Abstract
Deep brain stimulation (DBS) is a surgical treatment for advanced Parkinson’s disease (PD) that has undergone technological evolution that parallels an expansion in clinical phenotyping, neurophysiology, and neuroimaging of the disease state. Machine learning (ML) has been successfully used in a wide range of healthcare problems, including DBS. As computational power increases and more data become available, the application of ML in DBS is expected to grow. We review the literature of ML in DBS and discuss future opportunities for such applications. Specifically, we perform a comprehensive review of the literature from PubMed, the Institute for Scientific Information’s Web of Science, Cochrane Database of Systematic Reviews, and Institute of Electrical and Electronics Engineers’ (IEEE) Xplore Digital Library for ML applications in DBS. These studies are broadly placed in the following categories: (1) DBS candidate selection; (2) programming optimization; (3) surgical targeting; and (4) insights into DBS mechanisms. For each category, we provide and contextualize the current body of research and discuss potential future directions for the application of ML in DBS.
Highlights
Deep brain stimulation (DBS) has evolved into an effective therapeutic agent for Parkinson’s disease (PD) patients suffering from medication refractory tremor, motor fluctuations, and/or troublesome dyskinesia
The application of Machine learning (ML) in DBS treatment for PD is a rapidly growing, interdisciplinary field that has unprecedented potential to transform all aspects of DBS
This review examined the role of ML
Summary
Deep brain stimulation (DBS) has evolved into an effective therapeutic agent for Parkinson’s disease (PD) patients suffering from medication refractory tremor, motor fluctuations, and/or troublesome dyskinesia. Brain Sci. 2020, 10, 809 technology, including improvements in spatial control from segmented electrodes and image guided programming, along with brain sensing (a likely precursor for closed loop DBS) can benefit from the application of data driven modeling to enhance clinical decision making and outcome prediction. The algorithm’s general goal is to learn a mapping of inputs to outputs To accomplish this task, the algorithm is provided a training set of input–output pairs. The algorithm is provided a training set of input–output pairs These inputs are generally multi-dimensional vectors representing features of the data. The performance of a supervised learning algorithm is objectively evaluated on a testing set, a subset of the data that is never examined by the algorithm when training. We compartmentalize these approaches into four groups: (1) DBS candidate selection; (2) programming approaches; (3) surgical targeting; and (4) insights into DBS mechanisms
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
