Artifact Detection in Chronically Recorded Local Field Potentials using Long-Short Term Memory Neural Network

Abstract

The process of recording local fields potentials (LFP) can be contaminated by different internal and external sources of noise. To successfully use these recordings, noise must be removed, for which an automatic detection tool is needed to speed up the detection process. This work presents the use of a specific configuration of the recurrent neural network based machine learning approach, known as the long-short term memory (LSTM), in two different settings to identify artifacts and compares the obtained results to a feed forward neural network both in terms of classification performance and computational time. Using spontaneous LFP signals recorded chronically by multisite neuronal probes in behaving rats, our results show that the LSTM model with and without drop out can achieve an accuracy of 87.1%.