A smart Alzheimer's patient monitoring system with IoT-assisted technology through enhanced deep learning approach

Earlier detection of Alzheimer's disease is more significant for improving the quality of the patient's life. This aspect may reduce the fatality rate among the population and also maximize the average life expectancy. Thus, this paper introduces a new Alzheimer's prediction model using IoT and deep structured architectures. A new smart Alzheimer's patient monitoring system is developed by processing healthcare data using IoT devices. Initially, Alzheimer's patients are detected from the set of patients using "enhanced deep residual network-long short-term memory (DRN-LSTM)." Here, the detection process is done with the data associated with the patients. The optimal feature selection phase and enhanced deep convolutional network (DCN) and deep residual network (DRN)-based detection are accomplished by parameter-improved horse herd optimization algorithm (PI-HHO). The monitored data involve audio, data, and video from the sensors based on the location and movements of patients. Next, the gathered data are forwarded to the optimal feature selection with the same algorithm and predicted the abnormalities through enhanced DNN + LSTM using PI-HHO. Thirdly, the abnormal patients are alerted to the nearby hospital for appropriate treatment and monitoring. All through the result evaluation, the accuracy and precision rate of the recommended Alzheimer's patient monitoring system attain 98% and 97%. Thus, this smart patient prediction model ensures the high-quality results in terms of standard performance metrics while evaluating with other algorithms.