Invasive neurophysiology and whole brain connectomics for neural decoding in patients with brain implants

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Neuroscience

Background:

• Brain-computer interfaces (BCIs) offer potential for closed-loop neuromodulation therapies.
• Precise treatment of brain disorders requires decoding dynamic patient states from neural signals.
• A standardized framework for invasive brain signal decoding from neural implants is currently lacking.

Purpose of the Study:

• To develop and validate a platform integrating brain signal decoding with magnetic resonance imaging (MRI) connectomics.
• To enable rapid, high-accuracy decoding for adaptive neurotherapies.
• To demonstrate the clinical utility of brain signal decoding for deep brain stimulation (DBS) and responsive neurostimulation (RNS).

Main Methods:

• Development of a platform integrating brain signal decoding with MRI connectomics.
• Analysis of 123 hours of invasive brain data from 73 neurosurgical patients with implants for movement disorders, depression, and epilepsy.
• Introduction of connectomics-informed decoders for movement, emotion, and seizure detection.

Main Results:

• Connectomics-informed movement decoders demonstrated generalization across patient cohorts with Parkinson's disease and epilepsy from diverse geographical locations.
• Identification of specific network targets (left prefrontal and cingulate circuits) for emotion decoding in deep brain stimulation patients with major depression.
• Demonstrated potential for improved seizure detection in responsive neurostimulation for epilepsy.

Conclusions:

• The developed platform facilitates clinical applications of brain signal decoding for deep brain stimulation and responsive neurostimulation.
• The methods enable rapid, high-accuracy decoding, supporting precision medicine approaches.
• Dynamic adaptation of neurotherapies based on individual patient needs is achievable.