Development of Fully Synthetic Medical Database Shuffling Method

The importance of having a comprehensive and accurate medical database cannot be overstated. Such databases are crucial for various purposes including research, analysis, and decision-making in the field of medicine. However, collecting real-world medical data can be challenging due to privacy concerns and limited access to sensitive information. To overcome these challenges, our proposed method suggests using statistical data as a basis for developing a synthetic medical database. By employing a special shuffle algorithm, it is aimed to modify and enhance the primary database until it reaches an acceptable level of quality. This algorithm ensures that the resulting dataset maintains its statistical properties while also preserving privacy and confidentiality. Moreover, evaluating the resulting dataset using a neural network adds another layer of validation to ensure its reliability and accuracy. Through the utilization of a proposed method, a robust database has been developed to predict the risk of developing type 2 diabetes five years ahead. With an accuracy rate as high as 94.45% during neural network training, this dataset holds immense promise for improving patient outcomes by enabling early intervention and prevention strategies. This research represents a significant step forward in addressing the global burden posed by type 2 diabetes.