Fast Extraction Of Somatosensory Evoked Potential By Using Multi-Adaptive Filter Based On FPGA

Somatosensory evoked potentials (SEP) is an electrophysiological response to external stimulation, reflecting the integrity of the neurological pathway. SEP measurement is an essential test in spinal intraoperative monitoring.In clinical practices, SEP signal is usually accompanied by different kinds of noises, which make extremely poor signal-to-noise ration (SNR) during recording. Therefore, it usually uses conventional ensemble averaging (EA) to extract the SEP signal. However, this method requires a huge number of inputs in order to extract a reliable SEP, which may result in permanent damage of the spinal cord. Therefore, study of real time SEP extraction technique is necessary.This study presents a multi-adaptive filter (MAF) combining the technique of adaptive filter (AF) with neural network. The raw SEP is first denosied by adaptive noise canceller (ANC). Afterward, its output is estimated by radial basis function neural network (RBFNN) for extracting SEP features. The main purpose of this study is to evaluate the performance of new developed MAF in a large scale experiments, including simulation study and animal study. Simulation experiment provided a quantitative assessment of the MAF performance. Animal experiment verified the effectiveness of this method in real SEP signals. The experiment of damaging spinal cord even proved that MAF can provide the information of spinal cord damage much fast than the EA method.For further clinical study, we develop a real-time signal processing system based on Field Programmable Gate Array (FPGA). We implement MAF on this system, and verified the effectiveness of this real-time FPGA based MAF for SEP extraction.