首页|Affiliated Hospital of Qingdao University Reports Findings in Acute Kidney Injur y (Personalized Prediction of Long-Term Renal Function Prognosis Following Nephr ectomy Using Interpretable Machine Learning Algorithms: Case-Control Study)
Affiliated Hospital of Qingdao University Reports Findings in Acute Kidney Injur y (Personalized Prediction of Long-Term Renal Function Prognosis Following Nephr ectomy Using Interpretable Machine Learning Algorithms: Case-Control Study)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
2024 OCT 03 (NewsRx)-By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-New research on Kidney Diseases and Conditions - Acute Kidney Injury is the subject of a report. According to news reporting out of Qingdao, People's Republic of China, by NewsRx editors, research stated, "Acu te kidney injury (AKI) is a common adverse outcome following nephrectomy. The pr ogression from AKI to acute kidney disease (AKD) and subsequently to chronic kid ney disease (CKD) remains a concern; yet, the predictive mechanisms for these tr ansitions are not fully understood." Our news journalists obtained a quote from the research from the Affiliated Hosp ital of Qingdao University, "Interpretable machine learning (ML) models offer in sights into how clinical features influence long-term renal function outcomes af ter nephrectomy, providing a more precise framework for identifying patients at risk and supporting improved clinical decision-making processes. This study aime d to (1) evaluate postnephrectomy rates of AKI, AKD, and CKD, analyzing long-ter m renal outcomes along different trajectories; (2) interpret AKD and CKD models using Shapley Additive Explanations values and Local Interpretable Model-Agnosti c Explanations algorithm; and (3) develop a web-based tool for estimating AKD or CKD risk after nephrectomy. We conducted a retrospective cohort study involving patients who underwent nephrectomy between July 2012 and June 2019. Patient dat a were randomly split into training, validation, and test sets, maintaining a ra tio of 76.5:8.5:15. Eight ML algorithms were used to construct predictive models for postoperative AKD and CKD. The performance of the best-performing models wa s assessed using various metrics. We used various Shapley Additive Explanations plots and Local Interpretable Model-Agnostic Explanations bar plots to interpret the model and generated directed acyclic graphs to explore the potential causal relationships between features. Additionally, we developed a web-based predicti on tool using the top 10 features for AKD prediction and the top 5 features for CKD prediction. The study cohort comprised 1559 patients. Incidence rates for AK I, AKD, and CKD were 21.7 % (n=330), 15.3% (n=238), a nd 10.6% (n=165), respectively. Among the evaluated ML models, the Light Gradient-Boosting Machine (LightGBM) model demonstrated superior performa nce, with an area under the receiver operating characteristic curve of 0.97 for AKD prediction and 0.96 for CKD prediction. Performance metrics and plots highli ghted the model's competence in discrimination, calibration, and clinical applic ability. Operative duration, hemoglobin, blood loss, urine protein, and hematocr it were identified as the top 5 features associated with predicted AKD. Baseline estimated glomerular filtration rate, pathology, trajectories of renal function , age, and total bilirubin were the top 5 features associated with predicted CKD . Additionally, we developed a web application using the LightGBM model to estim ate AKD and CKD risks. An interpretable ML model effectively elucidated its deci sion-making process in identifying patients at risk of AKD and CKD following nep hrectomy by enumerating critical features."
QingdaoPeople's Republic of ChinaAsi aAcute Kidney InjuryAlgorithmsClinical ResearchClinical Trials and Studi esCyborgsEmerging TechnologiesGastroenterologyHealth and MedicineKidne yKidney Diseases and ConditionsKidney FunctionMachine LearningNephrectom yNephrologyRenal FunctionRisk and PreventionSurgery
NETL
NSTL
