Fog-based framework for diabetes prediction using hybrid ANFIS model in cloud environment

• Published in: Personal and ubiquitous computing Vol. 27; no. 3; pp. 909 - 916
• Main Authors: Kumar, Dipesh, Mandal, Nirupama, Kumar, Yugal
• Format: Journal Article
• Language: English
• Published: London Springer London 01.06.2023; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Cloud computing; Communication; Computer Science; Data acquisition; Data collection; Diabetes; Edge computing; Health care; Medical personnel; Mobile Computing; Original Paper; Personal Computing; Prediction models; Professionals; Remote sensors; User Interfaces and Human Computer Interaction; Fog computing; Internet of Things (IoT); PSO (particle swarm optimization); ANFIS (adaptive neuro-fuzzy inference system); Diabetes; Remote healthcare; WOA (whale optimization algorithm)
• ISSN: 1617-4909; 1617-4917
• DOI: 10.1007/s00779-022-01678-w

With the present situation in the world, proper monitoring of health has become one of the important and critical parts of our day-to-day life. With increasing work pressure and improper eating habit in day-to-day life, diabetes has become a common disease among the people. The early prediction of diabetes is a major tool to diagnose diabetes in its early stage. With the development of cloud computing and fog computing, a nature-based algorithm can be used to create a framework which can predict the diabetes in its early phase. The early prediction of disease will allow immediate treatment and hence patient’s life can be saved. Cloud computing and fog computing have provided mobility and has facilitated that the patients’ data can be collected and is monitored properly by the healthcare professionals at any time. In this paper, a fog-based diabetes prediction model is proposed where the patient data are acquired from the sensors that are present in remote. Fog computing is used for gathering and processing the data at the device end and communicating immediately. The processed data is transferred to the cloud layer for the analysis. In the cloud layer, the proposed hybrid ANFIS-PSO-WOA algorithm is used for the detection of diabetes level. The collected data is stored in fog layer and is used for further analysis by the healthcare professionals. Based on the analysis, the healthcare professional will treat the patient. The nature-based algorithm helps to provide good accuracy. The proposed framework has experimented with diabetes data obtained from the UCI repository and the result obtained with high accuracy of 92% by comparing with other existing algorithms such as SVM, ANN, and ANFIS. The proposed algorithm-based framework can be used to detect diabetes more efficiently and accurately as compared to other nature-based frameworks and will help healthcare professionals to diagnose the disease in its early phase.