Performance of each model for prediction.
<div><p>Background</p><p>Lymphoma is a severe condition with high mortality rates, often requiring ICU admission. Traditional risk stratification tools like SOFA and APACHE scores struggle to capture complex clinical interactions. Machine learning (ML) models offer a more acc...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1852017443602432000 |
|---|---|
| author | Ling Xu (117663) |
| author2 | Guang Tu (22054865) Zhonglan Cai (22054874) Tianbi Lan (9280010) |
| author2_role | author author author |
| author_facet | Ling Xu (117663) Guang Tu (22054865) Zhonglan Cai (22054874) Tianbi Lan (9280010) |
| author_role | author |
| dc.creator.none.fl_str_mv | Ling Xu (117663) Guang Tu (22054865) Zhonglan Cai (22054874) Tianbi Lan (9280010) |
| dc.date.none.fl_str_mv | 2025-08-20T17:31:50Z |
| dc.identifier.none.fl_str_mv | 10.1371/journal.pone.0330197.t002 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/dataset/Performance_of_each_model_for_prediction_/29952921 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Medicine Biotechnology Developmental Biology Cancer Virology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified systolic blood pressure shapley additive explanation provide valuable insights medical information mart apache scores struggle analyzing large datasets model &# 8217 including logistic regression highest predictive performance lymphoma using data improve patient outcomes intensive care iv catboost classifier demonstrated adult patients admitted xlink "> lymphoma shap analysis highlighted high mortality rates bun ), platelets xlink "> model performance catboost classifier lasso regression clinical outcomes study aims severe condition risk stratification random forest primary diagnosis neural networks mortality prediction models offer machine learning laboratory parameters important factor identify high hospital mortality hospital deaths heart rate gradient boosting future work critical role clinical implementation clinical decision baseline characteristics accurate alternative 5 %) 1591 patients |
| dc.title.none.fl_str_mv | Performance of each model for prediction. |
| dc.type.none.fl_str_mv | Dataset info:eu-repo/semantics/publishedVersion dataset |
| description | <div><p>Background</p><p>Lymphoma is a severe condition with high mortality rates, often requiring ICU admission. Traditional risk stratification tools like SOFA and APACHE scores struggle to capture complex clinical interactions. Machine learning (ML) models offer a more accurate alternative for predicting outcomes by analyzing large datasets. However, their application in predicting in-hospital mortality for lymphoma patients remains limited.</p><p>Objective</p><p>This study aims to develop and validate machine learning models to predict in-hospital mortality in ICU patients with lymphoma using data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, thereby enhancing risk stratification and clinical decision-making.</p><p>Methods</p><p>We conducted a retrospective cohort study using data from the MIMIC-IV database, which includes detailed clinical data from adult patients admitted to the ICU. Patients with a primary diagnosis of lymphoma were included. Baseline characteristics, laboratory parameters, and clinical outcomes were extracted. Lasso regression was employed to screen for significant risk factors associated with in-hospital mortality. Fifteen machine learning models, including logistic regression, random forest, gradient boosting, and neural networks, were developed and compared using receiver operating characteristic (ROC) curves and area under the curve (AUC) analysis. Model performance was evaluated through cross-validation and SHapley Additive exPlanation (SHAP) values to interpret variable importance.</p><p>Results</p><p>A total of 1591 patients were included, with 342 (21.5%) in-hospital deaths. Lasso regression identified significant predictors of mortality, including blood urea nitrogen (BUN), platelets, PT, heart rate, systolic blood pressure, APTT, spo2, and bicarbonate. The CatBoost Classifier demonstrated the highest predictive performance with an AUC of 0.7766. SHAP analysis highlighted the critical role of BUN as the most important factor in mortality prediction, followed by platelets and PT. The SHAP force plot provided individualized risk assessments for patients, demonstrating the model’s ability to identify high-risk subgroups.</p><p>Conclusion</p><p>Machine learning models, particularly the CatBoost Classifier, effectively predict in-hospital mortality in ICU patients with lymphoma. These models outperform traditional statistical methods and provide valuable insights into risk stratification. Future work should focus on external validation and clinical implementation to improve patient outcomes in this high-risk population.</p></div> |
| eu_rights_str_mv | openAccess |
| id | Manara_0f4fd007f4aee2d079a6101e55db3ec5 |
| identifier_str_mv | 10.1371/journal.pone.0330197.t002 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/29952921 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Performance of each model for prediction.Ling Xu (117663)Guang Tu (22054865)Zhonglan Cai (22054874)Tianbi Lan (9280010)MedicineBiotechnologyDevelopmental BiologyCancerVirologyBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedsystolic blood pressureshapley additive explanationprovide valuable insightsmedical information martapache scores struggleanalyzing large datasetsmodel &# 8217including logistic regressionhighest predictive performancelymphoma using dataimprove patient outcomesintensive care ivcatboost classifier demonstratedadult patients admittedxlink "> lymphomashap analysis highlightedhigh mortality ratesbun ), plateletsxlink ">model performancecatboost classifierlasso regressionclinical outcomesstudy aimssevere conditionrisk stratificationrandom forestprimary diagnosisneural networksmortality predictionmodels offermachine learninglaboratory parametersimportant factoridentify highhospital mortalityhospital deathsheart rategradient boostingfuture workcritical roleclinical implementationclinical decisionbaseline characteristicsaccurate alternative5 %)1591 patients<div><p>Background</p><p>Lymphoma is a severe condition with high mortality rates, often requiring ICU admission. Traditional risk stratification tools like SOFA and APACHE scores struggle to capture complex clinical interactions. Machine learning (ML) models offer a more accurate alternative for predicting outcomes by analyzing large datasets. However, their application in predicting in-hospital mortality for lymphoma patients remains limited.</p><p>Objective</p><p>This study aims to develop and validate machine learning models to predict in-hospital mortality in ICU patients with lymphoma using data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, thereby enhancing risk stratification and clinical decision-making.</p><p>Methods</p><p>We conducted a retrospective cohort study using data from the MIMIC-IV database, which includes detailed clinical data from adult patients admitted to the ICU. Patients with a primary diagnosis of lymphoma were included. Baseline characteristics, laboratory parameters, and clinical outcomes were extracted. Lasso regression was employed to screen for significant risk factors associated with in-hospital mortality. Fifteen machine learning models, including logistic regression, random forest, gradient boosting, and neural networks, were developed and compared using receiver operating characteristic (ROC) curves and area under the curve (AUC) analysis. Model performance was evaluated through cross-validation and SHapley Additive exPlanation (SHAP) values to interpret variable importance.</p><p>Results</p><p>A total of 1591 patients were included, with 342 (21.5%) in-hospital deaths. Lasso regression identified significant predictors of mortality, including blood urea nitrogen (BUN), platelets, PT, heart rate, systolic blood pressure, APTT, spo2, and bicarbonate. The CatBoost Classifier demonstrated the highest predictive performance with an AUC of 0.7766. SHAP analysis highlighted the critical role of BUN as the most important factor in mortality prediction, followed by platelets and PT. The SHAP force plot provided individualized risk assessments for patients, demonstrating the model’s ability to identify high-risk subgroups.</p><p>Conclusion</p><p>Machine learning models, particularly the CatBoost Classifier, effectively predict in-hospital mortality in ICU patients with lymphoma. These models outperform traditional statistical methods and provide valuable insights into risk stratification. Future work should focus on external validation and clinical implementation to improve patient outcomes in this high-risk population.</p></div>2025-08-20T17:31:50ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1371/journal.pone.0330197.t002https://figshare.com/articles/dataset/Performance_of_each_model_for_prediction_/29952921CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/299529212025-08-20T17:31:50Z |
| spellingShingle | Performance of each model for prediction. Ling Xu (117663) Medicine Biotechnology Developmental Biology Cancer Virology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified systolic blood pressure shapley additive explanation provide valuable insights medical information mart apache scores struggle analyzing large datasets model &# 8217 including logistic regression highest predictive performance lymphoma using data improve patient outcomes intensive care iv catboost classifier demonstrated adult patients admitted xlink "> lymphoma shap analysis highlighted high mortality rates bun ), platelets xlink "> model performance catboost classifier lasso regression clinical outcomes study aims severe condition risk stratification random forest primary diagnosis neural networks mortality prediction models offer machine learning laboratory parameters important factor identify high hospital mortality hospital deaths heart rate gradient boosting future work critical role clinical implementation clinical decision baseline characteristics accurate alternative 5 %) 1591 patients |
| status_str | publishedVersion |
| title | Performance of each model for prediction. |
| title_full | Performance of each model for prediction. |
| title_fullStr | Performance of each model for prediction. |
| title_full_unstemmed | Performance of each model for prediction. |
| title_short | Performance of each model for prediction. |
| title_sort | Performance of each model for prediction. |
| topic | Medicine Biotechnology Developmental Biology Cancer Virology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified systolic blood pressure shapley additive explanation provide valuable insights medical information mart apache scores struggle analyzing large datasets model &# 8217 including logistic regression highest predictive performance lymphoma using data improve patient outcomes intensive care iv catboost classifier demonstrated adult patients admitted xlink "> lymphoma shap analysis highlighted high mortality rates bun ), platelets xlink "> model performance catboost classifier lasso regression clinical outcomes study aims severe condition risk stratification random forest primary diagnosis neural networks mortality prediction models offer machine learning laboratory parameters important factor identify high hospital mortality hospital deaths heart rate gradient boosting future work critical role clinical implementation clinical decision baseline characteristics accurate alternative 5 %) 1591 patients |