An Accurate and Robust Method for Spike Sorting Based on Convolutional Neural Networks

Abstract

In the fields of neuroscience and biomedical signal processing, spike sorting is a crucial step to extract the information of single neurons from extracellular recordings. In this paper, we propose a novel deep learning approach based on one-dimensional convolutional neural networks (1D-CNNs) to implement accurate and robust spike sorting. The results of the simulated data demonstrated that the clustering accuracy in most datasets was greater than 99%, despite the multiple levels of noise and various degrees of overlapped spikes. Moreover, the proposed method performed significantly better than the state-of-the-art method named "WMsorting" and a deep-learning-based multilayer perceptron (MLP) model. In addition, the experimental data recorded from the primary visual cortex of a macaque monkey were used to evaluate the proposed method in a practical application. It was shown that the method could successfully isolate most spikes of different neurons (ranging from two to five) by training the 1D-CNN model with a small number of manually labeled spikes. Considering the above, the deep learning method proposed in this paper is of great advantage for spike sorting with high accuracy and strong robustness. It lays the foundation for application in more challenging works, such as distinguishing overlapped spikes and the simultaneous sorting of multichannel recordings.

Keywords: convolutional neural network; deep learning; extracellular recording; spike sorting.

Figure 1

Architecture of the proposed convolutional neural network (CNN) model used in the simulated database. The blue region corresponds to convolutional processes, and the red region represents the pooling. The stride of all convolutional layers was 1.

Figure 2

Details of the simulated database. Different colors represent different clusters. The number in parentheses represents the number of overlapped spikes.

Figure 3

Details of the experimental datasets. Different colors represent different clusters. The number in parentheses represents the number of overlapped spikes.

Figure 4

Proportion of data in training set and testing set for the six experiments. E = experiment.

Figure 5

Confusion matrix: (a) C_Easy1_noise015; (b) C_Difficult1_noise015.

Figure 6

Waveforms of spikes in each cluster: (a–c) C_Easy1_noise015; (d–f) C_Difficult1_noise015.

Figure 7

Falsely classified spikes and randomly selected correctly classified spikes: (a,b) C_Easy1_noise015; (c–e) C_Difficult1_noise015.

Figure 8

The cross-entropy loss function on the training set and testing set: (a) C_Easy1_015; (b) C_Difficult1_015.

Figure 9

Classification accuracies of the recording channels with different numbers of spike clusters: (a) Three clusters; (b) two clusters; (c) four clusters; (d) five clusters.