An efficient IoMT based health monitoring using complex valued deep CNN and political optimizer. Issue 12 (19th August 2022)

• Record Type: Journal Article
• Authors: Kavitha, D.; Vidhya, A.; Prema, V.; Priyadharshini, M.; Kumaresan, G.; Sangeetha, G.
• Is Part Of: Transactions on emerging telecommunications technologies. Volume 33:Issue 12(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-08-19
• Subjects: Telecommunication -- Periodicals; 384.05
• DOI: 10.1002/ett.4610
• Languages: English
• ISSNs: 2161-5748

Abstract

Abstract: In the field of prevalent computing, human living becomes smart with the recent developments on the IoMT, telecommunication methods and wearable sensors to provide smart healthcare systems. The IoMT plays a significant role in decreasing the mortality rate through the early recognition of disease. The prediction of the disease is the key problem in medical database analysis. In this paper, the novel smart healthcare structure for observing the physical actions of the matured people with the help of the IoMT and deep learning algorithms are utilized for quicker examination and improved treatment suggestions. The data is gathered from several wearable sensors positioned on the person's chest, right arm and left ankle is broadcasted via the IoMT gadgets to the merged cloud as well as the data analysis layer. The complex‐valued deep convolutional neural network with an improved political optimizer (CDCNNIP) approach perfectly suits the MapReduce model is employed for identifying the movement practiced by various body portions, which offer high flexibility and improved performance when compared to other classifiers. The accuracy of the complex‐valued deep Convolutional neural network is enhanced using the improved political optimizer (IPO). The proposed structure detects 12 physical actions and produces an accuracy of 98.5%. This highest‐value accuracy is considered as the best solution for the identification of physical actions hence observing the medical settings of the aged people. Abstract : Proposing the Deep learning techniques for monitoring the physical activities of the aged people. Utilizing the complex‐valued deep Convolutional neural network with Improved political optimizer (CDCNNIP) for the recognition of the motion practiced by the several body portions. To enhance the classification accuracy of the Cv‐DCNN by using an improved political optimizer (IPO) algorithm. Computing the performance of the proposed method by comparing it with various parameters such as region of convergence, precision, accuracy, and recall.