Case Study: Battelle achieves restoration of arm and hand movement for a paralyzed man with MathWorks MATLAB

Battelle Neural Bypass Technology Restores Movement To A Paralyzed Man’s Arm And Hand

Battelle developed the NeuroLife™ neural bypass to restore arm and hand movement for a quadriplegic participant, but faced the challenge of translating huge volumes of intracortical data (96 electrodes at 30,000 samples/sec) into decoded movements and actuating a 130‑channel neuromuscular electrical stimulator with under 100 ms latency. To meet these real‑time constraints, Battelle engineers used MathWorks tools—principally MATLAB and Wavelet Toolbox—to extract features and build decoding algorithms.

Using MathWorks’ MATLAB, Battelle implemented wavelet decomposition and computed mean wavelet power to reduce features from each 100 ms window, trained a custom SVM to classify intended movements, and ran the full pipeline in real time to control the NMES sleeve. The MathWorks‑based solution achieved end‑to‑end processing in about 60–70 ms and restored basic control of the participant’s right forearm and hand—enabling tasks such as gripping a bottle, pouring, and stirring—while also supporting rapid interdisciplinary development.

Battelle

David Friedenberg

Principal Research Statistician