Machine Learning Approaches to Predict Patient’s Length of Stay in Emergency Department

Applied Computational Intelligence and Soft Computing, Journal Year: 2023, Volume and Issue: 2023, P. 1 - 13

Published: Oct. 27, 2023

As the COVID-19 pandemic has affected globe, health systems worldwide have also been significantly affected. This impacted many sectors, including in Kingdom of Jordan. Crises that put heavy pressure on systems’ shoulders include emergency departments (ED), most demanded hospital resources during normal conditions, and critical crises. However, managing efficiently achieving best planning allocation their EDs’ becomes crucial to improve capabilities accommodate crisis’s impact. Knowing factors affecting patient length stay prediction is reducing risks prolonged waiting clustering inside EDs. That is, by focusing these analyzing effect each. research aims determine predict outcome: stay, i.e., predictor variables. Therefore, patients’ EDs across time duration categorized as (low, medium, high) using supervised machine learning (ML) approaches. Unsupervised algorithms applied classify patient’s local The Arab Medical Centre Hospital selected a case study justify performance proposed ML model. Data spans interval 22 months, covering period before after COVID-19, used train feedforward network. model compared with other approaches its superiority. Also, comparative correlation analyses are conducted considered attributes (inputs) help LOS ED. be trees such decision stump, REB tree, Random Forest multilayer perceptron (with batch sizes 50 0.001 rate) for this specific problem. Results showed better terms accuracy easiness implementation.

Language: Английский

AGO-FT: An adaptive guided oversampling based on fast space division and trustworthy sampling space for imbalanced noisy datasets

2021 IEEE International Conference on Big Data (Big Data), Journal Year: 2024, Volume and Issue: unknown, P. 529 - 538

Published: Dec. 15, 2024

Language: Английский

Prediction of Insufficient Accuracy for Patients Length of Stay using Deep Belief Network

2022 6th International Conference on Computing Methodologies and Communication (ICCMC), Journal Year: 2023, Volume and Issue: unknown, P. 211 - 216

Published: Feb. 23, 2023

The research study is to predict the patient's length of stay at healthcare centers and analyze their patient discharge by admission using machine learning algorithms achieve higher accuracy also determine data unplanned readmission in Intensive Care Unit (ICU). framework classify Support Vector Machine (SVM) Logistic Regression (LR) perform all measures. This used Novel Deep Belief Network Convolutional Neural operations give best exactness centers. Their patients' health investigations were gathered from numerous web sources with current results threshold 0.05%, confidence interval 95% mean, standard deviation for study, which 47 samples two groups a G-power 80%. To analysis, algorithm has found 90.25% accuracy, therefore this needs find better prediction learning. 87.11% analysis significant value tailed tests 0.045 (p<0.05) interval. concludes that significantly than algorithm.

Language: Английский