Machine-learning prediction of impaired outcome in diabetic patients undergoing non-cardiac surgery
Xiaojun Liu, Xueqing Chen, Lin Liu, Yuanyuan Lv · 2026
Read Machine-learning prediction of impaired outcome in diabetic patients undergoing non-cardiac surgery by Xiaojun Liu, Xueqing Chen, Lin Liu, Yuanyuan Lv (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/pmc-13279090.txt.
Description
Xiaojun Liu and colleagues published a 2026 study building machine-learning models to predict poor outcomes in diabetic patients having surgery outside the heart. Using a de-identified hospital dataset, they analyzed 4,293 diabetic patients, of whom 1,117 met the dataset's definition of an impaired outcome combining heart and brain complications. Five methods were compared, including penalized logistic regression and several boosting methods, and software showed which factors drove each prediction.
The AdaBoost model performed best, reaching an AUC of 0.82 on internal testing, a fair level of accuracy for risk around surgery. Prior stroke, prior heart attack, kidney function, albumin, inflammation markers, age, and blood-pressure patterns during surgery ranked among the top predictors. The authors read the impaired outcomes as the product of combined vascular disease, weak nutrition and kidney reserve, and unstable blood pressure during operations. They call for testing in other hospitals before clinical use.
Catalog record
| CREATOR | Xiaojun Liu, Xueqing Chen, Lin Liu, Yuanyuan Lv |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |