Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction
This study aims to address the critical issue of emergency department (ED) overcrowding, which negatively affects patient outcomes, wait times, and resource efficiency. Accurate prediction of ED length of stay (LOS) can streamline operations and improve care delivery. We utilized the MIMIC IV-ED dat...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Digital Health |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fdgth.2024.1498939/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823856483526246400 |
|---|---|
| author | Waqar A. Sulaiman Charithea Stylianides Andria Nikolaou Andria Nikolaou Zinonas Antoniou Ioannis Constantinou Lakis Palazis Anna Vavlitou Theodoros Kyprianou Theodoros Kyprianou Efthyvoulos Kyriacou Antonis Kakas Antonis Kakas Marios S. Pattichis Andreas S. Panayides Constantinos S. Pattichis Constantinos S. Pattichis |
| author_facet | Waqar A. Sulaiman Charithea Stylianides Andria Nikolaou Andria Nikolaou Zinonas Antoniou Ioannis Constantinou Lakis Palazis Anna Vavlitou Theodoros Kyprianou Theodoros Kyprianou Efthyvoulos Kyriacou Antonis Kakas Antonis Kakas Marios S. Pattichis Andreas S. Panayides Constantinos S. Pattichis Constantinos S. Pattichis |
| author_sort | Waqar A. Sulaiman |
| collection | DOAJ |
| description | This study aims to address the critical issue of emergency department (ED) overcrowding, which negatively affects patient outcomes, wait times, and resource efficiency. Accurate prediction of ED length of stay (LOS) can streamline operations and improve care delivery. We utilized the MIMIC IV-ED dataset, comprising over 400,000 patient records, to classify ED LOS into short (≤4.5 hours) and long (>4.5 hours) categories. Using machine learning models, including Gradient Boosting (GB), Random Forest (RF), Logistic Regression (LR), and Multilayer Perceptron (MLP), we identified GB as the best performing model outperforming the other models with an AUC of 0.730, accuracy of 69.93%, sensitivity of 88.20%, and specificity of 40.95% on the original dataset. In the balanced dataset, GB had an AUC of 0.729, accuracy of 68.86%, sensitivity of 75.39%, and specificity of 58.59%. To enhance interpretability, a novel rule extraction method for GB model was implemented using relevant important predictors, such as triage acuity, comorbidity scores, and arrival methods. By combining predictive analytics with interpretable rule-based methods, this research provides actionable insights for optimizing patient flow and resource allocation. The findings highlight the importance of transparency in machine learning applications for healthcare, paving the way for future improvements in model performance and clinical adoption. |
| format | Article |
| id | doaj-art-2cec9ce2cb0f42d5a5343a341ece1246 |
| institution | Kabale University |
| issn | 2673-253X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Digital Health |
| spelling | doaj-art-2cec9ce2cb0f42d5a5343a341ece12462025-02-12T07:27:18ZengFrontiers Media S.A.Frontiers in Digital Health2673-253X2025-02-01610.3389/fdgth.2024.14989391498939Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay predictionWaqar A. Sulaiman0Charithea Stylianides1Andria Nikolaou2Andria Nikolaou3Zinonas Antoniou4Ioannis Constantinou5Lakis Palazis6Anna Vavlitou7Theodoros Kyprianou8Theodoros Kyprianou9Efthyvoulos Kyriacou10Antonis Kakas11Antonis Kakas12Marios S. Pattichis13Andreas S. Panayides14Constantinos S. Pattichis15Constantinos S. Pattichis16Department of Computer Science and Biomedical Engineering Research Centre, University of Cyprus, Nicosia, CyprusCYENS Centre of Excellence, Nicosia, CyprusDepartment of Computer Science and Biomedical Engineering Research Centre, University of Cyprus, Nicosia, CyprusCYENS Centre of Excellence, Nicosia, CyprusResearch & Development Department, 3AHealth, Nicosia, CyprusResearch & Development Department, 3AHealth, Nicosia, CyprusDepartment of Intensive Care Medicine, Limassol General Hospital, State Health Services Organisation, Nicosia, CyprusDepartment of Intensive Care Medicine, Limassol General Hospital, State Health Services Organisation, Nicosia, CyprusDepartment of Critical Care and Emergency Medicine, Medical School, University of Nicosia, Nicosia, CyprusDepartment of Critical Care, St Thomas's Hospital NHS, London, United KingdomDepartment of Electrical Engineering, Computer Engineering and Informatics, Cyprus University of Technology, Limassol, CyprusDepartment of Computer Science and Biomedical Engineering Research Centre, University of Cyprus, Nicosia, CyprusCYENS Centre of Excellence, Nicosia, CyprusDepartment of Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM, United StatesCYENS Centre of Excellence, Nicosia, CyprusDepartment of Computer Science and Biomedical Engineering Research Centre, University of Cyprus, Nicosia, CyprusCYENS Centre of Excellence, Nicosia, CyprusThis study aims to address the critical issue of emergency department (ED) overcrowding, which negatively affects patient outcomes, wait times, and resource efficiency. Accurate prediction of ED length of stay (LOS) can streamline operations and improve care delivery. We utilized the MIMIC IV-ED dataset, comprising over 400,000 patient records, to classify ED LOS into short (≤4.5 hours) and long (>4.5 hours) categories. Using machine learning models, including Gradient Boosting (GB), Random Forest (RF), Logistic Regression (LR), and Multilayer Perceptron (MLP), we identified GB as the best performing model outperforming the other models with an AUC of 0.730, accuracy of 69.93%, sensitivity of 88.20%, and specificity of 40.95% on the original dataset. In the balanced dataset, GB had an AUC of 0.729, accuracy of 68.86%, sensitivity of 75.39%, and specificity of 58.59%. To enhance interpretability, a novel rule extraction method for GB model was implemented using relevant important predictors, such as triage acuity, comorbidity scores, and arrival methods. By combining predictive analytics with interpretable rule-based methods, this research provides actionable insights for optimizing patient flow and resource allocation. The findings highlight the importance of transparency in machine learning applications for healthcare, paving the way for future improvements in model performance and clinical adoption.https://www.frontiersin.org/articles/10.3389/fdgth.2024.1498939/fullemergency departmentlength of staymachine learninggradient boostingrule extractionpredictive modeling |
| spellingShingle | Waqar A. Sulaiman Charithea Stylianides Andria Nikolaou Andria Nikolaou Zinonas Antoniou Ioannis Constantinou Lakis Palazis Anna Vavlitou Theodoros Kyprianou Theodoros Kyprianou Efthyvoulos Kyriacou Antonis Kakas Antonis Kakas Marios S. Pattichis Andreas S. Panayides Constantinos S. Pattichis Constantinos S. Pattichis Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction Frontiers in Digital Health emergency department length of stay machine learning gradient boosting rule extraction predictive modeling |
| title | Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction |
| title_full | Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction |
| title_fullStr | Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction |
| title_full_unstemmed | Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction |
| title_short | Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction |
| title_sort | leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction |
| topic | emergency department length of stay machine learning gradient boosting rule extraction predictive modeling |
| url | https://www.frontiersin.org/articles/10.3389/fdgth.2024.1498939/full |
| work_keys_str_mv | AT waqarasulaiman leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT charitheastylianides leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT andrianikolaou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT andrianikolaou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT zinonasantoniou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT ioannisconstantinou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT lakispalazis leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT annavavlitou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT theodoroskyprianou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT theodoroskyprianou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT efthyvouloskyriacou leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT antoniskakas leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT antoniskakas leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT mariosspattichis leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT andreasspanayides leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT constantinosspattichis leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction AT constantinosspattichis leveragingmachinelearningandruleextractionforenhancedtransparencyinemergencydepartmentlengthofstayprediction |