Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit
ObjectiveTo identify and validate the clinical phenotypes of patients with sepsis in the intensive care unit(ICU).MethodsWe applied unsupervised machine learning algorithms (K-means clusteringand hierarchical clustering) to identify the phenotypes of sepsis patients in the Medical Information Mart f...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Medical Journal of Peking Union Medical College Hospital
2025-01-01
|
| Series: | Xiehe Yixue Zazhi |
| Subjects: | |
| Online Access: | https://xhyxzz.pumch.cn/article/doi/10.12290/xhyxzz.2024-0353 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849470991914237952 |
|---|---|
| author | GONG Chao YU Na CHEN Haoran |
| author_facet | GONG Chao YU Na CHEN Haoran |
| author_sort | GONG Chao |
| collection | DOAJ |
| description | ObjectiveTo identify and validate the clinical phenotypes of patients with sepsis in the intensive care unit(ICU).MethodsWe applied unsupervised machine learning algorithms (K-means clusteringand hierarchical clustering) to identify the phenotypes of sepsis patients in the Medical Information Mart for Intensive Care Ⅳ (MIMIC-Ⅳ) 2.2 database, based on 89 clinical features including demographic characteristics, laboratory indicators and treatment measures on the first day in ICU. Then, supervised machine learning algorithms (lightweight gradient boosting machine) were used for the prediction of the patient's phenotypes, and were further combined with SHAP (Shapely Additive eXplanations) for the identification of important features. Finally, traditional statistical methods were used to validate the differences in clinical characteristics and clinical outcomes among the phenotypes.ResultsWe identified three phenotypes in 22 517 sepsis patients. The phenotype 1 patients had the highest risk of death (28-day mortality of 46.4%), dominated by abnormal renal function and elevated disease severity scores, while the phenotype 3 patients had the lowest risk of death (28-day mortality of 11.2%), and the best neurological function score. Using interpretable machine learning, we identified six features (all the worst value on the first day) that showed good performance in phenotypic identification(AUC≥0.89) and phenotypic prognostic prediction (AUC≥0.74): anion gap, blood urea nitrogen, creatinine, Glasgow Coma Scale score, prothrombin time, and Sequential Organ Failure Assessment score. The mortality risk of phenotype 3 patients was the lowest at 28 days, 60 days, 90 days, and 1 year after ICU discharge (HR < 1).ConclusionUsing machine learning methods, we successfully identified three clinical phenotypes of sepsis patients with different clinical characteristics and prognosis and screened out six key clinical features, which are expected to play an important role in the phenotype classification and prognostic assessment of sepsis and are conducive to individualized treatment. |
| format | Article |
| id | doaj-art-bda46b2496a44428b56791ff9feaa8af |
| institution | Kabale University |
| issn | 1674-9081 |
| language | zho |
| publishDate | 2025-01-01 |
| publisher | Editorial Office of Medical Journal of Peking Union Medical College Hospital |
| record_format | Article |
| series | Xiehe Yixue Zazhi |
| spelling | doaj-art-bda46b2496a44428b56791ff9feaa8af2025-08-20T03:24:59ZzhoEditorial Office of Medical Journal of Peking Union Medical College HospitalXiehe Yixue Zazhi1674-90812025-01-0116371072110.12290/xhyxzz.2024-0353Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care UnitGONG ChaoYU Na0CHEN Haoran1Institute of Medical Biology, Chinese Academy of Medical Sciences & Peking Union Medical College, Kunming 650118, ChinaInstitute of Medical Information/Library, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100020, ChinaObjectiveTo identify and validate the clinical phenotypes of patients with sepsis in the intensive care unit(ICU).MethodsWe applied unsupervised machine learning algorithms (K-means clusteringand hierarchical clustering) to identify the phenotypes of sepsis patients in the Medical Information Mart for Intensive Care Ⅳ (MIMIC-Ⅳ) 2.2 database, based on 89 clinical features including demographic characteristics, laboratory indicators and treatment measures on the first day in ICU. Then, supervised machine learning algorithms (lightweight gradient boosting machine) were used for the prediction of the patient's phenotypes, and were further combined with SHAP (Shapely Additive eXplanations) for the identification of important features. Finally, traditional statistical methods were used to validate the differences in clinical characteristics and clinical outcomes among the phenotypes.ResultsWe identified three phenotypes in 22 517 sepsis patients. The phenotype 1 patients had the highest risk of death (28-day mortality of 46.4%), dominated by abnormal renal function and elevated disease severity scores, while the phenotype 3 patients had the lowest risk of death (28-day mortality of 11.2%), and the best neurological function score. Using interpretable machine learning, we identified six features (all the worst value on the first day) that showed good performance in phenotypic identification(AUC≥0.89) and phenotypic prognostic prediction (AUC≥0.74): anion gap, blood urea nitrogen, creatinine, Glasgow Coma Scale score, prothrombin time, and Sequential Organ Failure Assessment score. The mortality risk of phenotype 3 patients was the lowest at 28 days, 60 days, 90 days, and 1 year after ICU discharge (HR < 1).ConclusionUsing machine learning methods, we successfully identified three clinical phenotypes of sepsis patients with different clinical characteristics and prognosis and screened out six key clinical features, which are expected to play an important role in the phenotype classification and prognostic assessment of sepsis and are conducive to individualized treatment.https://xhyxzz.pumch.cn/article/doi/10.12290/xhyxzz.2024-0353sepsisphenotypic recognitionmachine learningprecision medicine |
| spellingShingle | GONG Chao YU Na CHEN Haoran Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit Xiehe Yixue Zazhi sepsis phenotypic recognition machine learning precision medicine |
| title | Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit |
| title_full | Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit |
| title_fullStr | Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit |
| title_full_unstemmed | Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit |
| title_short | Clinical Phenotype Identification and Validation of Patients with Sepsis in the Intensive Care Unit |
| title_sort | clinical phenotype identification and validation of patients with sepsis in the intensive care unit |
| topic | sepsis phenotypic recognition machine learning precision medicine |
| url | https://xhyxzz.pumch.cn/article/doi/10.12290/xhyxzz.2024-0353 |
| work_keys_str_mv | AT gongchao clinicalphenotypeidentificationandvalidationofpatientswithsepsisintheintensivecareunit AT yuna clinicalphenotypeidentificationandvalidationofpatientswithsepsisintheintensivecareunit AT chenhaoran clinicalphenotypeidentificationandvalidationofpatientswithsepsisintheintensivecareunit |