The potential value of low-level serum interleukin-38 for the clinical diagnosis and risk prediction of hyperuricemia

The development and validation of a non-invasive prediction model of hyperuricemia based on modifiable risk factors: baseline findings of a health examination population cohort

“…9 There have been some prediction models for HUA, for example, Zheng et al 10 established a prediction model based on three methods: logistic, convolutional neural network (CNN) and XGBoost, and found that XGBoost had the best performance in HUA prediction; using RF algorithm, Gao et al developed a prediction model among a Chinese health examination population; 11 and Huang et al explored the potential valuable biomarker in predicting HUA using the logistic model. 12 In these models, one thing in common is that the predictors included in the models all involved biochemical indices, which required individuals to undergo an invasive examination. Although including more predictors in a prediction model can improve the performance, its feasibility in community-based high-risk population screening will be limited.…”