This study developed and validated a machine learning model to accurately predict acute kidney injury (AKI) after non-cardiac surgery, aiming to improve patient outcomes by assessing its clinical feasibility and generalizability. We conducted a retrospective cohort study using data from 76,032 adults who underwent non-cardiac surgery at a single tertiary medical center between March 2019 and February 2021, and used data from 5512 patients from the VitalDB open dataset for external model validation. The predictive variables for model training consisted of demographic, preoperative laboratory, and intraoperative data, including calculated statistical values such as the minimum, maximum, and mean intraoperative blood pressure. When predicting postoperative AKI, our gradient boosting machine model incorporating all the variables achieved the best results, with AUROC values of 0.868 and 0.757 for the internal and external validations using the VitalDB dataset, respectively. The model using intraoperative data performed best in internal validation, while the model with preoperative data excelled in external validation. In this study, we developed a predictive model for postoperative AKI in adult patients undergoing non-cardiac surgery using preoperative and intraoperative data, and external validation demonstrated the efficacy of open datasets for generalization in medical artificial modeling research.

Lee SW, et al. J Korean Med Sci. 2024 Jul;39:e241. https://doi.org/10.3346/jkms.2024.39.e241

Lee SW, et al. Yonsei Med J. 2026 Jan;67:e7. https://doi.org/10.3349/ymj.2025.0105

Background and Objective: Lung transplantation is the only life-extending therapy for end-stage pulmonary disease patients, but its risks necessitate an understanding of outcome predictors, with the frailty index and nutritional status being key assessment tools. This study aims to evaluate the relationship between preoperative frailty and nutritional indexes and the postoperative mortality rate in patients receiving lung transplants, and to determine which measure is a more potent predictor of outcomes. Materials and Methods: This study reviewed 185 adults who received lung transplants at a single medical center between January 2013 and May 2023. We primarily focused on postoperative 7-year overall survival. Other outcomes measured were short-term mortalities, acute rejection, kidney complications, infections, and re-transplantation. We compared the predictive abilities of preoperative nutritional and frailty indicators for survival using receiver operating characteristic curve analysis and identified factors affecting survival through regression analyses. Results: There were no significant differences in preoperative nutritional indicators between survivors and non-survivors. However, preoperative frailty indicators did differ significantly between these groups. Multivariate analysis revealed that the American Society of Anesthesiologists Class V, clinical frailty scale, and Charlson Comorbidity Index (CCI) were key predictors of 7-year overall survival. Of these, the CCI had the strongest predictive ability with an area under the curve of 0.755, followed by the modified frailty index at 0.731. Conclusions: Our study indicates that for critically ill patients undergoing lung transplantation, frailty indexes derived from preoperative patient history and functional autonomy are more effective in forecasting postoperative outcomes, including survival, than indexes related to preoperative nutritional status.