loading page

Heuristic Approaches for Outlier Reduction and Disease Prediction in Chronic Kidney Disease through IoT
  • Kalpana Murugan,
  • Radha M
Kalpana Murugan
Kalasalingam Academy of Research and Education (Deemed to be University)

Corresponding Author:[email protected]

Author Profile
Radha M
Kalasalingam Academy of Research and Education (Deemed to be University)
Author Profile

Abstract

The development of the Internet of Things (IoT), which enables communication between people, things, data, and virtual platforms in the environment, is a result of the exponential rise of information technology (IT). Recently, numerous decision support systems in the medical industry have been offered via IoT and cloud-based e-health services. Owing to the developments in IoT-enabled medical gadgets and sensing devices. Chronic diseases are often considered as a major source of concern and a threat to public health on a global scale. The kidneys are one of the body’s most complicated organs and perform several tasks. The elimination of waste materials by the kidneys during the formation of urine helps to cleanse the blood. When the kidneys begin to lose function, Chronic Kidney Disease (CKD) occurs. The accurate diagnosis of CKD is an important and crucial step in medical informatics. An earlier diagnosis can save more human lives. But, it is complicated task due to similar symptoms, inaccurate data and information, lack of knowledge, etc. It is also seen that the presence of missing values and outliers can also complicate the prediction task and, in turn, produce less accurate results. The primary objective of this research work is to develop an IoT framework for collecting the real time CKD dataset and also to handle the two well-known issues of medical data, i.e. outlier detection and CKD prediction. This work proposes the Chicken Swarm Optimization (CSO) algorithm to enhance the raw data and diagnose CKD diseases using k-NN.