PROJECT TITLE :

Predicting Sleep Quality in Osteoporosis Patients Using Electronic Health Records and Heart Rate Variability

ABSTRACT:

One of the most well-known aspects in everyday work performance is sleep quality (SQ). Polysomnography (PSG) is a method of analyzing sleep by attaching electrodes to the bodies of participants, which is potentially sleep-destructive. As a result, a trendy issue right now is looking into SQ utilizing a more user-friendly and cost-effective manner. To prevent overfitting problems, one possible way for forecasting SQ is to reduce the amount of signals used. We present three techniques based on electronic health records and heart rate variability in this research (HRV). Several tests were conducted utilizing the Osteoporotic Fractures in Men (MrOS) sleep dataset to evaluate the performance of the suggested techniques. The results show that utilizing only ECG signals recorded during PSG, a deep neural network algorithm can predict light, medium, and deep SQ with an accuracy of 0.6. This result suggests that HRV properties, which are easily quantifiable by simple and inexpensive wearable equipment, can be used to predict SQ.