Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model
This study aims to develop an interpretable prediction model based on explainable artificial intelligence to predict bacterial sepsis and discover important biomarkers. A total of 1572 adult patients, 560 of whom were sepsis positive and 1012 of whom were negative, who were admitted to the emergency...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/14/5/457 |
| _version_ | 1797264737343373312 |
|---|---|
| author | Umran Aygun Fatma Hilal Yagin Burak Yagin Seyma Yasar Cemil Colak Ahmet Selim Ozkan Luca Paolo Ardigò |
| author_facet | Umran Aygun Fatma Hilal Yagin Burak Yagin Seyma Yasar Cemil Colak Ahmet Selim Ozkan Luca Paolo Ardigò |
| author_sort | Umran Aygun |
| collection | DOAJ |
| description | This study aims to develop an interpretable prediction model based on explainable artificial intelligence to predict bacterial sepsis and discover important biomarkers. A total of 1572 adult patients, 560 of whom were sepsis positive and 1012 of whom were negative, who were admitted to the emergency department with suspicion of sepsis, were examined. We investigated the performance characteristics of sepsis biomarkers alone and in combination for confirmed sepsis diagnosis using Sepsis-3 criteria. Three different tree-based algorithms—Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Adaptive Boosting (AdaBoost)—were used for sepsis prediction, and after examining comprehensive performance metrics, descriptions of the optimal model were obtained with the SHAP method. The XGBoost model achieved accuracy of 0.898 (0.868–0.929) and area under the ROC curve (AUC) of 0.940 (0.898–0.980) with a 95% confidence interval. The five biomarkers for predicting sepsis were age, respiratory rate, oxygen saturation, procalcitonin, and positive blood culture. SHAP results revealed that older age, higher respiratory rate, procalcitonin, neutrophil–lymphocyte count ratio, C-reactive protein, plaque, leukocyte particle concentration, as well as lower oxygen saturation, systolic blood pressure, and hemoglobin levels increased the risk of sepsis. As a result, the Explainable Artificial Intelligence (XAI)-based prediction model can guide clinicians in the early diagnosis and treatment of sepsis, providing more effective sepsis management and potentially reducing mortality rates and medical costs. |
| first_indexed | 2024-04-25T00:33:39Z |
| format | Article |
| id | doaj.art-b96b333daefd44c3bda9673e94f1ef94 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-04-25T00:33:39Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-b96b333daefd44c3bda9673e94f1ef942024-03-12T16:41:49ZengMDPI AGDiagnostics2075-44182024-02-0114545710.3390/diagnostics14050457Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction ModelUmran Aygun0Fatma Hilal Yagin1Burak Yagin2Seyma Yasar3Cemil Colak4Ahmet Selim Ozkan5Luca Paolo Ardigò6Department of Anesthesiology and Reanimation, Malatya Yesilyurt Hasan Calık State Hospital, Malatya 44929, TurkeyDepartment of Biostatistics and Medical Informatics, Faculty of Medicine, Inonu University, Malatya 44280, TurkeyDepartment of Biostatistics and Medical Informatics, Faculty of Medicine, Inonu University, Malatya 44280, TurkeyDepartment of Biostatistics and Medical Informatics, Faculty of Medicine, Inonu University, Malatya 44280, TurkeyDepartment of Biostatistics and Medical Informatics, Faculty of Medicine, Inonu University, Malatya 44280, TurkeyDepartment of Anesthesiology and Reanimation, Malatya Turgut Ozal University School of Medicine, Malatya 44210, TurkeyDepartment of Teacher Education, NLA University College, 0166 Oslo, NorwayThis study aims to develop an interpretable prediction model based on explainable artificial intelligence to predict bacterial sepsis and discover important biomarkers. A total of 1572 adult patients, 560 of whom were sepsis positive and 1012 of whom were negative, who were admitted to the emergency department with suspicion of sepsis, were examined. We investigated the performance characteristics of sepsis biomarkers alone and in combination for confirmed sepsis diagnosis using Sepsis-3 criteria. Three different tree-based algorithms—Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Adaptive Boosting (AdaBoost)—were used for sepsis prediction, and after examining comprehensive performance metrics, descriptions of the optimal model were obtained with the SHAP method. The XGBoost model achieved accuracy of 0.898 (0.868–0.929) and area under the ROC curve (AUC) of 0.940 (0.898–0.980) with a 95% confidence interval. The five biomarkers for predicting sepsis were age, respiratory rate, oxygen saturation, procalcitonin, and positive blood culture. SHAP results revealed that older age, higher respiratory rate, procalcitonin, neutrophil–lymphocyte count ratio, C-reactive protein, plaque, leukocyte particle concentration, as well as lower oxygen saturation, systolic blood pressure, and hemoglobin levels increased the risk of sepsis. As a result, the Explainable Artificial Intelligence (XAI)-based prediction model can guide clinicians in the early diagnosis and treatment of sepsis, providing more effective sepsis management and potentially reducing mortality rates and medical costs.https://www.mdpi.com/2075-4418/14/5/457sepsismachine learningexplainable artificial intelligencebiomarker |
| spellingShingle | Umran Aygun Fatma Hilal Yagin Burak Yagin Seyma Yasar Cemil Colak Ahmet Selim Ozkan Luca Paolo Ardigò Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model Diagnostics sepsis machine learning explainable artificial intelligence biomarker |
| title | Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model |
| title_full | Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model |
| title_fullStr | Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model |
| title_full_unstemmed | Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model |
| title_short | Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model |
| title_sort | assessment of sepsis risk at admission to the emergency department clinical interpretable prediction model |
| topic | sepsis machine learning explainable artificial intelligence biomarker |
| url | https://www.mdpi.com/2075-4418/14/5/457 |
| work_keys_str_mv | AT umranaygun assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel AT fatmahilalyagin assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel AT burakyagin assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel AT seymayasar assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel AT cemilcolak assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel AT ahmetselimozkan assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel AT lucapaoloardigo assessmentofsepsisriskatadmissiontotheemergencydepartmentclinicalinterpretablepredictionmodel |