A patient-centric decision guidance system for detecting glycemia of diabetes patients

We propose the development of a patient-centric decision guidance system for detecting glycemia of diabetes patients. The proposed approach combines the strengths of both domain-knowledge-based and machine-learning-based approaches to learn the detection parameters. The contributions of this work are four-fold: 1) develop the new extended glycemic expert query parametric estimation (G-EQPE) model and the sequential-parallel-modularised (SPM) architecture to learn the optimal hypo- and hyperglycemic detection parameters simultaneously with a lower computational cost; 2) provide the user-friendly GUI tools and the well-defined JavaScript object notation (JSON) schemas for healthcare system specialists to maintain the system operations; 3) enable the specialists to construct the customised JSONiq queries for data manipulation, data monitoring, parameter learning, and report generation; 4) conduct the experimental study and present the superior detection results in terms of accuracy, sensitivity, and specificity by using the learned detection parameters (i.e., 99 mg/dL and 172 mg/dL).