首页|Researchers from University of Bologna Describe Findings in Machine Learning (Co mparison of nine machine learning regression models in predicting hospital lengt h of stay for patients admitted to a general medicine department)
Researchers from University of Bologna Describe Findings in Machine Learning (Co mparison of nine machine learning regression models in predicting hospital lengt h of stay for patients admitted to a general medicine department)
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 Daily News Daily News – New research on artificial intelligenc e is the subject of a new report. According to news originating from Bologna, It aly, by NewsRx correspondents, research stated, “The General Medicine (GM) depar tment has the highest patient volume and heterogeneity among other hospital spec ialties. Closely examining hospitalization data is crucial because patients come with various conditions or traits. Length of stay (LoS) in hospitals is often u sed as an efficiency indicator.” Our news correspondents obtained a quote from the research from University of Bo logna: “It is influenced by various factors, including the patient’s medical bac kground, demographics, and type of diseases/signs/symptoms at the triage. LoS is a variable that can vary widely, making it difficult to estimate it promptly an d accurately, but doing so is highly beneficial. Moreover, efficiently grouping and managing patients based on their expected LoS remains a significant challeng e for healthcare organizations. This study aimed to compare the predictive abili ty of nine Machine Learning (ML) regression models in estimating the actual numb er of LoS days using demographics and clinical information recorded at admission as independent variables. We analyzed data collected on patients hospitalized a t the GM department of the Sant’Orsola-Malpighi University Hospital in Bologna, Italy, who were admitted through the Emergency Department. The data were collect ed from January 1, 2022, to October 26, 2022. Nine ML regression models were use d to predict LoS by analyzing historical data and patient information. The model s’ performance was assessed through root mean squared prediction error (RMSPE) a nd mean absolute prediction error (MAPE). Moreover, we used K-means clustering t o group patients’ medical and organizational criticalities (such as diseases, si gns, symptoms, and administrative problems) into four clusters. Feature Importan ce plots and SHAP (SHapley Additive exPlanations) values were employed to identi fy the more essential features and enhance the interpretability of the results. We analyzed the LoS of 3757 eligible patients, which showed an average of 13 day s and a standard deviation of 11.8 days. We randomly divided patients into a tra ining cohort of 2630 (70 %) and a test cohort of 1127 (30 % ). The predictive performance of the different models was between 11.00 and 16.1 6 days for RMSPE and between 7.52 and 10.78 days for MAPE. The eXtreme Gradient Boosting Regression (XGBR) model had the lowest prediction error, both in terms of RMSPE (11.00 days) and MAE (7.52 days). Sex, arrival via own vehicle/walk-in, ambulance arrival, light blue risk category, age 70 or older, and orange risk c ategory are some of the top features.”
University of BolognaBolognaItalyE uropeCyborgsEmerging TechnologiesHospitalsMachine Learning
NETL
NSTL
