<![CDATA[Non-communicable diseases (NCDs) present a considerable worldwide health dilemma, resulting in considerable expenses for treatment and heightened rates of mortality. Conditions like diabetes mellitus, cardiovascular diseases, cancer, and chronic respiratory diseases are primary causes of global mortality, making up 71% of total global deaths in 2016, as reported by the World Health Organization (WHO). Diabetes Mellitus (DM), marked by prolonged elevated blood glucose levels, stands out as a significant metabolic disorder. This research delves into the implementation of Recurrent Neural Networks (RNNs) utilizing the Adaptive Moment Estimation (Adam) optimizer for classifying Diabetes Mellitus (DM). RNNs, a subset of artificial neural networks tailored for sequential data processing, are employed to make predictions by incorporating recurrent connections. Situated within the dynamic landscape of Artificial Intelligence and Machine Learning, the research exhibits promising outcomes via k-fold cross-validation, confusion matrix analysis, loss graph examination, and classification report. The RNN-Adam model showcases commendable overall performance, achieving an average accuracy of 80.20% through k-fold cross-validation and 81.60% accuracy as revealed by the confusion matrix. This research offers valuable insights into the effectiveness of the RNN-Adam model for diabetes classification.]]>
Copyrights © 2024
