Intelligent sleep staging system based on lightweight convolutional neural network

Abstract

The invention belongs to the technical field of healthy sleep management, and particularly relates to an intelligent sleep staging system based on a lightweight convolutional neural network. The intelligent sleep staging system comprises an electroencephalogram signal acquisition module, an electroencephalogram signal preprocessing module, an electroencephalogram signal analysis algorithm module and a wireless signal transmission module. A result is wirelessly transmitted to an intelligent terminal for display through Bluetooth; the electroencephalogram signal analysis module is a specially-designed lightweight convolutional neural network and is composed of a convolutional layer, a maximum pooling layer, a ReLU activation function, a batch normalization module and a full connection layer; the computing resource overhead can be reduced, the computing resource overhead is mapped to a chip for sleep electroencephalogram data analysis, and end-to-end feature extraction and data analysis functions are realized under the condition of low power consumption; and the algorithm module can be implemented on any hardware based on a lightweight convolutional neural network and an improved version thereof.

Description

technical field [0001] The invention belongs to the technical field of healthy sleep management, and in particular relates to an intelligent sleep staging system based on a lightweight convolutional neural network. Background technique [0002] Sleep health has received widespread attention in recent years. Sleep staging is an important basis for the assessment of sleep quality and the diagnosis of related sleep diseases, such as insomnia, narcolepsy and sleep apnea syndrome. EEG signals are often used for sleep staging. The EEG signals of the whole night's sleep last for a long time, and it is a heavy task for doctors to perform sleep staging through EEG signals. Therefore, the use of deep learning methods based on EEG signals for sleep staging has important clinical value. [0003] However, most of the current deep learning methods for sleep staging based on EEG signals only stay at the software implementation level, and are rarely implemented at the hardware level. Th...

AI Technical Summary

Problems solved by technology

[0005] In addition, the existing EEG signal analysis equipment is usually connected in a wired way, which greatly affects the experience of the subjects in the process of EEG signal acquisition and analysis

In addition, due to the subject's side lying, turning and other movements, the electrode falls off, which will cause the interruption and loss of data collection, and adversely affect the sleep staging of the whole night.

Images