Artificial Intelligence in Movement Disorders: Transforming Early Detection, Monitoring, and Precision Treatment
Artificial Intelligence in Movement Disorders: Transforming Early Detection, Monitoring, and Precision Treatment
Where did the research take place?
The study site has not been established. Author addresses may differ from where the research occurred.
A plain-language reading has not been prepared for this paper yet.
Original abstract
Movement disorders are a growing global health issue, affecting millions worldwide with progressive neurological conditions such as Parkinson's disease, essential tremor, Huntington's disease, and dystonia. Traditional diagnostic methods rely heavily on subjective clinical assessments, often detecting these conditions only after significant neurological damage has occurred. The rise of artificial intelligence offers transformative potential for early detection, ongoing monitoring, and optimised precision treatment. This scoping review compiles current evidence on AI applications across the movement disorder care pathway and presents new experimental benchmarks for leading AI architectures. Following PRISMA guidelines, I conducted a thorough search across PubMed, Scopus, Web of Science, and IEEE Xplore, identifying 127 studies that met the inclusion criteria. The experimental evaluation utilised multiple modalities, including neuroimaging, wearable sensors, audio recordings, and digital handwriting, using publicly available datasets. Key findings indicate that AI-enhanced approaches significantly improve prodromal disease detection, with multimodal integration strategies consistently outperforming single-source models. Deep learning architectures achieved diagnostic accuracies exceeding 94% across various datasets, representing notable improvements over traditional clinical assessment methods. The results confirm the superior performance of transformer-based fusion strategies and graph neural networks in capturing complex spatiotemporal patterns of disease. Despite these advancements, substantial challenges remain regarding algorithmic generalisability, clinical explainability, ethical deployment, and fair implementation in resource-limited environments. This review demonstrates that artificial intelligence is fundamentally changing movement disorder care, shifting from reactive symptom management to proactive, personalised, and predictive neurology. Future developments in digital twins, foundation models, federated learning, and explainable AI frameworks are poised to accelerate clinical translation and further enhance global accessibility.