Design of biosensor for synchronized identification of diabetes using deep learning
Ammar Armghan, J. Logeshwaran, S.M. Sutharshan, Khaled Aliqab, Meshari Alsharari, Shobhit K. Patel
Abstract
A highly sensitive biosensor for the synchronized identification of diabetes is designed using a deep learning approach. The biosensor is built on a cutting-edge platform that combines nanotechnology with electrochemical processes. This platform makes it possible to diagnose diabetes accurately and quickly by detecting both diabetes in a single test. The novel platform creates a biosensor that is very sensitive and selective for detecting diabetes by combining nanotechnology and electrochemical techniques. With high precision, this biosensor can identify the presence of glucose molecules in the blood. Also, it can differentiate between diabetes, which is crucial for a precise diagnosis of diabetes. The biosensor is far more sensitive than other techniques now in use, and it can detect glucose levels at a range of 0.5–5 mmol per liter. This study uses a medical deep-learning model to identify diabetes using a highly sensitive biosensor. A significant development in diabetes identification is the creation of this highly sensitive biosensor. It will aid in enhancing the precision and effectiveness of diabetes diagnosis, enabling improved management of the condition. Also, it will lower the cost of diabetes diagnosis, increasing accessibility for individuals who require it. The diagnosis and management of diabetes might be revolutionized by this new platform, making it simpler and more efficient. In an evaluation point, the proposed MDML achieved 96.21% accuracy, 91.53% precision, 94.21% recall, 97.98% f1-score and 89.90% diagnostic odds ratio.