Muhammad Faisal, Unggul Pundjung Juswono, Didik Rahadi Santoso, Chomsin Sulistya Widodo, Saihan Nabawi, Muhammad Masdar Mahasin
Diabetes mellitus is an urgent challenge for global health. According to data from the International Diabetes Federation (IDF), the prevalence of diabetes has increased significantly in the last 5 years and is predicted to reach 700 million cases by 2045. Recently, a diabetes screening method based on the electrical properties of cells using the Electrical Impedance Spectroscopy (EIS) technique was proposed. In previous studies, the EIS technique was only used to identify cell and tissue damage but was not yet able to classify a disease such as diabetes. This study aims to develop the EIS technique so that it can be used to classify diabetes mellitus using machine learning with the backpropagation method. The data source was obtained from direct measurements in the laboratory using 90 mice (Mus musculus) that were made to suffer from diabetes mellitus. Mice were confirmed to have diabetes mellitus through a fasting blood sugar test (FBST) as a reference, then their cell electrical properties were measured using EIS. The measurement data will be made into a dataset for a machine learning model with the backpropagation training method consisting of input layers, hidden layers, and output layers. Paired data parameters used as input and output layers are frequency, phase, and impedance with blood glucose levels. The results of making a machine learning model to develop the EIS technique in classifying diabetes mellitus produced a fairly good performance index with an accuracy of 98.39 %, precision of 99 %, recall of 97 %, F1-score of 98 %, and specificity of 99 %. EIS development with the help of machine learning that utilizes the backpropagation method can be used to classify diabetes mellitus. The results of developing EIS using machine learning can be used to classify diabetes mellitus independently in a short time but have high accuracy. © 2025, Walailak University. All rights reserved.
Department of Physic, Brawijaya University, Jawa Timur, 65145, Indonesia; Department of Informatics Engineering, Brawijaya University, Jawa Timur, 65145, Indonesia
