A Study on Variational AutoEncoder to Extract Characteristic Patterns from Electroencephalograms During Sleep

Abstract

Meniere’s disease, a type of inner ear disease, is thought to be caused by ischemic lesions in the inner ear. On the other hand, Meniere’s disease is often associated with sleep apnea syndrome, and the relationship between the two has been pointed out. In recent years, many patients with Meniere’s disease have shown improvement in their symptoms after discontinuation or suppression of medication and sleep therapy. In this study, we hypothesized that the Electroencephalogram (EEG) during sleep in patients with Meniere’s disease has a characteristic pattern that is not seen in normal subjects. The EEGs of normal subjects and patients with Meniere’s disease were converted to lower dimensions using a variational auto-encoder (VAE), and the existence of characteristic differences was verified. Sub-sequence was extracted from the EEGs of 20 subjects, which was input to a variational autoencoder and was converted to lower dimensions. The machine learning was conducted for each channel. Latent variables obtained from the VAE were classified using Support Vector Machine (SVM). The results showed that the electrodes located at the back of the head had a higher correct response rate and F value.