Sleep apnea syndrome recognition device

Abstract

The invention discloses a sleep apnea syndrome recognition device, and the device is characterized in that the input of a convolution kernel channel of a convolutional neural network is improved in a targeted manner, and electrocardio data and respiratory wave data are pre-processed to be served as the input of the neural network; an operation layer of a model is optimized, the number of convolution kernels is adjusted to be the relatively optimal number, the convolution stride is adjusted to be 2, and a Dropout layer is added in front of a full connection layer; and a model output layer is improved, the model is adjusted to be a binary classification result to be output, and a judgment result about whether the sleep apnea event occurs or not is obtained. According to the invention, the electrocardiogram data and the respiratory wave data are fused, so that the detection capability of the model is improved; the traditional LeNet-5 model is optimized, so that the sleep apnea syndrome recognition capability of the model is improved; and the model detection time is shorter than that of a deep neural network, so that the application cost is low, and the application range is wide.

Description

technical field [0001] The invention relates to the technical field, in particular to a sleep apnea syndrome identification device. Background technique [0002] Sleep apnea syndrome (Sleep apnea syndrome, SAS) is a relatively common and typical sleep-breathing disease. The affected people often experience symptoms such as snoring, breath holding, breathing stop, body twitching and even shock during nighttime sleep. Complete (apnea) or partial (hypopnea) obstruction of the upper airway during sleep that significantly affects the quality of sleep, resulting in daytime drowsiness, headaches, difficulty concentrating, and reduced learning efficiency , After the disease develops seriously, it can also cause various diseases such as high blood pressure and myocardial infarction. Studies have shown that the incidence of sleep disorders is about 20%-40%, among which SAS is the most common, and SAS affects about 2%-4% of human sleep quality, and the proportion of affected elderly p...

AI Technical Summary

Problems solved by technology

[0004]1. Polysomnography requires the use of more sensing equipment, which greatly interferes with the sleep of patients, and the instruments are relatively complicated. Currently, it is only suitable for hospitals or experiments room, and the detection cost is high, it is not suitable for normal use at home

[0005]2. The number of hospital beds that can apply polysomnography and the resources of professionally trained sleep testing technicians at home and abroad are limited. If ordinary patients want to use PSG equipment Equipment needs to wait in line, and this time is generally very long

Images