Fig. 1

Overall framework of the proposed work. (1) Correlation analysis was conducted to find out the statistically significant factors to the GSD.(2)—(5) Incorporate the selected factors into machine learning to build the GSD model: The weight of each sample is adjusted according to the NOR, and a new research cohort that reflects the weight of sample is constructed; Based on machine learning method, the ANN-based GSD model and SVR-based GSD model are built; The prediction error and regression coefficient of the two models are compared, and the one, i.e., the ANN-based GSD model, that has better modeling performance is selected. (6) Based on the prediction results of the GSD model, the influencing factor is quantitatively and accurately evaluated. (7) 81 new patients were involved in the clinical application to verify the effectiveness of the proposed model