance in dyskinesia detection, while InceptionTime slightly outperforms in tremor and rigidity assessment; both significantly surpass conventional MLP baselines. Leveraging ROCKET’s random convolutional kernels, InceptionTime’s deep convolutional architecture, and ridge regression or MLP regressors—optimized via random hyperparameter search—all models attain moderate accuracy (mean absolute error ≤ 0.8 clinical severity levels). Results demonstrate the efficacy and clinical potential of lightweight time-series models for long-term, at-home PD monitoring.
📝 Abstract
Parkinson's disease (PD) is a neurodegenerative condition characterized by frequently changing motor symptoms, necessitating continuous symptom monitoring for more targeted treatment. Classical time series classification and deep learning techniques have demonstrated limited efficacy in monitoring PD symptoms using wearable accelerometer data due to complex PD movement patterns and the small size of available datasets. We investigate InceptionTime and RandOm Convolutional KErnel Transform (ROCKET) as they are promising for PD symptom monitoring. InceptionTime's high learning capacity is well-suited to modeling complex movement patterns, while ROCKET is suited to small datasets. With random search methodology, we identify the highest-scoring InceptionTime architecture and compare its performance to ROCKET with a ridge classifier and a multi-layer perceptron (MLP) on wrist motion data from PD patients. Our findings indicate that all approaches can learn to estimate tremor severity and bradykinesia presence with moderate performance but encounter challenges in detecting dyskinesia. Among the presented approaches, ROCKET demonstrates higher scores in identifying dyskinesia, whereas InceptionTime exhibits slightly better performance in tremor and bradykinesia estimation. Notably, both methods outperform the multi-layer perceptron. In conclusion, InceptionTime can classify complex wrist motion time series and holds potential for continuous symptom monitoring in PD with further development.
