Hyperparameter optimization based deep learning model for medical data classification in internet of things enabled cloud environment

RAGUPATHI T; GOVINDARAJAN M; PRIYARADHIKADEVI T

No Detail Found!

Hyperparameter optimization based deep learning model for medical data classification in internet of things enabled cloud environment

Author(s)

RAGUPATHI T Department of Computer Science and Engineering, Annamalai University, Annamalai Nagar, 608002, Chidambaram, Tamil Nadu, India

GOVINDARAJAN M Department of Computer Science and Engineering, Annamalai University, Annamalai Nagar, 608002, Chidambaram, Tamil Nadu, India

PRIYARADHIKADEVI T Department of Computer Science and Engineering, Mailam Engineering College, Mailam, Tamil Nadu, 604304, India

Abstract

Recently, the integration of internet of things (IoT) and advanced medicinal sensors can be contributed to improving the quality of healthcare services. In this way, the cloud and IoT technologies are completely utilized to design intelligent healthcare systems which can support real time applications by the use of artificial intelligence techniques. The recently developed deep learning (DL) approaches pave the way to design effective medical data classification models to diagnose diseases. With this motivation, this paper presents a new hyper parameter optimized DL technique for medical data classification in IoT enabled cloud environment. The proposed model enables the IoT devices to collect healthcare data and diagnose diseases in the cloud server. Primarily, the IoT devices are used for data acquisition and data preprocessing takes place to enhance the data quality. In addition, a convolutional neural network-long short term memory (CNN-BILSTM) method is employed for classification purposes, which identifies the presence of diseases or not. For boosting the classification performance of the CNN-BILSTM model, a black widow optimization (BWO) technique is applied to determine the optimal learning rate of the CNN-BILSTM model. A wide range of simulations take place on three benchmark medical datasets and the experimental results highlighted the promising efficiency of the proposed method over the other techniques.

Publication Details

Page(s) 5960-5972

DOI DOI not available

Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 20, Year: 2022

Request the full text from PASTIC

Keywords

Healthcare, Cloud Computing, IoT, Deep learning, Learning Rate, Remote diagnosis, Data classification

Citations

Downloads

Views