The Journal of surgical research · 2025
Using the Geriatric Emergency Perioperative Risk Index Derived From Artificial Intelligence Algorithms to Predict Outcomes of Geriatric Emergency General Surgery.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Xu Dequan, Zhou Haoxin, Rong Jie, Xie Xin, Hou Limin
Аннотация
INTRODUCTION: The objective of this study was to employ artificial intelligence (AI) technology for the development of a model that can accurately forecast the outcome of emergency general surgery (EGS) in elderly patients. Additionally, an innovative visual scoring system called geriatric emergency perioperative risk index (GEPR) was devised based on this model.
METHODS: A retrospective database of geriatric patients who had undergone EGS was used for the development of the AI model and GEPR. The study employed a specialized algorithm, comprising of four sequential steps namely scale prototype selection, clinical data collection and collation, AI model development, and GEPR development.
RESULTS: In total, 1500 patients with the mean age of 69.8 ys were enrolled in the study. RandomForestClassifier algorithm outperformed the other AI models. Based on the feature importance, GEPR was derived, with a total score range of 0-26. The C-statistic of GEPR for in-hospital mortality was 0.872 (95% confidence interval, 0.840-0.905). The observed probability of in-hospital mortality gradually increased from 0% at a score of 0 to 63.3% at a score of 12 and 100% at a score of 15.
CONCLUSIONS: Using patient-related and technical parameters, a GEPR model derived from AI algorithms for prediction of surgical complications in geriatric EGS was developed. The GEPR model reliably predicts postoperative in-hospital mortality in geriatric EGS patients. Clinical studies are currently being conducted to validate the stability and precision of the GEPR model utilizing the MIMIC-IV database. Further prospective multicenter trials are needed to externally validate the developed model.
METHODS: A retrospective database of geriatric patients who had undergone EGS was used for the development of the AI model and GEPR. The study employed a specialized algorithm, comprising of four sequential steps namely scale prototype selection, clinical data collection and collation, AI model development, and GEPR development.
RESULTS: In total, 1500 patients with the mean age of 69.8 ys were enrolled in the study. RandomForestClassifier algorithm outperformed the other AI models. Based on the feature importance, GEPR was derived, with a total score range of 0-26. The C-statistic of GEPR for in-hospital mortality was 0.872 (95% confidence interval, 0.840-0.905). The observed probability of in-hospital mortality gradually increased from 0% at a score of 0 to 63.3% at a score of 12 and 100% at a score of 15.
CONCLUSIONS: Using patient-related and technical parameters, a GEPR model derived from AI algorithms for prediction of surgical complications in geriatric EGS was developed. The GEPR model reliably predicts postoperative in-hospital mortality in geriatric EGS patients. Clinical studies are currently being conducted to validate the stability and precision of the GEPR model utilizing the MIMIC-IV database. Further prospective multicenter trials are needed to externally validate the developed model.