Development and comparative evaluation of machine learning algorithms and Cox regression for predicting fasting plasma glucose-defined incident prediabetes: a longitudinal cohort study
Su Hu, Changshun Yan, Xinyuan Tian, Menghan Xu, Bo Cheng, Tong Xu, Guiqiu Cao · 2026
Read Development and comparative evaluation of machine learning algorithms and Cox regression for predicting fasting plasma glucose-defined incident prediabetes: a longitudinal cohort study by Su Hu, Changshun Yan, Xinyuan Tian, Menghan Xu, Bo Cheng, Tong Xu, Guiqiu Cao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13645560.txt.
Description
Su Hu and six colleagues build and test prediction models for prediabetes, the stage where fasting blood sugar is rising but diabetes has not yet begun. Their work rests on a large public health-check cohort from China, plus a separate hospital cohort for outside testing. They compared six machine learning models with standard survival models called Cox regression. Candidate measures were first narrowed with a selection method that drops weak predictors.
Among 10,592 adults, about one in eight developed prediabetes over roughly three years of follow-up. A random forest model looked best in training but faded on new data, while a Cox model held up more evenly. The final model kept age, body mass index, blood pressure, fasting sugar, and family history of diabetes. The team also built an online calculator so clinics can try the model in practice.
Catalog record
| CREATOR | Su Hu, Changshun Yan, Xinyuan Tian, Menghan Xu, Bo Cheng, Tong Xu, Guiqiu Cao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |