Wearable Heat Stroke Detection With Stages Using Tinyml And Iot

Abstract

Heat stroke, a potentially fatal medical emergency, is defined as a fast rise in body temperature caused by prolonged exposure to high temperatures or excessive physical exertion in hot surroundings. To avoid serious repercussions or deaths, early diagnosis and action are important. Using Tiny Machine Learning (TinyML) algorithms and Internet of Things (IoT) technologies, this study proposes a unique approach for real-time wearable heat stroke diagnosis. The suggested wearable device has many sensors for monitoring vital signs and environmental conditions. A heart rate monitor, skin temperature sensor, and ambient temperature and humidity sensors are among the sensors. The gadget is lightweight, low-power, and non-invasive, making it suited for continuous use during outdoor activities or work in hot environments. TinyML, which uses the power of machine learning models built for resource-constrained contexts, processes the data gathered by the wearable device locally. TinyML algorithms analyze sensor data quickly to identify early indicators of heat stroke, allowing the device to categorize the disease into stage 1, stage 2 and stage 3 depending on the severity of the heat stress. The wearable gadget is incorporated into an IoT ecosystem to improve the system's capabilities. The processed data is wirelessly transferred to a central server or cloud platform for additional analysis and storage. This allows healthcare experts, emergency responders, and concerned people to follow the wearer's health state in real time and remotely. The suggested system's performance is tested by thorough real-world testing in a variety of heat-intensive circumstances.