首页|Capital Medical University Reports Findings in Stroke (Predicting 3-month poor f unctional outcomes of acute ischemic stroke in youngpatients using machine lear ning)
Capital Medical University Reports Findings in Stroke (Predicting 3-month poor f unctional outcomes of acute ischemic stroke in youngpatients using machine lear ning)
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
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News – New research on Cerebrovascular Diseas es and Conditions - Stroke is the subject ofa report. According to news reporti ng from Beijing, People’s Republic of China, by NewsRx journalists,research sta ted, “Prediction of short-term outcomes in young patients with acute ischemic st roke (AIS)may assist in making therapy decisions. Machine learning (ML) is incr easingly used in healthcare due toits high accuracy.”The news correspondents obtained a quote from the research from Capital Medical University, “Thisstudy aims to use a ML-based predictive model for poor 3-month functional outcomes in young AIS patientsand to compare the predictive perform ance of ML models with the logistic regression model. We enrolledAIS patients a ged between 18 and 50 years from the Third Chinese National Stroke Registry (CNS R-III),collected between 2015 and 2018. A modified Rankin Scale (mRS) 3 was a p oor functional outcome at 3months. Four ML tree models were developed: The extr eme Gradient Boosting (XGBoost), Light GradientBoosted Machine (lightGBM), Rand om Forest (RF), and The Gradient Boosting Decision Trees (GBDT),compared with l ogistic regression. We assess the model performance based on both discrimination andcalibration. A total of 2268 young patients with a mean age of 44.3 ± 5.5 y ears were included. Amongthem, (9%) had poor functional outcomes. The mRS at admission, living alone conditions, and highNational Institutes of H ealth Stroke Scale (NIHSS) at discharge remained independent predictors of poor3-month outcomes. The best AUC in the test group was XGBoost (AUC = 0.801), foll owed by GBDT,RF, and lightGBM (AUCs of 0.795, 0, 794, and 0.792, respectively). The XGBoost, RF, and lightGBMmodels were significantly better than logistic re gression (P <0.05). ML outperformed logistic regression,w here XGBoost the boost was the best model for predicting poor functional outcome s in young AISpatients.”
BeijingPeople’s Republic of ChinaAsi aCerebrovascularDiseases and ConditionsCyborgsEmerging TechnologiesHeal th and MedicineMachine LearningStroke
NETL
NSTL
