Enhancing Patient Selection in Sepsis Clinical Trials Design Through an AI Enrichment Strategy: Algorithm Development and Validation

Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Journal of medical Internet research Ročník 26; číslo 8; s. e54621
Hlavní autori:	Yang, Meicheng, Zhuang, Jinqiang, Hu, Wenhan, Li, Jianqing, Wang, Yu, Zhang, Zhongheng, Liu, Chengyu, Chen, Hui
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Canada Journal of Medical Internet Research 04.09.2024 JMIR Publications
Predmet:	Aged Algorithms Artificial Intelligence Clinical trials Clinical Trials as Topic - methods Data mining Decision-making Female Humans Infection Machine learning Male Medical centers Medical research Medicine, Experimental Middle Aged Online databases Original Paper Patient Selection Retrospective Studies Sepsis - therapy disease progression trajectories conformal prediction predictive modeling artificial intelligence enrichment strategy sepsis
ISSN:	1438-8871, 1439-4456, 1438-8871
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown. We aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level. In this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model. The multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients. The implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model's reliability and assists in making informed decisions based on the predicted outcomes.
AbstractList	Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown. We aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level. In this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model. The multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients. The implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model’s reliability and assists in making informed decisions based on the predicted outcomes. BackgroundSepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown. ObjectiveWe aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level. MethodsIn this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model. ResultsThe multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients. ConclusionsThe implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model’s reliability and assists in making informed decisions based on the predicted outcomes. Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown.BACKGROUNDSepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown.We aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level.OBJECTIVEWe aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level.In this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model.METHODSIn this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model.The multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients.RESULTSThe multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients.The implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model's reliability and assists in making informed decisions based on the predicted outcomes.CONCLUSIONSThe implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model's reliability and assists in making informed decisions based on the predicted outcomes. Background Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown. Objective We aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level. Methods In this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model. Results The multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients. Conclusions The implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model’s reliability and assists in making informed decisions based on the predicted outcomes. Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial intelligence (AI) has facilitated the identification of homogeneous subgroups, but how to estimate the uncertainty of the model outputs when applying AI to clinical decision-making remains unknown. We aimed to design an AI-based model for purposeful patient enrollment, ensuring that a patient with sepsis recruited into a trial would still be persistently ill by the time the proposed therapy could impact patient outcome. We also expected that the model could provide interpretable factors and estimate the uncertainty of the model outputs at a customized confidence level. In this retrospective study, 9135 patients with sepsis requiring vasopressor treatment within 24 hours after sepsis onset were enrolled from Beth Israel Deaconess Medical Center. This cohort was used for model development, and 10-fold cross-validation with 50 repeats was used for internal validation. In total, 3743 patients with sepsis from the eICU Collaborative Research Database were used as the external validation cohort. All included patients with sepsis were stratified based on disease progression trajectories: rapid death, recovery, and persistent ill. A total of 148 variables were selected for predicting the 3 trajectories. Four machine learning algorithms with 3 different setups were used. We estimated the uncertainty of the model outputs using conformal prediction (CP). The Shapley Additive Explanations method was used to explain the model. The multiclass gradient boosting machine was identified as the best-performing model with good discrimination and calibration performance in both validation cohorts. The mean area under the receiver operating characteristic curve with SD was 0.906 (0.018) for rapid death, 0.843 (0.008) for recovery, and 0.807 (0.010) for persistent ill in the internal validation cohort. In the external validation cohort, the mean area under the receiver operating characteristic curve (SD) was 0.878 (0.003) for rapid death, 0.764 (0.008) for recovery, and 0.696 (0.007) for persistent ill. The maximum norepinephrine equivalence, total urine output, Acute Physiology Score III, mean systolic blood pressure, and the coefficient of variation of oxygen saturation contributed the most. Compared to the model without CP, using the model with CP at a mixed confidence approach reduced overall prediction errors by 27.6% (n=62) and 30.7% (n=412) in the internal and external validation cohorts, respectively, as well as enabled the identification of more potentially persistent ill patients. The implementation of our model has the potential to reduce heterogeneity and enroll more homogeneous patients in sepsis clinical trials. The use of CP for estimating the uncertainty of the model outputs allows for a more comprehensive understanding of the model's reliability and assists in making informed decisions based on the predicted outcomes.
Audience	Academic
Author	Zhuang, Jinqiang Chen, Hui Hu, Wenhan Wang, Yu Zhang, Zhongheng Li, Jianqing Liu, Chengyu Yang, Meicheng
AuthorAffiliation	4 Jiangsu Provincial Key Laboratory of Critical Care Medicine, Department of Critical Care Medicine, Zhongda Hospital School of Medicine Southeast University Nanjing China 6 Department of Emergency Medicine, Key Laboratory of Precision Medicine in Diagnosis and Monitoring Research of Zhejiang Province Sir Run Run Shaw Hospital Zhejiang University School of Medicine Hangzhou China 5 School of Biomedical Engineering and Informatics Nanjing Medical University Nanjing China 3 Key Laboratory of Big Data Analysis and Knowledge Services of Yangzhou City Yangzhou University Yangzhou China 2 Emergency Intensive Care Unit (EICU) The Affiliated Hospital of Yangzhou University Yangzhou University Yangzhou China 1 State Key Laboratory of Digital Medical Engineering School of Instrument Science and Engineering Southeast University Nanjing China
AuthorAffiliation_xml	– name: 2 Emergency Intensive Care Unit (EICU) The Affiliated Hospital of Yangzhou University Yangzhou University Yangzhou China – name: 4 Jiangsu Provincial Key Laboratory of Critical Care Medicine, Department of Critical Care Medicine, Zhongda Hospital School of Medicine Southeast University Nanjing China – name: 3 Key Laboratory of Big Data Analysis and Knowledge Services of Yangzhou City Yangzhou University Yangzhou China – name: 6 Department of Emergency Medicine, Key Laboratory of Precision Medicine in Diagnosis and Monitoring Research of Zhejiang Province Sir Run Run Shaw Hospital Zhejiang University School of Medicine Hangzhou China – name: 1 State Key Laboratory of Digital Medical Engineering School of Instrument Science and Engineering Southeast University Nanjing China – name: 5 School of Biomedical Engineering and Informatics Nanjing Medical University Nanjing China
Author_xml	– sequence: 1 givenname: Meicheng orcidid: 0000-0002-8093-799X surname: Yang fullname: Yang, Meicheng – sequence: 2 givenname: Jinqiang orcidid: 0000-0002-0368-7923 surname: Zhuang fullname: Zhuang, Jinqiang – sequence: 3 givenname: Wenhan orcidid: 0009-0005-5656-3253 surname: Hu fullname: Hu, Wenhan – sequence: 4 givenname: Jianqing orcidid: 0000-0002-3524-8933 surname: Li fullname: Li, Jianqing – sequence: 5 givenname: Yu orcidid: 0000-0002-0635-4104 surname: Wang fullname: Wang, Yu – sequence: 6 givenname: Zhongheng orcidid: 0000-0002-2336-5323 surname: Zhang fullname: Zhang, Zhongheng – sequence: 7 givenname: Chengyu orcidid: 0000-0003-1965-3020 surname: Liu fullname: Liu, Chengyu – sequence: 8 givenname: Hui orcidid: 0000-0002-1344-3084 surname: Chen fullname: Chen, Hui
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/39231425$$D View this record in MEDLINE/PubMed
BookMark	eNptkltrFDEUgAep2Iv9CxIQQZGtk8smmb7Isq66UFTo6mvIZJKZlEyyJrPFvvvDzc620hHJQ27f-U4OOafFkQ9eF8U5LC8QrOi7OaEIPilOIMF8xjmDR4_Wx8VpSjdliUpSwWfFMa4QhgTNT4rfK99Jr6xvwTc5WO0HcK2dVoMNHlifN9tkE1g6662SDmyilS6BDzrZ1oNNF8Ou7YD0YLEGKx-t6vrRMUQ56PbuEixcG6Iduj7H3GoXtuO99A34IZ1t5D7R8-KpyVZ9fj-fFd8_rjbLz7Orr5_Wy8XVTBFC4YwYIisOpaRGK94oCTk2nNS1pgwR0mjDa0owojVkUiPaoIoTSpu6niPKG4PPivXB2wR5I7bR9jLeiSCtGA9CbIWMg1VOC8YUYxDDnIkRolVlKkapgQ3DZTlHPLveH1zbXd3rRuWqonQT6fTG20604VZASCBECGfD63tDDD93Og2it0lp56TXYZcEhmVZQcw5zOjLA9rK_DbrTchKtcfFgpcMMYarKlMX_6HyaHRvVW4XY_P5JODNJCAzg_41tHKXklhff5myLx7X-7fQh1bKwNsDoGJIKWojlB3G782vsE7AUuwbVYyNmulX_9APwin3B8105Jo
CitedBy_id	crossref_primary_10_1016_j_ecoenv_2025_118136 crossref_primary_10_4103_ajim_ajim_32_25 crossref_primary_10_1038_s41598_025_95604_8
Cites_doi	10.1097/CCM.0000000000004145 10.1001/jama.2020.24505 10.1164/rccm.201512-2544CP 10.1097/CCM.0000000000004000 10.1038/s41746-021-00504-6 10.1038/s41467-022-34945-8 10.1038/s41597-022-01899-x 10.1097/CCM.0000000000004550 10.1038/s41746-021-00505-5 10.1056/NEJMp2304839 10.1109/iccv.2015.172 10.1016/S2213-2600(19)30417-5 10.1016/S0140-6736(19)32989-7 10.1097/CCM.0000000000002749 10.1097/CCM.0000000000003891 10.1097/CCM.0000000000003624 10.1038/s41591-022-01843-x 10.1016/j.artmed.2021.102024 10.1001/jama.2016.0287 10.1038/sdata.2018.178 10.1016/j.artmed.2024.102785 10.1097/01.ccm.0000145229.59014.6c 10.1145/2939672.2939785 10.1007/s00134-022-06890-z 10.1038/s41591-020-0789-4 10.1109/jproc.2015.2494218 10.1001/jama.2019.5791 10.1038/s42256-019-0138-9 10.1056/NEJMoa1005372 10.1001/jamainternmed.2023.0780 10.1016/S2213-2600(20)30124-7 10.1016/S2213-2600(21)00461-6
ContentType	Journal Article
Copyright	Meicheng Yang, Jinqiang Zhuang, Wenhan Hu, Jianqing Li, Yu Wang, Zhongheng Zhang, Chengyu Liu, Hui Chen. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.09.2024. COPYRIGHT 2024 Journal of Medical Internet Research Meicheng Yang, Jinqiang Zhuang, Wenhan Hu, Jianqing Li, Yu Wang, Zhongheng Zhang, Chengyu Liu, Hui Chen. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.09.2024. 2024
Copyright_xml	– notice: Meicheng Yang, Jinqiang Zhuang, Wenhan Hu, Jianqing Li, Yu Wang, Zhongheng Zhang, Chengyu Liu, Hui Chen. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.09.2024. – notice: COPYRIGHT 2024 Journal of Medical Internet Research – notice: Meicheng Yang, Jinqiang Zhuang, Wenhan Hu, Jianqing Li, Yu Wang, Zhongheng Zhang, Chengyu Liu, Hui Chen. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.09.2024. 2024
DBID	AAYXX CITATION CGR CUY CVF ECM EIF NPM ISN 7X8 5PM DOA
DOI	10.2196/54621
DatabaseName	CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed Gale In Context: Canada MEDLINE - Academic PubMed Central (Full Participant titles) DOAJ Directory of Open Access Journals
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic MEDLINE
Database_xml	– sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Medicine Library & Information Science
EISSN	1438-8871
ExternalDocumentID	oai_doaj_org_article_77c77131fec744ec9f9766f1d7300528 PMC11411223 A807277399 39231425 10_2196_54621
Genre	Validation Study Research Support, Non-U.S. Gov't Journal Article
GroupedDBID	--- .4I .DC 29L 2WC 36B 53G 5GY 5VS 77I 77K 7RV 7X7 8FI 8FJ AAFWJ AAKPC AAWTL AAYXX ABDBF ABIVO ABUWG ACGFO ADBBV AEGXH AENEX AFFHD AFKRA AFPKN AIAGR ALMA_UNASSIGNED_HOLDINGS ALSLI AOIJS BAWUL BCNDV BENPR CCPQU CITATION CNYFK CS3 DIK DU5 DWQXO E3Z EAP EBD EBS EJD ELW EMB EMOBN ESX F5P FRP FYUFA GROUPED_DOAJ GX1 HMCUK HYE IAO ICO IEA IHR INH ISN ITC KQ8 M1O M48 NAPCQ OK1 OVT P2P PGMZT PHGZM PHGZT PIMPY PPXIY PQQKQ PRQQA RNS RPM SJN SV3 TR2 UKHRP XSB CGR CUY CVF ECM EIF NPM PUEGO 7X8 5PM
ID	FETCH-LOGICAL-c4461-4f4a981aa6fec8dca183f84bbe67244def8b64326b17ae26d298466dbb5268df3
IEDL.DBID	DOA
ISSN	1438-8871 1439-4456
IngestDate	Fri Oct 03 12:37:52 EDT 2025 Tue Nov 04 02:04:58 EST 2025 Sat Nov 01 15:05:14 EDT 2025 Sat Nov 29 13:57:56 EST 2025 Sat Nov 29 10:37:51 EST 2025 Wed Nov 26 11:31:02 EST 2025 Tue Sep 09 02:33:17 EDT 2025 Tue Nov 18 22:29:49 EST 2025 Sat Nov 29 03:22:53 EST 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	8
Keywords	disease progression trajectories conformal prediction predictive modeling artificial intelligence enrichment strategy sepsis
Language	English
License	Meicheng Yang, Jinqiang Zhuang, Wenhan Hu, Jianqing Li, Yu Wang, Zhongheng Zhang, Chengyu Liu, Hui Chen. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.09.2024. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research (ISSN 1438-8871), is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included.
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c4461-4f4a981aa6fec8dca183f84bbe67244def8b64326b17ae26d298466dbb5268df3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 ObjectType-Undefined-3
ORCID	0000-0002-0635-4104 0000-0003-1965-3020 0000-0002-0368-7923 0000-0002-8093-799X 0000-0002-1344-3084 0000-0002-3524-8933 0000-0002-2336-5323 0009-0005-5656-3253
OpenAccessLink	https://doaj.org/article/77c77131fec744ec9f9766f1d7300528
PMID	39231425
PQID	3100913881
PQPubID	23479
ParticipantIDs	doaj_primary_oai_doaj_org_article_77c77131fec744ec9f9766f1d7300528 pubmedcentral_primary_oai_pubmedcentral_nih_gov_11411223 proquest_miscellaneous_3100913881 gale_infotracmisc_A807277399 gale_infotracacademiconefile_A807277399 gale_incontextgauss_ISN_A807277399 pubmed_primary_39231425 crossref_citationtrail_10_2196_54621 crossref_primary_10_2196_54621
PublicationCentury	2000
PublicationDate	20240904
PublicationDateYYYYMMDD	2024-09-04
PublicationDate_xml	– month: 9 year: 2024 text: 20240904 day: 4
PublicationDecade	2020
PublicationPlace	Canada
PublicationPlace_xml	– name: Canada – name: Toronto, Canada
PublicationTitle	Journal of medical Internet research
PublicationTitleAlternate	J Med Internet Res
PublicationYear	2024
Publisher	Journal of Medical Internet Research JMIR Publications
Publisher_xml	– name: Journal of Medical Internet Research – name: JMIR Publications
References	ref13 ref35 ref12 ref34 ref15 ref14 ref31 ref30 ref11 ref33 ref10 ref32 ref2 ref1 ref17 ref16 ref19 ref18 ref24 ref23 ref26 ref25 ref20 ref22 ref21 ref28 ref27 ref29 ref8 ref7 ref9 ref4 ref3 ref6 ref5
References_xml	– ident: ref16 doi: 10.1097/CCM.0000000000004145 – ident: ref29 doi: 10.1001/jama.2020.24505 – ident: ref28 doi: 10.1164/rccm.201512-2544CP – ident: ref4 doi: 10.1097/CCM.0000000000004000 – ident: ref30 doi: 10.1038/s41746-021-00504-6 – ident: ref13 doi: 10.1038/s41467-022-34945-8 – ident: ref14 doi: 10.1038/s41597-022-01899-x – ident: ref21 doi: 10.1097/CCM.0000000000004550 – ident: ref11 doi: 10.1038/s41746-021-00505-5 – ident: ref32 doi: 10.1056/NEJMp2304839 – ident: ref18 doi: 10.1109/iccv.2015.172 – ident: ref7 doi: 10.1016/S2213-2600(19)30417-5 – ident: ref2 doi: 10.1016/S0140-6736(19)32989-7 – ident: ref26 doi: 10.1097/CCM.0000000000002749 – ident: ref20 doi: 10.1097/CCM.0000000000003891 – ident: ref34 – ident: ref8 doi: 10.1097/CCM.0000000000003624 – ident: ref5 doi: 10.1038/s41591-022-01843-x – ident: ref22 doi: 10.1016/j.artmed.2021.102024 – ident: ref1 doi: 10.1001/jama.2016.0287 – ident: ref15 doi: 10.1038/sdata.2018.178 – ident: ref31 doi: 10.1016/j.artmed.2024.102785 – ident: ref27 doi: 10.1097/01.ccm.0000145229.59014.6c – ident: ref17 doi: 10.1145/2939672.2939785 – ident: ref10 doi: 10.1007/s00134-022-06890-z – ident: ref19 doi: 10.1038/s41591-020-0789-4 – ident: ref24 doi: 10.1109/jproc.2015.2494218 – ident: ref9 doi: 10.1001/jama.2019.5791 – ident: ref23 doi: 10.1038/s42256-019-0138-9 – ident: ref35 – ident: ref6 doi: 10.1056/NEJMoa1005372 – ident: ref25 doi: 10.1001/jamainternmed.2023.0780 – ident: ref33 – ident: ref3 doi: 10.1016/S2213-2600(20)30124-7 – ident: ref12 doi: 10.1016/S2213-2600(21)00461-6
SSID	ssj0020491
Score	2.4317203
Snippet	Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial... Background Sepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials.... BackgroundSepsis is a heterogeneous syndrome, and enrollment of more homogeneous patients is essential to improve the efficiency of clinical trials. Artificial...
SourceID	doaj pubmedcentral proquest gale pubmed crossref
SourceType	Open Website Open Access Repository Aggregation Database Index Database Enrichment Source
StartPage	e54621
SubjectTerms	Aged Algorithms Artificial Intelligence Clinical trials Clinical Trials as Topic - methods Data mining Decision-making Female Humans Infection Machine learning Male Medical centers Medical research Medicine, Experimental Middle Aged Online databases Original Paper Patient Selection Retrospective Studies Sepsis - therapy
Title	Enhancing Patient Selection in Sepsis Clinical Trials Design Through an AI Enrichment Strategy: Algorithm Development and Validation
URI	https://www.ncbi.nlm.nih.gov/pubmed/39231425 https://www.proquest.com/docview/3100913881 https://pubmed.ncbi.nlm.nih.gov/PMC11411223 https://doaj.org/article/77c77131fec744ec9f9766f1d7300528
Volume	26
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 1438-8871 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020491 issn: 1438-8871 databaseCode: DOA dateStart: 19990101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVPQU databaseName: Health & Medical Collection customDbUrl: eissn: 1438-8871 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020491 issn: 1438-8871 databaseCode: 7X7 dateStart: 20010101 isFulltext: true titleUrlDefault: https://search.proquest.com/healthcomplete providerName: ProQuest – providerCode: PRVPQU databaseName: Library Science Database customDbUrl: eissn: 1438-8871 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020491 issn: 1438-8871 databaseCode: M1O dateStart: 20010101 isFulltext: true titleUrlDefault: https://search.proquest.com/libraryscience providerName: ProQuest – providerCode: PRVPQU databaseName: Nursing & Allied Health Database customDbUrl: eissn: 1438-8871 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020491 issn: 1438-8871 databaseCode: 7RV dateStart: 20010101 isFulltext: true titleUrlDefault: https://search.proquest.com/nahs providerName: ProQuest – providerCode: PRVPQU databaseName: ProQuest Central customDbUrl: eissn: 1438-8871 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020491 issn: 1438-8871 databaseCode: BENPR dateStart: 20010101 isFulltext: true titleUrlDefault: https://www.proquest.com/central providerName: ProQuest – providerCode: PRVPQU databaseName: Publicly Available Content Database customDbUrl: eissn: 1438-8871 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0020491 issn: 1438-8871 databaseCode: PIMPY dateStart: 20010101 isFulltext: true titleUrlDefault: http://search.proquest.com/publiccontent providerName: ProQuest
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3Pb9MwFH6CgSYkhGD8KmyVQROcqi2uGzvcOuhEJVqirVTlFNmO3Vbq0qlpkbjzh_PsuFUDBy5cfIifo9jv2f6e8_w9gFOWUGupYi2ed0SLKekOmqRqiQ4VWndEHvkDt_EXPhyKySRJ91J9uZiwih64GrgzzjVHRyqyRnPGjE4sbqCxjXJPtE79Nd9znmydqeBqIe6NDuGhC3RGEzvrsJhGtZ3HE_T_vQzv7UP1GMm9TefyMTwKaJF0q698AndMcQQn4a4BeUfCZSI3uCTM0iM4HIT_5U_hV6-YOUKNYkrSikCVXPvEN67BHNuY23JekkAOuiAjb47kkw_rIKMqhw-RBen2Sa_AJXN2499RUdr-_EC6i-lyNV_Pbshe9BE2yMkYAX6Vr-kZfLvsjT5-boW8Cy2NziG6lJbJBDUmYxx0kWuJ094KppSJOaKB3FihEMjQWEVcGhrnNEEUE-dKOe6Y3Lafw0GxLMxLINaYSCGkUFobxqhUMXrjVDJmhWPSpw043eok04GU3OXGWGTonDjVZV51DWjuxG4rFo4_BS6cQneVjjTbP0BTyoIpZf8ypQa8deaQOVqMwsXdTOWmLLP-9TDrinMEehzRXAPeByG7xC_VMlxjwP46Jq2a5HFNEuetrlW_2Vpd5qpcsFthlpsyc_9ckqgtBHbrRWWFu461HSDHdbYBomaftZ7Xa4r5zNOGo-eL4Jq2X_2PsXoNDyjCOx9tx47hYL3amBO4r3-s5-WqCXf51diVE-5L0YR7F71hetX00xTLQfQVn6X9Qfr9N8r5QU0
linkProvider	Directory of Open Access Journals
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Enhancing+Patient+Selection+in+Sepsis+Clinical+Trials+Design+Through+an+AI+Enrichment+Strategy%3A+Algorithm+Development+and+Validation&rft.jtitle=Journal+of+medical+Internet+research&rft.au=Yang%2C+Meicheng&rft.au=Zhuang%2C+Jinqiang&rft.au=Hu%2C+Wenhan&rft.au=Li%2C+Jianqing&rft.date=2024-09-04&rft.eissn=1438-8871&rft.volume=26&rft.spage=e54621&rft_id=info:doi/10.2196%2F54621&rft_id=info%3Apmid%2F39231425&rft.externalDocID=39231425
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1438-8871&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1438-8871&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1438-8871&client=summon