Advanced Diabetes Diagnosis Using Machine Learning and Deep Learning Approaches

Diabetes is a chronic disease that weakens the body’s ability to regulate blood sugar (glucose) levels. Severe complications may occur if diabetes re-mains untreated or undiagnosed. The traditional method to identify diabetes is to visit a diagnostic center or consult a doctor in the field. However, with the advancements in machine learning (ML) and deep learning (DL) approaches, this critical problem can be addressed more efficiently. In the pre-sent work, we have designed an intelligent predictive model that enables the detection of diabetes in patients with high accuracy. For this purpose, we have used several machine learning algorithms in the first stage, including LR, LDA, KNN, CART, NB, and SVM. In the second stage, we have developed a multi-layer ANN model to improve detection accuracy. Finally, we established a comparison between ML and DL algorithms in terms of detection accuracy.