Analysis on Benefits and Costs of Machine Learning-Based Early Hospitalization Prediction

Kim, Eunbi; Han, Kap Su; Cheong, Taesu; Lee, Sung Woo; Eun, Joonyup; Kim, Su Jin

doi:10.1109/access.2022.3160742

Cited by 3 publications

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pak et al reduced the number of patients with

greater than 30 min by more than 42% with the predictive data using the quantile regression ML model [ 49 ]. A study using Logistic Regression, Extreme Gradient Boosting, Natural Gradient Boosting, SVM, and Decision Tree ML models reduced LOS by 12.3 min (

reduction was achieved indirectly) from a general hospital’s emergency department in South Korea [ 50 ]. In terms of estimating

, ML models are often used to estimate the number of patients for a disease type.…”

Section: Discussionmentioning

confidence: 99%

Integration of Machine Learning Algorithms and Discrete-Event Simulation for the Cost of Healthcare Resources

2022

View full text Add to dashboard Cite

A healthcare resource allocation generally plays a vital role in the number of patients treated (pnt) and the patient waiting time (wt) in healthcare institutions. This study aimed to estimate pnt and wt as output variables by considering the number of healthcare resources employed and analyze the cost of health resources to the hospital depending on the cost coefficient (δi) in an emergency department (ED). The integration of the discrete-event simulation (DES) model and machine learning (ML) algorithms, namely random forest (RF), gradient boosting (GB), and AdaBoost (AB), was used to calculate the estimation of the output variables depending on the δi of resources cost. The AB algorithm performed best in almost all scenarios based on the results of the analysis. According to the AB algorithm based on the δ0.0, δ0.1, δ0.2, and δ0.3, the accuracy data were calculated as 0.9838, 0.9843, 0.9838, and 0.9846 for pnt; 0.9514, 0.9517, 0.9514, and 0.9514 for wt, respectively in the training stage. The GB algorithm had the best performance value, except for the results of the δ0.2 (AB had a better accuracy at 0.8709 based on the value of δ0.2 for pnt) in the test stage. According to the AB algorithm based on the δ0.0, δ0.1, δ0.2, and δ0.3, the accuracy data were calculated as 0.7956, 0.9298, 0.8288, and 0.7394 for pnt; 0.8820, 0.8821, 0.8819, and 0.8818 for wt in the training phase, respectively. All scenarios created by the δi coefficient should be preferred for ED since the income provided by the pnt value to the hospital was more than the cost of healthcare resources. On the contrary, the wt estimation results of ML algorithms based on the δi coefficient differed. Although wt values in all ML algorithms with δ0.0 and δ0.1 coefficients reduced the cost of the hospital, wt values based on δ0.2 and δ0.3 increased the cost of the hospital.

show abstract

“…Pak et al reduced the number of patients with

reduction was achieved indirectly) from a general hospital’s emergency department in South Korea [ 50 ]. In terms of estimating

, ML models are often used to estimate the number of patients for a disease type.…”

Section: Discussionmentioning

confidence: 99%

Integration of Machine Learning Algorithms and Discrete-Event Simulation for the Cost of Healthcare Resources

2022

View full text Add to dashboard Cite

show abstract

Machine learning-based prediction of length of stay (LoS) in the neonatal intensive care unit using ensemble methods

Erdogan Yildirim,

Canayaz

2024

Neural Comput & Applic

View full text Add to dashboard Cite

Neonatal medical data holds critical information within the healthcare industry, and it is important to analyze this data effectively. Machine learning algorithms offer powerful tools for extracting meaningful insights from the medical data of neonates and improving treatment processes. Knowing the length of hospital stay in advance is very important for managing hospital resources, healthcare personnel, and costs. Thus, this study aims to estimate the length of stay for infants treated in the Neonatal Intensive Care Unit (NICU) using machine learning algorithms. Our study conducted a two-class prediction for long and short-term lengths of stay utilizing a unique dataset. Adopting a hybrid approach called Classifier Fusion-LoS, the study involved two stages. In the initial stage, various classifiers were employed including classical models such as Logistic Regression, ExtraTrees, Random Forest, KNN, Support Vector Classifier, as well as ensemble models like AdaBoost, GradientBoosting, XGBoost, and CatBoost. Random Forest yielded the highest validation accuracy at 0.94. In the subsequent stage, the Voting Classifier—an ensemble method—was applied, resulting in accuracy increasing to 0.96. Our method outperformed existing studies in terms of accuracy, including both neonatal-specific length of stay prediction studies and other general length of stay prediction research. While the length of stay estimation offers insights into the potential suitability of the incubators in the NICUs, which are not universally available in every city, for patient admission, it plays a pivotal role in delineating the treatment protocols of patients. Additionally, the research provides crucial information to the hospital management for planning such as beds, equipment, personnel, and costs.

show abstract

Evaluation of the Prediction Algorithms for the Diagnosis of Hepatic Dysfunction

ARPACI,

VARLI

2023

Nevşehir Bilim Ve Teknoloji Dergisi

View full text Add to dashboard Cite

Acute liver failure develops due to liver dysfunction. Early diagnosis is crucial for acute liver failure, which develops in a short time and causes serious damage to the body. Prediction processes based on machine learning methods can provide assistance to the physician in the decision-making process in order for the physician to make a diagnosis earlier. This study aims to evaluate three recently presented algorithms with high predictive capabilities that can assist the doctor in determining the existence of acute liver failure. In this study, the prediction performances of the XGBoost, LightGBM, and NGBoost methods are examined on publicly available data sets. In this research, two datasets are used; the first dataset was gathered in the “JPAC Health Diagnostic and Control Center” during the periods 2008–2009 and 2014–2015. The dataset includes a total of 8785 patients' information, and it mostly does not contain patients' information that "acute liver failure" was developing. Furthermore, a dataset collected by Iesu et al., containing information on patients who developed or did not develop "acute liver dysfunction," is used for the second evaluation. According to the information obtained from the data set, "acute liver dysfunction" developed in 208 patients, while this situation did not develop in 166 patients. It is observed within the scope of the evaluations that all three algorithms give high estimation results during the training and testing stages, and moreover, the LightGBM method achieves results in a shorter time while the NGBoost method provides results in a longer time compared to other algorithms.

show abstract

Analysis on Benefits and Costs of Machine Learning-Based Early Hospitalization Prediction

Cited by 3 publications

References 33 publications

Integration of Machine Learning Algorithms and Discrete-Event Simulation for the Cost of Healthcare Resources

Integration of Machine Learning Algorithms and Discrete-Event Simulation for the Cost of Healthcare Resources

Machine learning-based prediction of length of stay (LoS) in the neonatal intensive care unit using ensemble methods

Evaluation of the Prediction Algorithms for the Diagnosis of Hepatic Dysfunction

Contact Info

Product

Resources

About