MyMedR

Displaying all 4 publications

Abstract:

Sort:

Fulltext A Deep Learning Model for Automated Sleep Stages Classification Using PSG Signals

Yildirim O, Baloglu UB, Acharya UR

Int J Environ Res Public Health, 2019 02 19;16(4).
PMID: 30791379 DOI: 10.3390/ijerph16040599

Sleep disorder is a symptom of many neurological diseases that may significantly affect the quality of daily life. Traditional methods are time-consuming and involve the manual scoring of polysomnogram (PSG) signals obtained in a laboratory environment. However, the automated monitoring of sleep stages can help detect neurological disorders accurately as well. In this study, a flexible deep learning model is proposed using raw PSG signals. A one-dimensional convolutional neural network (1D-CNN) is developed using electroencephalogram (EEG) and electrooculogram (EOG) signals for the classification of sleep stages. The performance of the system is evaluated using two public databases (sleep-edf and sleep-edfx). The developed model yielded the highest accuracies of 98.06%, 94.64%, 92.36%, 91.22%, and 91.00% for two to six sleep classes, respectively, using the sleep-edf database. Further, the proposed model obtained the highest accuracies of 97.62%, 94.34%, 92.33%, 90.98%, and 89.54%, respectively for the same two to six sleep classes using the sleep-edfx dataset. The developed deep learning model is ready for clinical usage, and can be tested with big PSG data.

Matched MeSH terms: Sleep Wake Disorders/physiopathology*
Fulltext SleepEEGNet: Automated sleep stage scoring with sequence to sequence deep learning approach

Mousavi S, Afghah F, Acharya UR

PLoS One, 2019;14(5):e0216456.
PMID: 31063501 DOI: 10.1371/journal.pone.0216456

Electroencephalogram (EEG) is a common base signal used to monitor brain activities and diagnose sleep disorders. Manual sleep stage scoring is a time-consuming task for sleep experts and is limited by inter-rater reliability. In this paper, we propose an automatic sleep stage annotation method called SleepEEGNet using a single-channel EEG signal. The SleepEEGNet is composed of deep convolutional neural networks (CNNs) to extract time-invariant features, frequency information, and a sequence to sequence model to capture the complex and long short-term context dependencies between sleep epochs and scores. In addition, to reduce the effect of the class imbalance problem presented in the available sleep datasets, we applied novel loss functions to have an equal misclassified error for each sleep stage while training the network. We evaluated the performance of the proposed method on different single-EEG channels (i.e., Fpz-Cz and Pz-Oz EEG channels) from the Physionet Sleep-EDF datasets published in 2013 and 2018. The evaluation results demonstrate that the proposed method achieved the best annotation performance compared to current literature, with an overall accuracy of 84.26%, a macro F1-score of 79.66% and κ = 0.79. Our developed model can be applied to other sleep EEG signals and aid the sleep specialists to arrive at an accurate diagnosis. The source code is available at https://github.com/SajadMo/SleepEEGNet.

Matched MeSH terms: Sleep Wake Disorders/physiopathology*
The social impact of oral and facial pain in an industrial population

Jaafar N, Razak IA, Zain RB

Ann Acad Med Singap, 1989 Sep;18(5):553-5.
PMID: 2619246

The aim of this study is to determine the social impact of oral and facial pain in a sample involving an industrial population. Out of a total of 355 subjects interviewed, nearly one-half claimed to have oral and facial pain in the previous one month prior to the survey. The most common type of pain was that related to hot or cold fluids or sweet things followed by toothache. On the average, the pain lasted for 4.2 days (SD = 4.9) per person in the past one-month. About one in five persons with pain reported that it was severe enough to disrupt sleep. About one in ten persons reporting pain had to take sick leave because of the pain. However, only one in four persons with pain consulted a doctor or dentist. More than one-third tried to cope with the pain and did nothing while the rest tried various means of self-cure. It is therefore postulated that in this population, pain per se is a poor predictor of utilisation of dental services. Further research into pain coping behaviour and how this affects of pattern of utilisation of dental services is indicated in order to formulate a strategy to encourage the habit of seeking prompt dental care by the target population.

Matched MeSH terms: Sleep Wake Disorders/physiopathology
Fulltext Serum potassium and handgrip strength as predictors of sleep quality among hemodialysis patients in Malaysia

Ling LL, Chan YM, Mat Daud Z'

Asia Pac J Clin Nutr, 2019;28(2):401-410.
PMID: 31192570 DOI: 10.6133/apjcn.201906_28(2).0023

BACKGROUND AND OBJECTIVES: Poor sleep quality is prevalent among hemodialysis (HD) patients and leads to adverse health outcomes. This study investigated the association of nutritional parameters with sleep quality among Malaysian HD patients.
METHODS AND STUDY DESIGN: A cross-sectional study was conducted among 184 Malaysian HD patients. Anthropometric measurements and handgrip strength (HGS) were obtained using standardized protocols. Relevant biochemical indicators were retrieved from patients' medical records. Nutritional status was assessed using the dialysis malnutrition score. The sleep quality of patients was determined using the Pittsburgh Sleep Quality Index questionnaire on both dialysis and non-dialysis days.
RESULTS: Slightly more than half of the HD patients were poor sleepers, with approximately two-third of them having a sleep duration of <7 hours per day. Sleep latency (1.5±1.2) had the highest sleep component score, whereas sleep medicine use (0.1±0.6) had the lowest score. Significantly longer sleep latency and shorter sleep duration were observed in the poor sleepers, regardless of whether it was a dialysis day or not (p<0.001). Poor sleep quality was associated with male sex, old age, small triceps skinfold, hypoproteinemia, hyperkalemia, hyperphosphatemia, and poorer nutritional status. In a multivariate analysis model, serum potassium (β=1.41, p=0.010), male sex (β=2.15, p=0.003), and HGS (β=-0.088, p=0.021) were found as independent predictors of sleep quality.
CONCLUSIONS: Poor sleep quality was evident among the HD patients in Malaysia. The sleep quality of the HD patients was associated with nutritional parameters. Routine assessment of sleep quality and nutritional parameters indicated that poor sleepers have a risk of malnutrition and may benefit from appropriate interventions.

Matched MeSH terms: Sleep Wake Disorders/physiopathology