Development of an IoT-based sleep pattern monitoring system for sleep disorder detection
Inadequate sleep can cause various health problems including heart disease and obesity. In this work, a sleep monitoring system that monitors human sleep patterns is developed using the internet of things (IoT) and Raspberry Pi. The system is designed to record any detected movements and process th...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
Institute Of Advanced Engineering And Science (IAES)
2024
|
| Online Access: | http://eprints.utem.edu.my/id/eprint/28714/2/01779101220241255251348.pdf http://eprints.utem.edu.my/id/eprint/28714/ https://ijeecs.iaescore.com/index.php/IJEECS/article/view/38035 http://doi.org/10.11591/ijeecs.v36.i2.pp777-784 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Inadequate sleep can cause various health problems including heart disease and obesity. In this work, a sleep monitoring system that monitors human sleep patterns is developed using the internet of things (IoT) and Raspberry
Pi. The system is designed to record any detected movements and process the data using machine learning to provide valuable insight into a person’s sleep patterns including sleep duration, the time taken to fall asleep, and the frequency of waking up. This information is very useful to provide the sleep disorder diagnostics of an individual including restless leg, parasomnia and insomnia syndrome besides giving recommendations to improve their sleep
quality. Also, the system allows the processed data to be stored in the cloud database which can be accessed through a mobile application or web interface. The performance of the system is evaluated in terms of its accuracy and reliability in detecting sleep order diagnostics. Based on the confusion matrix, the results show the accuracy of the system is 90.32%, 91.80%, and 91.80% in detecting the restless leg, parasomnia and insomnia syndrome, respectively. Meanwhile, the system showed high reliability in monitoring 10 participants for 8 hours and updated the recorded data and its analysis in the cloud. |
|---|