USING DEEP LEARNING TECHNOLOGY FOR HEALTHCARE APPLICATIONS IN INTERNET OF THINGS SENSOR MONITORING SYSTEM

Abstract

A healthcare management platform for a personal wearable device can be set up in an app on a mobile phone or on a platform in a smart classroom. However, a large number of Internet of Things (IoT) devices being simultaneously connected to a central management server may negatively affect the server. Combining an IoT platform with deep learning technology facilitates monitoring the state of the input device, allowing for corresponding adjustments. The operation consists of collecting sensor data, optimizing the dataset using deep learning models, and integrating the deep learning model with the IoT platform. To optimize the data, noise data from human factors must be eliminated. Through a re-evaluation of the deep learning results, data with known errors are removed and data with higher accuracy are added. The optimized dataset can be used to construct a higher accuracy model and integrate it into the IoT platform. The IoT platform can analyze outputs through the deep learning model. The results show that the current status of each device can be identified. This is helpful for the user of a healthcare applications system.