Advances in Fuzzy-Based Internet of Medical Things (IoMT)

11Implementation of a Neuro-Fuzzy-Based Classifier for the Detection of Types 1 and 2 Diabetes

Chamandeep Kaur¹, Mohammed Saleh Al Ansari², Vijay Kumar Dwivedi³ and D. Suganthi⁴*

¹Department of Computer Science and IT, Jazan University, Jazan, Saudi Arabia

²College of Engineering, Department of Chemical Engineering, University of Bahrain, Sakhir Bahrain

³Department of Mathematics, Vishwavidyalaya Engineering College, Ambikapur, Surguja, India

⁴Department of Computer Science, Saveetha College of Liberal Arts and Sciences, SIMATS, Thandalam, Chennai, India

Abstract

In 2019, the death rate increased above 1.5 million throughout the globe due to diabetes, according to data published by the Globe Health Organization, and it affected about 450 million people worldwide. It has been noted that many people with diabetes failed to recognize their condition in its earliest stages, leading to a rise in the commonness of type 2 diabetes. To prevent this from happening, a novel neural classifier based on fuzzy logic for diagnosing diabetes at an early stage. We add some unknown neuro-fuzzy rules with temporal bounds for preliminary sorting. Fuzzy cognitive maps (FCMs) with time intervals improve these levels before making a final categorization judgment. The suggested model’s primary focus is on time-based detection of diabetes severity. In addition, the decision-making procedure in diabetes forecasting uses a set of neuro-fuzzy criteria for picking the most relevant variables. Trials done ...