A Flexible Pulse Generator Based on a Field Programmable Gate Array Architecture for Functional Electrical Stimulation

Abstract

Non-invasive Functional Electrical Stimulation (FES) is a technique applied for motor rehabilitation of patients with central nervous system injury. This technique requires programmable multichannel systems to configure the stimulation parameters (amplitude, frequency, and pulse width). Most FES systems are based on microcontrollers with fixed architecture; this limits the control of the parameters and the scaling to multiple channels. Although field programmable gate arrays (FPGA) have been used in FES systems as alternative to microcontrollers, most of them focus on signal acquisition, processing, or communication functions, or are for invasive stimulation. A few FES systems report using FPGAs for parameter configuration and pulse generation in non-invasive FES. However, generally they limit the value of the frequency or amplitude parameters to enable multichannel operation. This restricts free selection of parameters and implementation of modulation patterns, previously reported to delay FES-induced muscle fatigue. To overcome those limitations, this paper presents a proof-of-concept (technology readiness level three-TRL 3) regarding the technical feasibility and potential use of an FPGA-based pulse generator for non-invasive FES applications (PG-nFES). The main aims were: (1) the development of a flexible pulse generator for FES applications and (2) to perform a proof-of-concept of the system, comprising: electrical characterization of the stimulation parameters, and verification of its potential for upper limb FES applications. Biphasic stimulation pulses with high linearity (r2 > 0.9998) and repeatability (>0.81) were achieved by combining the PG-nFES with a current-controlled output stage. Average percentage error in the characterizations was under 3% for amplitude (1–48 mA) and pulse width (20–400 μs), and 0% for frequency (10–150 Hz). A six-channel version of the PG-nFES was implemented to demonstrate the scalability feature. The independence of parameters was tested with three patterns of co-modulation of two parameters. Moreover, two complete FES channels were implemented and the claimed features of the PG-nFES were verified by performing upper limb functional movements involving the hand and the arm. Finally, the system enabled implementation of a stimulation pattern with co-modulation of frequency and pulse width, applied successfully for efficient elbow during repetitions of a functional movement.

Functional Electrostimulation System for a Prototype of a Human Hand Prosthesis Using Electromyography Signal Classification by Machine Learning Techniques

“…Although FES has been studied for several years, the signal's amplitude is still ambiguous, especially depending on the level of the patient's motion, which could not be complete due to the sickness condition [15,[30][31][32]. We considered a fundamental signal proposed as a reference signal satisfying a biphasic shape with a sawtooth shape of 30 Hz and a pulse width of 100 µs [15,[30][31][32]. At the same time, the trajectories of the simulated hand were obtained and used as the trajectory of reference.…”