A Reliable Model System Design to Predict Elderly People Falling Based on Drowsiness and Sleep Patterns

The Internet of Things aids in the monitoring of real-time sensing information via implanted or wearable sensors on the human body. A gateway will be connected to the internet and online data sensing will be possible thanks to the system it facilitates. Accidental or temporary loss of consciousness can result in a fall, and if help isn't given immediately, the results can be fatal for the elderly. Older people's mobility is limited since they need to be accompanied by a caretaker at all times out of concern that they will collapse. Those who cannot afford a career at home often have no choice but to go to institutional care. The need to seek assistance on an as-needed basis, rather than being constantly monitored, is the subject of this research. The sleep habits of the elderly are examined in this work since they are an important indicator of fatigue. Insufficient sleep is a common problem among the elderly. This factor allowed us to accurately forecast the elderly's physical condition and served as a useful piece of information in preventing elderly falls. As a result, we have applied a machine learning model to the problem of predicting the health of the elderly and are now able to do so. Model accuracy was confirmed by application to real-world data.