Estimating daytime sleepiness with previous night electroencephalography, electrooculography, and electromyography spectrograms in patients with suspected sleep apnea using a convolutional neural network

Abstract

A common symptom of obstructive sleep apnea (OSA) is excessive daytime sleepiness (EDS). The gold standard test for EDS is the multiple sleep latency test (MSLT). However, due to its high cost, MSLT is not routinely conducted for OSA patients and EDS is instead evaluated using sleep questionnaires. This is problematic however, since sleep questionnaires are subjective and correlate poorly with the MSLT. Therefore, new objective tools are needed for reliable evaluation of EDS. The aim of this study was to test our hypothesis that EDS can be estimated with neural network analysis of previous night polysomnographic signals. We trained a convolutional neural network (CNN) classifier using electroencephalography, electrooculography, and chin electromyography signals from 2,014 patients with suspected OSA. The CNN was trained to classify the patients into four sleepiness categories based on their mean sleep latency (MSL); severe (MSL < 5min), moderate (5 ≤ MSL < 10), mild (10 ≤ MSL < 15), and normal (MSL ≥ 15). The CNN classified patients to the four sleepiness categories with an overall accuracy of 60.6% and Cohen’s kappa value of 0.464. In two-group classification scheme with sleepy (MSL < 10 min) and non-sleepy (MSL ≥ 10) patients, the CNN achieved an accuracy of 77.2%, with sensitivity of 76.5%, and specificity of 77.9%. Our results show that previous night’s polysomnographic signals can be used for objective estimation of EDS with at least moderate accuracy. Since the diagnosis of OSA is currently confirmed by polysomnography, the classifier could be used simultaneously to get an objective estimate of the daytime sleepiness with minimal extra workload.

Convolutional Neural Networks based OSA Event Prediction from ECG Scalograms and Spectrograms

“…Some researchers modified convolutional neural networks models, which are originally used for 2D classification tasks to do 1D signal classification. On the other hand, time-frequency spectrum representations like wavelet based scalograms [12][13][14] and Short Time Fourier transform based spectrograms [15][16][17][18][19] provide multidimensional analysis of biological signals and they are generated to form 2D images to feed the network. In the literature, many studies focus on ECG signals which are found to be significantly associated with respiratory events.…”

Achieving Better Understanding of Obstructive Sleep Apnea Treatment Effects on Cardiovascular Disease Outcomes through Machine Learning Approaches: A Narrative Review

“…Layering automated analysis of physiologic responses on top of existing automation of traditional scoring will allow for a deeper understanding of patient-level data and the identification of additional features that contribute to meaningful disease outcomes in future research. For example, neural networks can meaningfully predict patient-relevant outcomes, such as daytime sleepiness [ 56 ]. Going beyond OSA itself, the data contained within a PSG and processed via neural networks can predict mortality, with much of the risk attributable to sleep fragmentation [ 57 ].…”

The Sleep Revolution project: the concept and objectives

“…As sleep recordings are now digitised, there is no need to be restricted by practical constraints as it was in the 1970s when sleep recordings were paper based. To overcome these limitations, we have already started to develop ML-based methods to analyse sleep studies and estimate the severity of OSA and related daytime symptoms (Huttunen et al, 2021;Korkalainen et al, 2019;Nikkonen et al, 2019Nikkonen et al, , 2020Nikkonen et al, , 2021. Moreover, we will continue this work and produce fully-automated deep-learning-based analysis algorithms for PSGs and HSATs using state-of-the-art methods such as convolutional neural networks and recurrent neural networks.…”