Objective: Develop an open source toolbox that facilitates data extraction, analysis, artifact correction and standardization for the DBS research community.

Background: Novel sensing enabled implantable pulse generators (IPG) for deep brain stimulation (DBS) like the Medtronic Percept provide an opportunity to record, extract and analyse brain signals directly from the target areas. However, data extraction and analysis methods are not provided by the manufacturers and the data format does not adhere to common research data standards.

Method: The toolbox has been developed in Matlab and creates standardized output from all research data modalities provided by the Medtronic Percept implant. The output structure is inspired by the Brain Imaging Data Structure (BIDS) to facilitate transparent and reproducible data communication.

Results: We have developed an open-source and widely adopted toolbox for the extraction and analysis of Medtronic Percept research data with a single Matlab command. Automatic preprocessing performs metadata anonymization, alignment of multiple data types, such as stimulation amplitude, peak power and raw data. Data recorded during stimulation is automatically run through an ECG artifact removal algorithm.

Conclusion: The Perceive toolbox provides researchers with a practical tool for Percept research data. In the future we aim to extend the usability of the tool for external synchronization, various and improved ECG removal algorithms and a more stringent BIDS standardization.

« Back to 2022 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/perceive-a-data-extraction-and-analysis-toolbox-for-brain-signals-from-sensing-enabled-deep-brain-stimulators/