Jingtao Ma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jingtao Ma has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
Jing Liu
The Fourth Hospital of Hebei Medical University
6 shared publicationsTiezhu Yao
The Fourth Hospital of Hebei Medical University
6 shared publicationsGuang Liu
Department of Cardiology, The Fourth Hospital of Hebei Medical University., Shijiazhuang, Hebei, China
5 shared publicationsLing Guo
Department of Cardiology, The Fourth Hospital of Hebei Medical University., Shijiazhuang, Hebei, China
5 shared publicationsZhengkun Guan
Department of Cardiology, The Fourth Hospital of Hebei Medical University., Shijiazhuang, Hebei, China
4 shared publicationsZhenli Li
Department of Cardiology, The Fourth Hospital of Hebei Medical University., Shijiazhuang, Hebei, China
4 shared publications
Latest Publications
- Open Access
- 27 February 2026
- Published by Wolters Kluwer Health in Medicine
- Vol. 105 (9) , e47916
Explore the potential mediating role of plasma metabolites and cytokines in the causal relationship between gut microbiota and the severity of immune-related adverse events: A Mendelian randomization study
The gut microbiota, particularly genus Ruminiclostridium6 , has been implicated in modulating immune-related adverse events (irAEs) associated with immune checkpoint inhibitor therapy. However, the mediating roles of circulating metabolites and cytokines in this relationship remain poorly understood. We performed a two-sample Mendelian randomization study to investigate causal links between genus Ruminiclostridium6 , 19 plasma metabolites, 41 cytokines, and high-grade irAEs. Genetic instruments were selected from publicly available genome-wide association studies summary statistics. Inverse variance weighted method served as the primary analysis, supplemented by sensitivity and mediation analyses. No robust causal relationships were found between genus Ruminiclostridium6 and the 19 metabolites. However, high-grade irAEs were causally linked to decreased oleoylcholine levels. Among cytokines, only IL-2RA showed a causal association with high-grade irAEs, but not with genus Ruminiclostridium6 . Consequently, two-step Mendelian randomization did not support a mediating role for the studied metabolites or cytokines in the genus Ruminiclostridium6 –irAEs pathway. Enrichment analysis highlighted glycine, serine, threonine, arginine, and proline metabolism as potential pathways of interest. This study did not support 19 plasma metabolites or 41 cytokines as major mediators of the protective effect of genus Ruminiclostridium6 against high-grade irAEs. The mechanism may involve local gut-level immunomodulation or microbial metabolites not captured in this study. These findings underscore the complexity of microbiota–irAEs interactions and highlight specific metabolic pathways for further investigation.
- 27 February 2026
- Published by Wolters Kluwer Health in Medicine
- Vol. 105 (9) , e47916
Open AccessExplore the potential mediating role of plasma metabolites and cytokines in the causal relationship between gut microbiota and the severity of immune-related adverse events: A Mendelian randomization study
The gut microbiota, particularly genus Ruminiclostridium6 , has been implicated in modulating immune-related adverse events (irAEs) associated with immune checkpoint inhibitor therapy. However, the mediating roles of circulating metabolites and cytokines in this relationship remain poorly understood. We performed a two-sample Mendelian randomization study to investigate causal links between genus Ruminiclostridium6 , 19 plasma metabolites, 41 cytokines, and high-grade irAEs. Genetic instruments were selected from publicly available genome-wide association studies summary statistics. Inverse variance weighted method served as the primary analysis, supplemented by sensitivity and mediation analyses. No robust causal relationships were found between genus Ruminiclostridium6 and the 19 metabolites. However, high-grade irAEs were causally linked to decreased oleoylcholine levels. Among cytokines, only IL-2RA showed a causal association with high-grade irAEs, but not with genus Ruminiclostridium6 . Consequently, two-step Mendelian randomization did not support a mediating role for the studied metabolites or cytokines in the genus Ruminiclostridium6 –irAEs pathway. Enrichment analysis highlighted glycine, serine, threonine, arginine, and proline metabolism as potential pathways of interest. This study did not support 19 plasma metabolites or 41 cytokines as major mediators of the protective effect of genus Ruminiclostridium6 against high-grade irAEs. The mechanism may involve local gut-level immunomodulation or microbial metabolites not captured in this study. These findings underscore the complexity of microbiota–irAEs interactions and highlight specific metabolic pathways for further investigation.
- Open Access
- 1 January 2026
- Published by SAGE Publications in Clinical and Applied Thrombosis/Hemostasis
- Vol. 32
Explore the Causal Effect of SGLT2 Inhibition on Venous Thromboembolism Events: A Drug-Target Mendelian Randomization Study With Pharmacovigilance Analysis
Background: Sodium-glucose cotransporter 2 (SGLT2) inhibitors are widely used antidiabetic agents with established cardiorenal benefits, yet their genetic causal association with venous thromboembolism (VTE) remains uncertain. Methods: We employed a multifaceted approach to investigate the causal relationship between SGLT2 inhibition and VTE, as well as deep vein thrombosis (DVT) and pulmonary embolism (PE). Drug-target Mendelian randomization (MR) and summary-data-based MR (SMR) analyses were conducted using genetic variants proxying SGLT2 inhibition, derived from SLC5A2 expression quantitative trait loci and glycemic traits. Positive control analyses confirmed validity using type 2 diabetes outcomes. Meta-analyses were performed across multiple independent genome-wide association study datasets. Complementary pharmacovigilance analysis was conducted using the FDA Adverse Event Reporting System (FAERS) to assess disproportionality of SGLT2 inhibitor-related VTE reports. Results: Genetically proxied SGLT2 inhibition was significantly associated with reduced risk of type 2 diabetes (P < 0.05), validating its robustness. Most MR analyses showed no consistent causal association with VTE, DVT, or PE across datasets, and meta-analyses yielded non-significant pooled estimates (all P > 0.05). SMR analyses revealed no association between SLC5A2 expression and VTE-related outcomes (all P_SMR > 0.05). FAERS disproportionality analysis also identified no safety signals for PE (reporting odds ratio [ROR] = 0.51; 95% CI: 0.42–0.64) or DVT (ROR = 0.54; 95% CI: 0.45–0.64). Conclusions: This integrative study provides consistent evidence that SGLT2 inhibition is not causally associated with VTE, DVT, or PE, supporting its thrombotic safety. However, generalizability to non-European populations requires future validation in diverse cohorts.
- 1 January 2026
- Published by SAGE Publications in Clinical and Applied Thrombosis/Hemostasis
- Vol. 32
Open AccessExplore the Causal Effect of SGLT2 Inhibition on Venous Thromboembolism Events: A Drug-Target Mendelian Randomization Study With Pharmacovigilance Analysis
Background: Sodium-glucose cotransporter 2 (SGLT2) inhibitors are widely used antidiabetic agents with established cardiorenal benefits, yet their genetic causal association with venous thromboembolism (VTE) remains uncertain. Methods: We employed a multifaceted approach to investigate the causal relationship between SGLT2 inhibition and VTE, as well as deep vein thrombosis (DVT) and pulmonary embolism (PE). Drug-target Mendelian randomization (MR) and summary-data-based MR (SMR) analyses were conducted using genetic variants proxying SGLT2 inhibition, derived from SLC5A2 expression quantitative trait loci and glycemic traits. Positive control analyses confirmed validity using type 2 diabetes outcomes. Meta-analyses were performed across multiple independent genome-wide association study datasets. Complementary pharmacovigilance analysis was conducted using the FDA Adverse Event Reporting System (FAERS) to assess disproportionality of SGLT2 inhibitor-related VTE reports. Results: Genetically proxied SGLT2 inhibition was significantly associated with reduced risk of type 2 diabetes (P < 0.05), validating its robustness. Most MR analyses showed no consistent causal association with VTE, DVT, or PE across datasets, and meta-analyses yielded non-significant pooled estimates (all P > 0.05). SMR analyses revealed no association between SLC5A2 expression and VTE-related outcomes (all P_SMR > 0.05). FAERS disproportionality analysis also identified no safety signals for PE (reporting odds ratio [ROR] = 0.51; 95% CI: 0.42–0.64) or DVT (ROR = 0.54; 95% CI: 0.45–0.64). Conclusions: This integrative study provides consistent evidence that SGLT2 inhibition is not causally associated with VTE, DVT, or PE, supporting its thrombotic safety. However, generalizability to non-European populations requires future validation in diverse cohorts.
- Open Access
- 7 November 2025
- Published by Wolters Kluwer Health in Medicine
- Vol. 104 (45) , e45714
Explore the relationships between gut microbiota and cardiomyopathies with the potential mediating effect of cytokines: A univariable and multivariable Mendelian randomization study
Emerging evidence suggests a potential link between gut microbiota and cardiomyopathies, but the causal relationships and the mediating role of cytokines remain unclear. We performed a two-sample Mendelian randomization (MR) analysis using genome-wide association study (GWAS) summary data for 211 gut microbiota taxa, 41 cytokines, and cardiomyopathy subtypes. Instrumental variables were selected at a genome-wide significance threshold ( P < 1 × 10 −5 ). Primary causal estimates were derived via inverse-variance weighted (IVW) regression, supplemented by sensitivity analyses. Multivariable MR was utilized to reduce bias from confounders. Mediation analysis tested whether cytokines linked gut microbiota to cardiomyopathies. Genetic liability in gut microbiota showed significant associations with cardiomyopathy risk. For dilated cardiomyopathy (DCM), genus Ruminococcaceae (OR = 1.24, 95% CI: 1.00–1.54, P = .045) and genus Flavonifractor (OR = 1.66, 95% CI: 1.15–2.35, P = .0067) increased risk, while family Peptostreptococcaceae was protective (OR = 0.71, 95% CI: 0.51–0.99, P = .041). For hypertrophic cardiomyopathy (HCM), genus Enterorhabdus raised risk (OR = 1.53, 95% CI: 1.00–2.33, P = .049), whereas order Methanobacteria lowered it (OR = 0.69, 95% CI: 0.52–0.93, P = .013). For myocarditis, genus Faecalibacterium reduced risk (OR = 0.53, 95% CI: 0.33–0.85, P = .009), but genus Family XIII UCG001 increased it (OR = 2.06, 95% CI: 1.15–3.69, P = .015). Cytokines including IL-9 and MIP-1β showed protective effects for DCM, while pro-inflammatory cytokines TNF-α and VEGF elevated myocarditis risk. Mediation analysis revealed cytokines did not mediate gut microbiota-cardiomyopathy pathways. Specific gut microbiota and cytokines exert causal effects on cardiomyopathy subtypes, independent of cytokine mediation. These findings highlight potential therapeutic targets and underscore the need to explore alternative mechanistic pathways.
- 7 November 2025
- Published by Wolters Kluwer Health in Medicine
- Vol. 104 (45) , e45714
Open AccessExplore the relationships between gut microbiota and cardiomyopathies with the potential mediating effect of cytokines: A univariable and multivariable Mendelian randomization study
Emerging evidence suggests a potential link between gut microbiota and cardiomyopathies, but the causal relationships and the mediating role of cytokines remain unclear. We performed a two-sample Mendelian randomization (MR) analysis using genome-wide association study (GWAS) summary data for 211 gut microbiota taxa, 41 cytokines, and cardiomyopathy subtypes. Instrumental variables were selected at a genome-wide significance threshold ( P < 1 × 10 −5 ). Primary causal estimates were derived via inverse-variance weighted (IVW) regression, supplemented by sensitivity analyses. Multivariable MR was utilized to reduce bias from confounders. Mediation analysis tested whether cytokines linked gut microbiota to cardiomyopathies. Genetic liability in gut microbiota showed significant associations with cardiomyopathy risk. For dilated cardiomyopathy (DCM), genus Ruminococcaceae (OR = 1.24, 95% CI: 1.00–1.54, P = .045) and genus Flavonifractor (OR = 1.66, 95% CI: 1.15–2.35, P = .0067) increased risk, while family Peptostreptococcaceae was protective (OR = 0.71, 95% CI: 0.51–0.99, P = .041). For hypertrophic cardiomyopathy (HCM), genus Enterorhabdus raised risk (OR = 1.53, 95% CI: 1.00–2.33, P = .049), whereas order Methanobacteria lowered it (OR = 0.69, 95% CI: 0.52–0.93, P = .013). For myocarditis, genus Faecalibacterium reduced risk (OR = 0.53, 95% CI: 0.33–0.85, P = .009), but genus Family XIII UCG001 increased it (OR = 2.06, 95% CI: 1.15–3.69, P = .015). Cytokines including IL-9 and MIP-1β showed protective effects for DCM, while pro-inflammatory cytokines TNF-α and VEGF elevated myocarditis risk. Mediation analysis revealed cytokines did not mediate gut microbiota-cardiomyopathy pathways. Specific gut microbiota and cytokines exert causal effects on cardiomyopathy subtypes, independent of cytokine mediation. These findings highlight potential therapeutic targets and underscore the need to explore alternative mechanistic pathways.
- Open Access
- 9 October 2025
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 15, 1676806
Blood-derived ratio indexes associated with severity and prognosis of immune checkpoint inhibitor-related cardiotoxicity: a retrospective analysis
Background: Immune checkpoint inhibitors (ICIs) have transformed cancer therapy but may cause immune checkpoint inhibitor-related cardiotoxicities (iRCs). Blood-derived inflammatory ratios may serve as practical prognostic tools for these life-threatening complications.Methods: We conducted a retrospective study of 105 iRC patients treated with ICIs between 2020 and 2023. Patients were classified by Common Terminology Criteria for Adverse Events (CTCAE) severity grades. We identified the most important blood-derived ratio indexes at iRC diagnosis associated with the severity of iRCs, 40-day major adverse cardiovascular events (MACEs), and long-term mortality using receiver operating characteristic (ROC) analyses and restricted cubic spline (RCS) curves. Kaplan–Meier survival curves, Cox regression, and subgroup analysis were also performed to evaluate them.Results: High-grade iRCs patients (n=40) showed a significantly higher system inflammation response index (SIRI) (8.21 vs. 2.21, p3.07 was strongly associated with 40-day MACEs [hazard ratios (HR)=6.56, p7.88 was associated with increased long-term mortality risk (HR = 2.33, p=0.003). Both SIRI and NLR remained significant after adjusting for cardiac biomarkers and clinical variables and were found associated with iRC severity-related cardiac biomarkers.Conclusion: SIRI and NLR are effectively associated with iRC severity and could stratify the risk of developing poor short- and long-term prognoses. These readily available inflammatory indexes could improve risk assessment and guide clinical decision-making for iRC patients. However, further prospective studies should validate their utility in diverse populations.
- 9 October 2025
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 15, 1676806
Open AccessBlood-derived ratio indexes associated with severity and prognosis of immune checkpoint inhibitor-related cardiotoxicity: a retrospective analysis
Background: Immune checkpoint inhibitors (ICIs) have transformed cancer therapy but may cause immune checkpoint inhibitor-related cardiotoxicities (iRCs). Blood-derived inflammatory ratios may serve as practical prognostic tools for these life-threatening complications.Methods: We conducted a retrospective study of 105 iRC patients treated with ICIs between 2020 and 2023. Patients were classified by Common Terminology Criteria for Adverse Events (CTCAE) severity grades. We identified the most important blood-derived ratio indexes at iRC diagnosis associated with the severity of iRCs, 40-day major adverse cardiovascular events (MACEs), and long-term mortality using receiver operating characteristic (ROC) analyses and restricted cubic spline (RCS) curves. Kaplan–Meier survival curves, Cox regression, and subgroup analysis were also performed to evaluate them.Results: High-grade iRCs patients (n=40) showed a significantly higher system inflammation response index (SIRI) (8.21 vs. 2.21, p3.07 was strongly associated with 40-day MACEs [hazard ratios (HR)=6.56, p7.88 was associated with increased long-term mortality risk (HR = 2.33, p=0.003). Both SIRI and NLR remained significant after adjusting for cardiac biomarkers and clinical variables and were found associated with iRC severity-related cardiac biomarkers.Conclusion: SIRI and NLR are effectively associated with iRC severity and could stratify the risk of developing poor short- and long-term prognoses. These readily available inflammatory indexes could improve risk assessment and guide clinical decision-making for iRC patients. However, further prospective studies should validate their utility in diverse populations.
- Open Access
- 23 July 2025
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 12, 1605437
Development and validation of a nomogram for predicting in-hospital mortality in older adult hip fracture patients with atrial fibrillation: a retrospective study
Background: Hip fracture is prevalent among older adult patients, which often results in intensive care unit (ICU) admission. When complicated with atrial fibrillation (AF), older adult patients with hip fractures were observed to have a high short-term mortality. However, few studies have focused specifically on such a cohort. This study aimed to develop and validate a nomogram to evaluate the in-hospital mortality risk of such a group in the ICU.Methods: We enrolled older adult patients with hip fractures complicated by AF in the Medical Information Mart for Intensive Care Database (MIMIC). Logistic regression (LR) and Least Absolute Shrinkage and Selection Operator (LASSO) algorithms were employed to screen features. We further used Extreme Gradient Boosting (XGBoost) based on features selected by LR and LASSO algorithms to assist in identifying the final model-established features. An Electronic Intensive Care Unit Collaborative Research Database (eICU-CRD) was utilized for external validation. The area under curves (AUC), calibration curves, Delong test, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were performed to evaluate the prediction performance. Ultimately, a visualized nomogram was constructed to provide convenient access for clinicians to evaluate mortality risk.Results: A total of 308 patients were enrolled in this study. We employed LR and LASSO algorithms to initially screen out 15 and 20 features, respectively. Next, 10 features, which were the intersection of features selected by the above methods, were further utilized to develop an XGBoost model to obtain the rank of feature importance. Finally, eight features were ultimately selected to develop a nomogram by comparing the AUCs of LR models originating from a “feature-adding by the feature rank” strategy. The nomogram exhibited superior predictive performance (AUC:0.834) than conventional scoring systems in the training set, with an AUC of 0.715 in external validation.Conclusion: Our study constructed a predictive model based on features selected by machine learning approaches to evaluate the in-hospital mortality risk of critically ill patients with hip fractures combined with AF. An accessible nomogram was offered to facilitate clinical decision-making.
- 23 July 2025
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 12, 1605437
Open AccessDevelopment and validation of a nomogram for predicting in-hospital mortality in older adult hip fracture patients with atrial fibrillation: a retrospective study
Background: Hip fracture is prevalent among older adult patients, which often results in intensive care unit (ICU) admission. When complicated with atrial fibrillation (AF), older adult patients with hip fractures were observed to have a high short-term mortality. However, few studies have focused specifically on such a cohort. This study aimed to develop and validate a nomogram to evaluate the in-hospital mortality risk of such a group in the ICU.Methods: We enrolled older adult patients with hip fractures complicated by AF in the Medical Information Mart for Intensive Care Database (MIMIC). Logistic regression (LR) and Least Absolute Shrinkage and Selection Operator (LASSO) algorithms were employed to screen features. We further used Extreme Gradient Boosting (XGBoost) based on features selected by LR and LASSO algorithms to assist in identifying the final model-established features. An Electronic Intensive Care Unit Collaborative Research Database (eICU-CRD) was utilized for external validation. The area under curves (AUC), calibration curves, Delong test, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were performed to evaluate the prediction performance. Ultimately, a visualized nomogram was constructed to provide convenient access for clinicians to evaluate mortality risk.Results: A total of 308 patients were enrolled in this study. We employed LR and LASSO algorithms to initially screen out 15 and 20 features, respectively. Next, 10 features, which were the intersection of features selected by the above methods, were further utilized to develop an XGBoost model to obtain the rank of feature importance. Finally, eight features were ultimately selected to develop a nomogram by comparing the AUCs of LR models originating from a “feature-adding by the feature rank” strategy. The nomogram exhibited superior predictive performance (AUC:0.834) than conventional scoring systems in the training set, with an AUC of 0.715 in external validation.Conclusion: Our study constructed a predictive model based on features selected by machine learning approaches to evaluate the in-hospital mortality risk of critically ill patients with hip fractures combined with AF. An accessible nomogram was offered to facilitate clinical decision-making.
- Open Access
- 11 February 2025
- Published by Springer Nature in Journal of Cancer Research and Clinical Oncology
- Vol. 151 (2) , 1
Correction: Nomograms based on ratio indexes to predict severity and prognosis in immune checkpoint inhibitors-related myocarditis: a retrospective analysis
- 11 February 2025
- Published by Springer Nature in Journal of Cancer Research and Clinical Oncology
- Vol. 151 (2) , 1
Open AccessCorrection: Nomograms based on ratio indexes to predict severity and prognosis in immune checkpoint inhibitors-related myocarditis: a retrospective analysis
- 4 January 2025
- Published by Elsevier in Journal of Hazardous Materials
- Vol. 486, 137114
Co-exposure to polystyrene nanoplastics and F-53B induces vascular endothelial cell pyroptosis through the NF-κB/NLRP3 pathway
- 4 January 2025
- Published by Elsevier in Journal of Hazardous Materials
- Vol. 486, 137114
Co-exposure to polystyrene nanoplastics and F-53B induces vascular endothelial cell pyroptosis through the NF-κB/NLRP3 pathway
- Open Access
- 27 May 2024
- Published by Springer Nature in Journal of Cancer Research and Clinical Oncology
- Vol. 150 (5) , 277
Nomograms based on ratio indexes to predict severity and prognosis in immune checkpoint inhibitors-related myocarditis: a retrospective analysis
Purpose: Immune checkpoint inhibitors-related myocarditis (ICI-M) is one of the immune-related adverse events (irAEs), which is rare and highly lethal. This study aimed to establish nomograms based on ratio biomarkers to predict the severity and prognosis of ICI-M. Methods: We retrospectively examined patients with advanced cancers who were also diagnosed with ICI-M at the Fourth Hospital of Hebei Medical University. The patients of ICI-M were divided into mild and severe groups and a 40-day following up was carried out. The major adverse cardiovascular events(MACEs) were regarded as the endpoint. Nomogram-based models were established and validated. Results: Seventy-seven patients were involved, including 31 severe cases(40.3%). Lactate dehydrogenase-to-albumin ratio(LAR) combined with the change rate from baseline to onset of LAR( $$\triangle$$ LAR) which performed best to diagnose the severe ICI-M was identified to establish the nomogram-based model. The bootstrap-corrected concordance index [0.752 95% confidence interval (CI): 0.635 $$-$$ 0.866] and calibration plot with good degree of fitting confirmed this diagnostic model. Neutrophil-to-high-density lipoprotein cholesterol ratio(NHR) and LAR were also screened into the nomogram-based model for 40-day MACEs after ICI-M, which performed well by validating for concordance index(0.779 95% CI: 0.677 $$-$$ 0.865)and calibration plots after being bootstrap-corrected. Moreover, a $$\ge$$ 101% increase in LAR significantly separated patients in MACE-free survival. Conclusion: Ratio indexes at onset and their change rates from baseline showed good diagnostic value for the severity of ICI-M and prognostic value for subsequent MACEs, particularly LAR, NHR and their change rates. The nomogram-based models of ratio indexes could provide a potential choice for early detection and monitor of the severe ICI-M and subsequent MACEs. Graphical- 27 May 2024
- Published by Springer Nature in Journal of Cancer Research and Clinical Oncology
- Vol. 150 (5) , 277
Open AccessNomograms based on ratio indexes to predict severity and prognosis in immune checkpoint inhibitors-related myocarditis: a retrospective analysis
Purpose: Immune checkpoint inhibitors-related myocarditis (ICI-M) is one of the immune-related adverse events (irAEs), which is rare and highly lethal. This study aimed to establish nomograms based on ratio biomarkers to predict the severity and prognosis of ICI-M. Methods: We retrospectively examined patients with advanced cancers who were also diagnosed with ICI-M at the Fourth Hospital of Hebei Medical University. The patients of ICI-M were divided into mild and severe groups and a 40-day following up was carried out. The major adverse cardiovascular events(MACEs) were regarded as the endpoint. Nomogram-based models were established and validated. Results: Seventy-seven patients were involved, including 31 severe cases(40.3%). Lactate dehydrogenase-to-albumin ratio(LAR) combined with the change rate from baseline to onset of LAR( $$\triangle$$ LAR) which performed best to diagnose the severe ICI-M was identified to establish the nomogram-based model. The bootstrap-corrected concordance index [0.752 95% confidence interval (CI): 0.635 $$-$$ 0.866] and calibration plot with good degree of fitting confirmed this diagnostic model. Neutrophil-to-high-density lipoprotein cholesterol ratio(NHR) and LAR were also screened into the nomogram-based model for 40-day MACEs after ICI-M, which performed well by validating for concordance index(0.779 95% CI: 0.677 $$-$$ 0.865)and calibration plots after being bootstrap-corrected. Moreover, a $$\ge$$ 101% increase in LAR significantly separated patients in MACE-free survival. Conclusion: Ratio indexes at onset and their change rates from baseline showed good diagnostic value for the severity of ICI-M and prognostic value for subsequent MACEs, particularly LAR, NHR and their change rates. The nomogram-based models of ratio indexes could provide a potential choice for early detection and monitor of the severe ICI-M and subsequent MACEs. Graphical - 9 May 2024
- Published by Elsevier in Toxicology Letters
- Vol. 397, 23-33
Transcriptomics coupled with proteomics reveals osimertinib-induced myocardial mitochondrial dysfunction
- 9 May 2024
- Published by Elsevier in Toxicology Letters
- Vol. 397, 23-33
Transcriptomics coupled with proteomics reveals osimertinib-induced myocardial mitochondrial dysfunction
- Open Access
- 1 January 2024
- Published by SAGE Publications in Clinical and Applied Thrombosis/Hemostasis
- Vol. 30
Interpretable Machine Learning Approach for Predicting 30-Day Mortality of Critical Ill Patients with Pulmonary Embolism and Heart Failure: A Retrospective Study
Background: Pulmonary embolism (PE) patients combined with heart failure (HF) have been reported to have a high short-term mortality. However, few studies have developed predictive tools of 30-day mortality for these patients in intensive care unit (ICU). This study aimed to construct and validate a machine learning (ML) model to predict 30-day mortality for PE patients combined with HF in ICU. Methods: We enrolled patients with PE combined with HF in the Medical Information Mart for Intensive Care Database (MIMIC) and developed six ML models after feature selection. Further, eICU Collaborative Research Database (eICU-CRD) was utilized for external vali- dation. The area under curves (AUC), calibration curves, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were performed to evaluate the prediction performance. Shapley additive explanation (SHAP) was performed to enhance the interpretability of our models. Results: A total of 472 PE patients combined with HF were included. We developed six ML models by the 13 selected features. After internal validation, the Support Vector Ma- chine (SVM) model performed best with an AUC of 0.835, a superior calibration degree, and a wider risk threshold (from 0% to 90%) for obtaining clinical benefit, which also outperformed traditional mortality risk evaluation systems,as evaluated by NRI and IDI. The SVM model was still reliable after external validation. SHAP was performed to explain the model. Moreover, an online application was developed for further clinical use. Conclusion: This study developed a potential tool for identify short-term mortality risk to guide clinical decision making for PE patients combined with HF in the ICU. The SHAP method also helped clinicians to better understand the model.
- 1 January 2024
- Published by SAGE Publications in Clinical and Applied Thrombosis/Hemostasis
- Vol. 30
Open AccessInterpretable Machine Learning Approach for Predicting 30-Day Mortality of Critical Ill Patients with Pulmonary Embolism and Heart Failure: A Retrospective Study
Background: Pulmonary embolism (PE) patients combined with heart failure (HF) have been reported to have a high short-term mortality. However, few studies have developed predictive tools of 30-day mortality for these patients in intensive care unit (ICU). This study aimed to construct and validate a machine learning (ML) model to predict 30-day mortality for PE patients combined with HF in ICU. Methods: We enrolled patients with PE combined with HF in the Medical Information Mart for Intensive Care Database (MIMIC) and developed six ML models after feature selection. Further, eICU Collaborative Research Database (eICU-CRD) was utilized for external vali- dation. The area under curves (AUC), calibration curves, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were performed to evaluate the prediction performance. Shapley additive explanation (SHAP) was performed to enhance the interpretability of our models. Results: A total of 472 PE patients combined with HF were included. We developed six ML models by the 13 selected features. After internal validation, the Support Vector Ma- chine (SVM) model performed best with an AUC of 0.835, a superior calibration degree, and a wider risk threshold (from 0% to 90%) for obtaining clinical benefit, which also outperformed traditional mortality risk evaluation systems,as evaluated by NRI and IDI. The SVM model was still reliable after external validation. SHAP was performed to explain the model. Moreover, an online application was developed for further clinical use. Conclusion: This study developed a potential tool for identify short-term mortality risk to guide clinical decision making for PE patients combined with HF in the ICU. The SHAP method also helped clinicians to better understand the model.