A Deep Learning Model for Classification of Uncontrolled Diabetics Based on Microwave Resonator Measurements of Urine

In this paper, an analysis of human urine using a microwave resonator was conducted and a deep learning model implemented to distinguish between uncontrolled diabetics and controlled diabetics was developed. It was necessary to collect urine samples from diabetic patients with poorly controlled diabetes, normal individuals, and diabetic patients with well controlled diabetes for the study. According to our findings, deep learning methods are more accurate than peak analysis techniques in classifying controlled diabetics in complex compounds measured by microwave resonators. This innovative approach has the potential to transform early disease detection and diagnosis, leading to more effective treatments and improved patient outcomes. Furthermore, the novel analytical method presented in our paper has broader applications, including the classification of intricate chemical components in foods and natural products.