Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis
Sepsis remains a major global health concern, causing high mortality rates, prolonged hospital stays, and substantial economic burdens. The accurate prediction of clinical outcomes, such as mortality and length of stay (LOS), is critical for optimizing hospital resource allocation and improving pati...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Computation |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-3197/13/1/8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588792059396096 |
|---|---|
| author | Houssem Ben Khalfallah Mariem Jelassi Jacques Demongeot Narjès Bellamine Ben Saoud |
| author_facet | Houssem Ben Khalfallah Mariem Jelassi Jacques Demongeot Narjès Bellamine Ben Saoud |
| author_sort | Houssem Ben Khalfallah |
| collection | DOAJ |
| description | Sepsis remains a major global health concern, causing high mortality rates, prolonged hospital stays, and substantial economic burdens. The accurate prediction of clinical outcomes, such as mortality and length of stay (LOS), is critical for optimizing hospital resource allocation and improving patient management. The present study investigates the potential of machine learning (ML) models to predict these outcomes using a dataset of 1492 sepsis patients with clinical, physiological, and demographic features. After rigorous preprocessing to address missing data and ensure consistency, multiple classifiers, including Random Forest, Extra Trees, and Gradient Boosting, were trained and validated. The results demonstrate that Random Forest and Extra Trees achieve high accuracy for LOS prediction, while Gradient Boosting and Bernoulli Naïve Bayes effectively predict mortality. Feature importance analysis identified ICU stay duration (ICU_DAYS_OBS) as the most influential predictor for both outcomes, alongside vital signs, white blood cell counts, and lactic acid levels. These findings highlight the potential of ML-driven clinical decision support systems (CDSSs) to enhance early risk assessment, optimize ICU resource planning, and support timely interventions. Future research should refine predictive features, integrate advanced biomarkers, and validate models across larger and more diverse datasets to improve scalability and clinical impact. |
| format | Article |
| id | doaj-art-b3790a65b816456c8ba54424d9080af5 |
| institution | Kabale University |
| issn | 2079-3197 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computation |
| spelling | doaj-art-b3790a65b816456c8ba54424d9080af52025-01-24T13:27:47ZengMDPI AGComputation2079-31972025-01-01131810.3390/computation13010008Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in SepsisHoussem Ben Khalfallah0Mariem Jelassi1Jacques Demongeot2Narjès Bellamine Ben Saoud3RIADI Laboratory, Ecole Nationale des Sciences de l’Informatique, Manouba University, La Manouba 2010, TunisiaRIADI Laboratory, Ecole Nationale des Sciences de l’Informatique, Manouba University, La Manouba 2010, TunisiaAGEIS Laboratory, University Grenoble Alpes, 38700 La Tronche, FranceRIADI Laboratory, Ecole Nationale des Sciences de l’Informatique, Manouba University, La Manouba 2010, TunisiaSepsis remains a major global health concern, causing high mortality rates, prolonged hospital stays, and substantial economic burdens. The accurate prediction of clinical outcomes, such as mortality and length of stay (LOS), is critical for optimizing hospital resource allocation and improving patient management. The present study investigates the potential of machine learning (ML) models to predict these outcomes using a dataset of 1492 sepsis patients with clinical, physiological, and demographic features. After rigorous preprocessing to address missing data and ensure consistency, multiple classifiers, including Random Forest, Extra Trees, and Gradient Boosting, were trained and validated. The results demonstrate that Random Forest and Extra Trees achieve high accuracy for LOS prediction, while Gradient Boosting and Bernoulli Naïve Bayes effectively predict mortality. Feature importance analysis identified ICU stay duration (ICU_DAYS_OBS) as the most influential predictor for both outcomes, alongside vital signs, white blood cell counts, and lactic acid levels. These findings highlight the potential of ML-driven clinical decision support systems (CDSSs) to enhance early risk assessment, optimize ICU resource planning, and support timely interventions. Future research should refine predictive features, integrate advanced biomarkers, and validate models across larger and more diverse datasets to improve scalability and clinical impact.https://www.mdpi.com/2079-3197/13/1/8sepsismachine learninglength of stay (LOS)mortality predictionclinical decision support systems (CDSSs)intensive care unit (ICU) |
| spellingShingle | Houssem Ben Khalfallah Mariem Jelassi Jacques Demongeot Narjès Bellamine Ben Saoud Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis Computation sepsis machine learning length of stay (LOS) mortality prediction clinical decision support systems (CDSSs) intensive care unit (ICU) |
| title | Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis |
| title_full | Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis |
| title_fullStr | Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis |
| title_full_unstemmed | Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis |
| title_short | Advancements in Predictive Analytics: Machine Learning Approaches to Estimating Length of Stay and Mortality in Sepsis |
| title_sort | advancements in predictive analytics machine learning approaches to estimating length of stay and mortality in sepsis |
| topic | sepsis machine learning length of stay (LOS) mortality prediction clinical decision support systems (CDSSs) intensive care unit (ICU) |
| url | https://www.mdpi.com/2079-3197/13/1/8 |
| work_keys_str_mv | AT houssembenkhalfallah advancementsinpredictiveanalyticsmachinelearningapproachestoestimatinglengthofstayandmortalityinsepsis AT mariemjelassi advancementsinpredictiveanalyticsmachinelearningapproachestoestimatinglengthofstayandmortalityinsepsis AT jacquesdemongeot advancementsinpredictiveanalyticsmachinelearningapproachestoestimatinglengthofstayandmortalityinsepsis AT narjesbellaminebensaoud advancementsinpredictiveanalyticsmachinelearningapproachestoestimatinglengthofstayandmortalityinsepsis |