Enhancing Gestational Diabetes Prediction Through Deep Learning and Traditional ML Techniques

Timely and accurate forecasting is essential for making well-informed decisions and taking proactive measures of Gestational Diabetes Mellitus (GDM) which is essential for interventions and better maternal and fetal health outcomes. This study investigates the effectiveness of deep learning and traditional machine learning methods in enhancing the accuracy of gestational diabetes mellitus prediction. Using a comprehensive data-set that demographic details, medical history, Glucose levels etc.,. The study developed and evaluated the performance of a proposed Gestational Diabetic Deep Learning Classifier, along with two base classifiers in machine learning like Decision Tree and Support Vector Machines. These models were thoroughly trained, and their performance was evaluated using a number of measures, like accuracy, precision, recall, and F-Score. The findings demonstrate that the proposed Gestational Diabetic Deep Learning Classifier exhibited superior performance compared to conventional methods, achieving an accuracy rate of 95% relative to the baseline classifiers. Furthermore, this study highlights the potential of deep learning techniques in enhancing the prediction of gestational diabetes mellitus, and provides valuable insights to guide the development of more effective screening approaches aimed at improving maternal and fetal health outcomes.