# SimpleKANSleepNet: a Kolmogorov–Arnold network based sleep stage classification method

## TL;DR

This paper introduces SimpleKANSleepNet, a new machine learning model for classifying sleep stages using EEG and other signals, achieving high accuracy.

## Contribution

The novel contribution is a KAN-based model with learnable activation functions for improved sleep stage classification.

## Key findings

- SimpleKANSleepNet outperforms CNN and GCN methods on sleep stage classification with high accuracy.
- The model achieves 0.928 F1-score and 0.910 Cohen’s kappa on the Sleep-EDF-153 dataset.
- Data balancing methods and feature types are tested to enhance classification performance.

## Abstract

A novel Kolmogorov–Arnold Network (KAN) based machine learning model is proposed for the automatic sleep stage classification task. The redefined architecture of the Multilayer Perceptron (MLP) aims to build a more flexible model by using learnable activation functions. In this study, an effective KAN model named SimpleKANSleepNet is evaluated on two different datasets with temporal features and frequency features extracted from electroencephalography (EEG), electromyogram (EMG), electrooculogram (EOG), and electrocardiogram (ECG) signals through a dual-stream convolutional neural network (CNN). Compared with existing CNN-based methods and graph convolutional networks (GCNs), the proposed model achieves an overall classification accuracy, F1-score, and Cohen’s kappa on the ISRUC-S1 and the Sleep-EDF-153 datasets of 0.812, 0.793, 0.757, 0.928, 0.929, and 0.910, respectively, which demonstrates its competitive classification performance and generality. Moreover, several data balancing methods are tested on Sleep-EDF-153 to further evaluate the potential for achieving the best results. Finally, the factors that may affect the classification ability are tested on the ISRUC-S1 dataset.