Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis
BackgroundSepsis is a life-threatening condition caused by a dysregulated response to infection, affecting millions of people worldwide. Early diagnosis and treatment are critical for managing sepsis and reducing morbidity and mortality rates.Materials and methodsA systematic design approach was emp...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-11-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2023.1284081/full |
| _version_ | 1797499705907740672 |
|---|---|
| author | Mohammed A. Mahyoub Mohammed A. Mahyoub Ravi R. Yadav Kacie Dougherty Ajit Shukla |
| author_facet | Mohammed A. Mahyoub Mohammed A. Mahyoub Ravi R. Yadav Kacie Dougherty Ajit Shukla |
| author_sort | Mohammed A. Mahyoub |
| collection | DOAJ |
| description | BackgroundSepsis is a life-threatening condition caused by a dysregulated response to infection, affecting millions of people worldwide. Early diagnosis and treatment are critical for managing sepsis and reducing morbidity and mortality rates.Materials and methodsA systematic design approach was employed to build a model that predicts sepsis, incorporating clinical feedback to identify relevant data elements. XGBoost was utilized for prediction, and interpretability was achieved through the application of Shapley values. The model was successfully deployed within a widely used Electronic Medical Record (EMR) system.ResultsThe developed model demonstrated robust performance pre-operations, with a sensitivity of 92%, specificity of 93%, and a false positive rate of 7%. Following deployment, the model maintained comparable performance, with a sensitivity of 91% and specificity of 94%. Notably, the post-deployment false positive rate of 6% represents a substantial reduction compared to the currently deployed commercial model in the same health system, which exhibits a false positive rate of 30%.DiscussionThese findings underscore the effectiveness and potential value of the developed model in improving timely sepsis detection and reducing unnecessary alerts in clinical practice. Further investigations should focus on its long-term generalizability and impact on patient outcomes. |
| first_indexed | 2024-03-10T03:51:16Z |
| format | Article |
| id | doaj.art-6b86f4ff658847af8f29a01694065e01 |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-03-10T03:51:16Z |
| publishDate | 2023-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-6b86f4ff658847af8f29a01694065e012023-11-23T11:05:05ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2023-11-011010.3389/fmed.2023.12840811284081Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsisMohammed A. Mahyoub0Mohammed A. Mahyoub1Ravi R. Yadav2Kacie Dougherty3Ajit Shukla4Advanced Analytics and Solutions, Virtua Health, Marlton, NJ, United StatesSystems Science and Industrial Engineering Department, Binghamton University, Binghamton, NY, United StatesAdvanced Analytics and Solutions, Virtua Health, Marlton, NJ, United StatesAdvanced Analytics and Solutions, Virtua Health, Marlton, NJ, United StatesAdvanced Analytics and Solutions, Virtua Health, Marlton, NJ, United StatesBackgroundSepsis is a life-threatening condition caused by a dysregulated response to infection, affecting millions of people worldwide. Early diagnosis and treatment are critical for managing sepsis and reducing morbidity and mortality rates.Materials and methodsA systematic design approach was employed to build a model that predicts sepsis, incorporating clinical feedback to identify relevant data elements. XGBoost was utilized for prediction, and interpretability was achieved through the application of Shapley values. The model was successfully deployed within a widely used Electronic Medical Record (EMR) system.ResultsThe developed model demonstrated robust performance pre-operations, with a sensitivity of 92%, specificity of 93%, and a false positive rate of 7%. Following deployment, the model maintained comparable performance, with a sensitivity of 91% and specificity of 94%. Notably, the post-deployment false positive rate of 6% represents a substantial reduction compared to the currently deployed commercial model in the same health system, which exhibits a false positive rate of 30%.DiscussionThese findings underscore the effectiveness and potential value of the developed model in improving timely sepsis detection and reducing unnecessary alerts in clinical practice. Further investigations should focus on its long-term generalizability and impact on patient outcomes.https://www.frontiersin.org/articles/10.3389/fmed.2023.1284081/fullsepsisearly detectionmachine learningXGBoostmodel interpretabilitymachine learning deployment |
| spellingShingle | Mohammed A. Mahyoub Mohammed A. Mahyoub Ravi R. Yadav Kacie Dougherty Ajit Shukla Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis Frontiers in Medicine sepsis early detection machine learning XGBoost model interpretability machine learning deployment |
| title | Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis |
| title_full | Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis |
| title_fullStr | Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis |
| title_full_unstemmed | Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis |
| title_short | Development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis |
| title_sort | development and validation of a machine learning model integrated with the clinical workflow for early detection of sepsis |
| topic | sepsis early detection machine learning XGBoost model interpretability machine learning deployment |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2023.1284081/full |
| work_keys_str_mv | AT mohammedamahyoub developmentandvalidationofamachinelearningmodelintegratedwiththeclinicalworkflowforearlydetectionofsepsis AT mohammedamahyoub developmentandvalidationofamachinelearningmodelintegratedwiththeclinicalworkflowforearlydetectionofsepsis AT raviryadav developmentandvalidationofamachinelearningmodelintegratedwiththeclinicalworkflowforearlydetectionofsepsis AT kaciedougherty developmentandvalidationofamachinelearningmodelintegratedwiththeclinicalworkflowforearlydetectionofsepsis AT ajitshukla developmentandvalidationofamachinelearningmodelintegratedwiththeclinicalworkflowforearlydetectionofsepsis |