Materiały źródłowe

• #1 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Predicting mortality after acute myocardial infarction (AMI) is crucial for timely prescription and treatment of AMI patients, but there are no appropriate AI systems for clinicians. Our primary goal is to develop a reliable and interpretable AI system and provide some valuable insights regarding short, and long-term mortality. […] RIAS reveals the significance of statin-based medications, beta-blockers, and age on mortality regardless of time period. […] RIAS addresses the black-box issue in AI by providing both global and local explanations based on SHAP values and reliable predictions, interpretable as actual likelihoods. […] The proposed framework provides reliable and interpretable predictions along with counterfactual examples. […] To deliver timely intervention and personalized therapy for AMI patients, it is crucial to assess prognosis and predict short-term mortality.

• #2

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #3

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #4

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #5 Machine learning-based long-term outcome prediction in patients undergoing percutaneous coronary intervention – Liu – Cardiovascular Diagnosis and Therapy

  https://cdt.amegroups.org/article/view/69917/html

  During the 5-year follow-up, 467 (4.82%) patients died. […] ML models improved the prediction of long-term all-cause mortality in patients with coronary artery disease before PCI. […] The performance of the RF model was better than that of the other models, providing a meaningful stratification. […] The RF-PCI score can be used to predict the long-term prognosis of PCI patients. […] Diastolic blood pressure, age, and brain natriuretic peptide are the top three risk factors for 5-year mortality in patients undergoing PCI. […] This study provides useful information for a predictive model for PCI recipients.

• #6 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Machine learning may be useful to characterize cardiovascular risk, predict outcomes and identify biomarkers in population studies. […] To test the ability of random survival forests (RF), a machine learning technique, to predict six cardiovascular outcomes in comparison to standard cardiovascular risk scores. […] The RF technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 1025%). […] Machine learning in conjunction with deep phenotyping improve prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. […] Increasing age, perhaps reflecting duration of risk exposure, was the most important predictor of all-cause death. […] Inflammation and immune response measured by increased interleukin-6, fibrinogen, homocysteine, TNF- SR and IL2 SR levels; and abnormal hemostasis measured by increased D-Dimer, plasmin-antiplasmin complex and factor VIII levels were among the top 20 markers of all-cause death underlining the role of inflammation and thrombosis as common pathways for chronic diseases leading to death.

• #7 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Machine learning may be useful to characterize cardiovascular risk, predict outcomes and identify biomarkers in population studies. […] To test the ability of random survival forests (RF), a machine learning technique, to predict six cardiovascular outcomes in comparison to standard cardiovascular risk scores. […] The RF technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 1025%). […] Machine learning in conjunction with deep phenotyping improve prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. […] Increasing age, perhaps reflecting duration of risk exposure, was the most important predictor of all-cause death. […] Inflammation and immune response measured by increased interleukin-6, fibrinogen, homocysteine, TNF- SR and IL2 SR levels; and abnormal hemostasis measured by increased D-Dimer, plasmin-antiplasmin complex and factor VIII levels were among the top 20 markers of all-cause death underlining the role of inflammation and thrombosis as common pathways for chronic diseases leading to death.

• #8 Machine learning-based long-term outcome prediction in patients undergoing percutaneous coronary intervention – Liu – Cardiovascular Diagnosis and Therapy

  https://cdt.amegroups.org/article/view/69917/html

  During the 5-year follow-up, 467 (4.82%) patients died. […] ML models improved the prediction of long-term all-cause mortality in patients with coronary artery disease before PCI. […] The performance of the RF model was better than that of the other models, providing a meaningful stratification. […] The RF-PCI score can be used to predict the long-term prognosis of PCI patients. […] Diastolic blood pressure, age, and brain natriuretic peptide are the top three risk factors for 5-year mortality in patients undergoing PCI. […] This study provides useful information for a predictive model for PCI recipients.

• #9 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Machine learning may be useful to characterize cardiovascular risk, predict outcomes and identify biomarkers in population studies. […] To test the ability of random survival forests (RF), a machine learning technique, to predict six cardiovascular outcomes in comparison to standard cardiovascular risk scores. […] The RF technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 1025%). […] Machine learning in conjunction with deep phenotyping improve prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. […] Increasing age, perhaps reflecting duration of risk exposure, was the most important predictor of all-cause death. […] Inflammation and immune response measured by increased interleukin-6, fibrinogen, homocysteine, TNF- SR and IL2 SR levels; and abnormal hemostasis measured by increased D-Dimer, plasmin-antiplasmin complex and factor VIII levels were among the top 20 markers of all-cause death underlining the role of inflammation and thrombosis as common pathways for chronic diseases leading to death.

• #10 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Machine learning may be useful to characterize cardiovascular risk, predict outcomes and identify biomarkers in population studies. […] To test the ability of random survival forests (RF), a machine learning technique, to predict six cardiovascular outcomes in comparison to standard cardiovascular risk scores. […] The RF technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 1025%). […] Machine learning in conjunction with deep phenotyping improve prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. […] Increasing age, perhaps reflecting duration of risk exposure, was the most important predictor of all-cause death. […] Inflammation and immune response measured by increased interleukin-6, fibrinogen, homocysteine, TNF- SR and IL2 SR levels; and abnormal hemostasis measured by increased D-Dimer, plasmin-antiplasmin complex and factor VIII levels were among the top 20 markers of all-cause death underlining the role of inflammation and thrombosis as common pathways for chronic diseases leading to death.

• #11 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Similarly, cardiac stress measured by increased NT pro-BNP levels, and myocyte damage by increased cardiac troponin T levels were among the top predictive markers of death reflecting the role of cardiac failure on mortality. […] Expectedly, a composite of atherosclerosis measures (low and high ABI, increased carotid IMT, decreased aortic distensibility) were among the most important predictors of CHD which represents a subset of CVD events, with CAC being by far the most important, reflecting the specific influence of coronary atherosclerosis. […] For incident HF as the endpoint, cardiac chamber stress (increased LV volume, and increased NT-proBNP levels), and decreased LV function from MRI were the most important markers. […] For incident AF as the endpoint, inflammation, higher levels of creatinine, atherosclerosis (CAC and ABI), and repolarization abnormalities were the most important markers.

• #12 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Similarly, cardiac stress measured by increased NT pro-BNP levels, and myocyte damage by increased cardiac troponin T levels were among the top predictive markers of death reflecting the role of cardiac failure on mortality. […] Expectedly, a composite of atherosclerosis measures (low and high ABI, increased carotid IMT, decreased aortic distensibility) were among the most important predictors of CHD which represents a subset of CVD events, with CAC being by far the most important, reflecting the specific influence of coronary atherosclerosis. […] For incident HF as the endpoint, cardiac chamber stress (increased LV volume, and increased NT-proBNP levels), and decreased LV function from MRI were the most important markers. […] For incident AF as the endpoint, inflammation, higher levels of creatinine, atherosclerosis (CAC and ABI), and repolarization abnormalities were the most important markers.

• #13 Machine learning-based long-term outcome prediction in patients undergoing percutaneous coronary intervention – Liu – Cardiovascular Diagnosis and Therapy

  https://cdt.amegroups.org/article/view/69917/html

  During the 5-year follow-up, 467 (4.82%) patients died. […] ML models improved the prediction of long-term all-cause mortality in patients with coronary artery disease before PCI. […] The performance of the RF model was better than that of the other models, providing a meaningful stratification. […] The RF-PCI score can be used to predict the long-term prognosis of PCI patients. […] Diastolic blood pressure, age, and brain natriuretic peptide are the top three risk factors for 5-year mortality in patients undergoing PCI. […] This study provides useful information for a predictive model for PCI recipients.

• #14 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Similarly, cardiac stress measured by increased NT pro-BNP levels, and myocyte damage by increased cardiac troponin T levels were among the top predictive markers of death reflecting the role of cardiac failure on mortality. […] Expectedly, a composite of atherosclerosis measures (low and high ABI, increased carotid IMT, decreased aortic distensibility) were among the most important predictors of CHD which represents a subset of CVD events, with CAC being by far the most important, reflecting the specific influence of coronary atherosclerosis. […] For incident HF as the endpoint, cardiac chamber stress (increased LV volume, and increased NT-proBNP levels), and decreased LV function from MRI were the most important markers. […] For incident AF as the endpoint, inflammation, higher levels of creatinine, atherosclerosis (CAC and ABI), and repolarization abnormalities were the most important markers.

• #15 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Similarly, cardiac stress measured by increased NT pro-BNP levels, and myocyte damage by increased cardiac troponin T levels were among the top predictive markers of death reflecting the role of cardiac failure on mortality. […] Expectedly, a composite of atherosclerosis measures (low and high ABI, increased carotid IMT, decreased aortic distensibility) were among the most important predictors of CHD which represents a subset of CVD events, with CAC being by far the most important, reflecting the specific influence of coronary atherosclerosis. […] For incident HF as the endpoint, cardiac chamber stress (increased LV volume, and increased NT-proBNP levels), and decreased LV function from MRI were the most important markers. […] For incident AF as the endpoint, inflammation, higher levels of creatinine, atherosclerosis (CAC and ABI), and repolarization abnormalities were the most important markers.

• #16 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Predicting mortality after acute myocardial infarction (AMI) is crucial for timely prescription and treatment of AMI patients, but there are no appropriate AI systems for clinicians. Our primary goal is to develop a reliable and interpretable AI system and provide some valuable insights regarding short, and long-term mortality. […] RIAS reveals the significance of statin-based medications, beta-blockers, and age on mortality regardless of time period. […] RIAS addresses the black-box issue in AI by providing both global and local explanations based on SHAP values and reliable predictions, interpretable as actual likelihoods. […] The proposed framework provides reliable and interpretable predictions along with counterfactual examples. […] To deliver timely intervention and personalized therapy for AMI patients, it is crucial to assess prognosis and predict short-term mortality.

• #17 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Recent studies have proposed advanced machine learning (ML) techniques using electronic health record (EHR) data to assess risk and predict mortality in AMI cases. […] RIAS generates an optimized model for a given dataset and ensures both reliability and interpretability in outcomes. […] The mortality rates for each clinical event are 3%, 5%, and 6% for in-hospital, 6-month, and 12-month mortality, respectively. […] Our study introduces the RIASan end-to-end framework that is designed with the above principles at its core. […] The system gained the most benefit in predicting in-hospital mortality with an AUROC of 0.990 and an F1 score of 0.833. […] RIAS has resolved these issues by providing global (overall) and local (individual patient) explanations based on SHAP values and reliable predictions which can be interpreted as the actual likelihood through confidence calibration. […] Predicting mortality after AMI is critical for timely and personalized interventions to reduce AMI mortality and hinder heart failure progression.

• #18 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Predicting mortality after acute myocardial infarction (AMI) is crucial for timely prescription and treatment of AMI patients, but there are no appropriate AI systems for clinicians. Our primary goal is to develop a reliable and interpretable AI system and provide some valuable insights regarding short, and long-term mortality. […] RIAS reveals the significance of statin-based medications, beta-blockers, and age on mortality regardless of time period. […] RIAS addresses the black-box issue in AI by providing both global and local explanations based on SHAP values and reliable predictions, interpretable as actual likelihoods. […] The proposed framework provides reliable and interpretable predictions along with counterfactual examples. […] To deliver timely intervention and personalized therapy for AMI patients, it is crucial to assess prognosis and predict short-term mortality.

• #19 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Recent studies have proposed advanced machine learning (ML) techniques using electronic health record (EHR) data to assess risk and predict mortality in AMI cases. […] RIAS generates an optimized model for a given dataset and ensures both reliability and interpretability in outcomes. […] The mortality rates for each clinical event are 3%, 5%, and 6% for in-hospital, 6-month, and 12-month mortality, respectively. […] Our study introduces the RIASan end-to-end framework that is designed with the above principles at its core. […] The system gained the most benefit in predicting in-hospital mortality with an AUROC of 0.990 and an F1 score of 0.833. […] RIAS has resolved these issues by providing global (overall) and local (individual patient) explanations based on SHAP values and reliable predictions which can be interpreted as the actual likelihood through confidence calibration. […] Predicting mortality after AMI is critical for timely and personalized interventions to reduce AMI mortality and hinder heart failure progression.

• #20 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Predicting mortality after acute myocardial infarction (AMI) is crucial for timely prescription and treatment of AMI patients, but there are no appropriate AI systems for clinicians. Our primary goal is to develop a reliable and interpretable AI system and provide some valuable insights regarding short, and long-term mortality. […] RIAS reveals the significance of statin-based medications, beta-blockers, and age on mortality regardless of time period. […] RIAS addresses the black-box issue in AI by providing both global and local explanations based on SHAP values and reliable predictions, interpretable as actual likelihoods. […] The proposed framework provides reliable and interpretable predictions along with counterfactual examples. […] To deliver timely intervention and personalized therapy for AMI patients, it is crucial to assess prognosis and predict short-term mortality.

• #21 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Machine learning may be useful to characterize cardiovascular risk, predict outcomes and identify biomarkers in population studies. […] To test the ability of random survival forests (RF), a machine learning technique, to predict six cardiovascular outcomes in comparison to standard cardiovascular risk scores. […] The RF technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 1025%). […] Machine learning in conjunction with deep phenotyping improve prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. […] Increasing age, perhaps reflecting duration of risk exposure, was the most important predictor of all-cause death. […] Inflammation and immune response measured by increased interleukin-6, fibrinogen, homocysteine, TNF- SR and IL2 SR levels; and abnormal hemostasis measured by increased D-Dimer, plasmin-antiplasmin complex and factor VIII levels were among the top 20 markers of all-cause death underlining the role of inflammation and thrombosis as common pathways for chronic diseases leading to death.

• #22 Machine learning-based long-term outcome prediction in patients undergoing percutaneous coronary intervention – Liu – Cardiovascular Diagnosis and Therapy

  https://cdt.amegroups.org/article/view/69917/html

  During the 5-year follow-up, 467 (4.82%) patients died. […] ML models improved the prediction of long-term all-cause mortality in patients with coronary artery disease before PCI. […] The performance of the RF model was better than that of the other models, providing a meaningful stratification. […] The RF-PCI score can be used to predict the long-term prognosis of PCI patients. […] Diastolic blood pressure, age, and brain natriuretic peptide are the top three risk factors for 5-year mortality in patients undergoing PCI. […] This study provides useful information for a predictive model for PCI recipients.

• #23 A deep learning-based electrocardiogram risk score for long term cardiovascular death and disease | npj Digital Medicine

  https://www.nature.com/articles/s41746-023-00916-6

  The electrocardiogram (ECG) is the most frequently performed cardiovascular diagnostic test, but it is unclear how much information resting ECGs contain about long term cardiovascular risk. Here we report that a deep convolutional neural network can accurately predict the long-term risk of cardiovascular mortality and disease based on a resting ECG alone. SEER predicts 5-year cardiovascular mortality with an area under the receiver operator characteristic curve (AUC) of 0.83 in a held-out test set at Stanford, and with AUCs of 0.78 and 0.83 respectively when independently evaluated at Cedars-Sinai Medical Center and Columbia University Irving Medical Center. SEER can also predict several other cardiovascular conditions such as heart failure and atrial fibrillation. SEER stratifies patients in five different evaluation cohorts from three different medical centers and two different ECG vendors, and out-performs predictions based on traditional risk factors. SEER uncovers 16% of patients misclassified as low-risk by the PCE, highlighting a new cohort of patients who may benefit from statins and would be missed following current practices. SEER also stratifies patients with intermediate risk according to the PCE score, showing potential to perform a similar role to what the CAC score currently plays. The relative low-cost and ubiquity of the ECG makes common application possible.

• #24 Machine learning for diagnosis of myocardial infarction using cardiac troponin concentrations | Nature Medicine

  https://www.nature.com/articles/s41591-023-02325-4

  CoDE-ACS used as a clinical decision support system has the potential to reduce hospital admissions and have major benefits for patients and health care providers. […] In patients presenting with possible acute myocardial infarction, we developed and validated the CoDE-ACS clinical decision support system using machine learning with single or serial high-sensitivity cardiac troponin measurements to inform the probability of acute myocardial infarction. […] Compared with guideline-recommended pathways using cardiac troponin thresholds and risk scores, CoDE-ACS identified twice as many patients as low probability of myocardial infarction at presentation with a similar negative predictive value and fewer patients as high probability with an improved positive predictive value. […] While our models were trained to estimate the probability of myocardial infarction during the index hospital admission, patients who were identified as low probability of myocardial infarction were also at low risk of death following discharge, with fewer than 1 in 300 having a cardiac death at 1year.

• #25 Artificial neural networks improve early outcome prediction and risk classification in out-of-hospital cardiac arrest patients admitted to intensive care | springermedizin.de

  https://www.springermedizin.de/artificial-neural-networks-improve-early-outcome-prediction-and-/18233080

  Pre-hospital circumstances, cardiac arrest characteristics, comorbidities and clinical status on admission are strongly associated with outcome after out-of-hospital cardiac arrest (OHCA). […] Early prediction of outcome may inform prognosis, tailor therapy and help in interpreting the intervention effect in heterogenous clinical trials. […] The outcome was predicted with an area under the receiver operating characteristic curve (AUC) of 0.891 using 54 clinical variables available on admission to hospital, categorised as background, pre-hospital and admission data. […] A supervised machine learning model using ANN predicted neurological recovery, including survival excellently, and outperformed a conventional model based on logistic regression. […] Among the data available at the time of hospitalisation, factors related to the pre-hospital setting carried most information.

• #26 Artificial neural networks improve early outcome prediction and risk classification in out-of-hospital cardiac arrest patients admitted to intensive care | springermedizin.de

  https://www.springermedizin.de/artificial-neural-networks-improve-early-outcome-prediction-and-/18233080

  The overall ANN model, based on the 54 variables available on admission, showed an excellent capability of outcome prediction during the internal validation training and performed even better on the test set with an AUC of 0.891. […] Using only the three most important independent factors (age, time to ROSC and first monitored rhythm, which are variables readily known on arrival in the emergency room) in an ANN led to a model with an excellent predictive ability on the test set with an AUC of 0.852 which is better compared to most proposed models in the field. […] Our supervised machine learning model of ANN predicted neurological recovery, including survival excellently and outperformed a conventional model based on logistic regression. […] ANN may stratify a heterogenous trial population in risk classes and help determine intervention effect across subgroups.

• #27 Early prediction of in-hospital mortality in patients with congestive heart failure in intensive care unit: a retrospective observational cohort study | BMJ Open

  https://bmjopen.bmj.com/content/12/7/e059761

  Objective Congestive heart failure (CHF) is a clinical syndrome in which the heart disease progresses to a severe stage. Early diagnosis and risk assessment of death of patients with CHF are critical to prognosis and treatment. The purpose of this study was to establish a nomogram that predicts the in-hospital death of patients with CHF in the intensive care unit (ICU). […] The 20 independent risk factors for in-hospital mortality of patients with CHF were age, race, norepinephrine, dopamine, phenylephrine, vasopressin, mechanical ventilation, intubation, HepF, heart rate, respiratory rate, temperature, SBP, AG, BUN, creatinine, chloride, MCV, RDW and WCC. The nomogram, which included these factors, accurately predicted the in-hospital mortality of patients with CHF. The novel nomogram has the potential for use in clinical practice as a tool to predict and assess mortality of patients with CHF in the ICU.

• #28 Development of a multidimensional prediction model for long-term prognostic risk in patients with acute coronary syndromes after percutaneous coronary intervention: A retrospective observational cohort study | PLOS One

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0318445

  The aim of this study is to examine the critical variables that impact the long-term prognosis of patients with acute coronary syndrome (ACS) after percutaneous coronary intervention (PCI) and to create a multidimensional predictive risk assessment model that can serve as a theoretical basis for accurate cardiac rehabilitation. […] We found white blood cell count (WBC) (OR: 4.110) and the effective average number of daily steps (ANS) (OR: 2.689) as independent prognostic risk factors for acute myocardial infarction (AMI). The independent risk factors for unstable angina prognosis were white blood cell count (OR: 6.257), VO2 at anaerobic threshold (OR: 4.294), and effective autonomic nervous system function (OR: 4.097). […] This study developed a multimodal predictive model that integrates the inflammatory response after onset, physical performance and exercise tolerance before discharge, and daily activity after discharge to predict the long-term prognosis of patients with ACS. The multidimensional model is more effective than the single-factor model for assessing risk in ACS patients.

• #29 Early prediction of in-hospital mortality in patients with congestive heart failure in intensive care unit: a retrospective observational cohort study | BMJ Open

  https://bmjopen.bmj.com/content/12/7/e059761

  Objective Congestive heart failure (CHF) is a clinical syndrome in which the heart disease progresses to a severe stage. Early diagnosis and risk assessment of death of patients with CHF are critical to prognosis and treatment. The purpose of this study was to establish a nomogram that predicts the in-hospital death of patients with CHF in the intensive care unit (ICU). […] The 20 independent risk factors for in-hospital mortality of patients with CHF were age, race, norepinephrine, dopamine, phenylephrine, vasopressin, mechanical ventilation, intubation, HepF, heart rate, respiratory rate, temperature, SBP, AG, BUN, creatinine, chloride, MCV, RDW and WCC. The nomogram, which included these factors, accurately predicted the in-hospital mortality of patients with CHF. The novel nomogram has the potential for use in clinical practice as a tool to predict and assess mortality of patients with CHF in the ICU.

• #30 Early prediction of in-hospital mortality in patients with congestive heart failure in intensive care unit: a retrospective observational cohort study | BMJ Open

  https://bmjopen.bmj.com/content/12/7/e059761

  Prognostic factors significantly related to in-hospital mortality of patients with CHF were age, race, norepinephrine, dopamine, epinephrine, phenylephrine, vasopressin, mechanical ventilation, intubation, AMI, HepF, heart rate, respiratory rate, temperature, SBP, AG, BUN, creatinine, calcium, chloride, haemoglobin, potassium, MCHC, MCV, RDW, RBC and WCC. Multivariate logistic regression analysis was then performed on these variables to control for confounding effects as much as possible. Finally, age, race, norepinephrine, dopamine, phenylephrine, vasopressin, mechanical ventilation, intubation, HepF, heart rate, respiratory rate, temperature, SBP, AG, BUN, creatinine, chloride, MCV, RDW and WCC were determined as the prognostic factors for patients with CHF. […] The C-index of the nomogram was 0.839 (95% CI 0.829 to 0.848), indicating good discrimination ability. The ROC curves of the SOFA, APSIII and GWTGHF scores and the novel nomogram used to predict in-hospital mortality are presented in figure 3. The Area Under Curve (AUC) values indicated that our novel nomogram model is better than the SOFA, APSIII and GWTGHF score models.

• #31 Development of a multidimensional prediction model for long-term prognostic risk in patients with acute coronary syndromes after percutaneous coronary intervention: A retrospective observational cohort study | PLOS One

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0318445

  The aim of this study is to examine the critical variables that impact the long-term prognosis of patients with acute coronary syndrome (ACS) after percutaneous coronary intervention (PCI) and to create a multidimensional predictive risk assessment model that can serve as a theoretical basis for accurate cardiac rehabilitation. […] We found white blood cell count (WBC) (OR: 4.110) and the effective average number of daily steps (ANS) (OR: 2.689) as independent prognostic risk factors for acute myocardial infarction (AMI). The independent risk factors for unstable angina prognosis were white blood cell count (OR: 6.257), VO2 at anaerobic threshold (OR: 4.294), and effective autonomic nervous system function (OR: 4.097). […] This study developed a multimodal predictive model that integrates the inflammatory response after onset, physical performance and exercise tolerance before discharge, and daily activity after discharge to predict the long-term prognosis of patients with ACS. The multidimensional model is more effective than the single-factor model for assessing risk in ACS patients.

• #32 Development of a multidimensional prediction model for long-term prognostic risk in patients with acute coronary syndromes after percutaneous coronary intervention: A retrospective observational cohort study | PLOS One

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0318445

  Overall, our comprehensive prognostic risk assessment for AMI assigns a total score of 5 points, categorizing 0 points as low risk, 2 and 3 points as intermediate risk, and 5 points as high risk; and for UA assigns a total score of 7 points, categorizing 03 points as low risk, 45 points as intermediate risk, and 7 points as high risk. […] This study found that the inflammatory response after onset, physical performance and exercise tolerance before discharge, and daily activity after discharge were the independent risk factors for predicting the long-term prognosis of patients with ACS. The multidimensional prognostic model to risk-stratify for the patients with ACS, was better than the single factor model. This study also provides a theoretical basis that the prognosis of potentially high-risk patients can be improved by precise and rational exercise prescription.

• #33 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Recent studies have proposed advanced machine learning (ML) techniques using electronic health record (EHR) data to assess risk and predict mortality in AMI cases. […] RIAS generates an optimized model for a given dataset and ensures both reliability and interpretability in outcomes. […] The mortality rates for each clinical event are 3%, 5%, and 6% for in-hospital, 6-month, and 12-month mortality, respectively. […] Our study introduces the RIASan end-to-end framework that is designed with the above principles at its core. […] The system gained the most benefit in predicting in-hospital mortality with an AUROC of 0.990 and an F1 score of 0.833. […] RIAS has resolved these issues by providing global (overall) and local (individual patient) explanations based on SHAP values and reliable predictions which can be interpreted as the actual likelihood through confidence calibration. […] Predicting mortality after AMI is critical for timely and personalized interventions to reduce AMI mortality and hinder heart failure progression.

• #34 Machine learning-based long-term outcome prediction in patients undergoing percutaneous coronary intervention – Liu – Cardiovascular Diagnosis and Therapy

  https://cdt.amegroups.org/article/view/69917/html

  During the 5-year follow-up, 467 (4.82%) patients died. […] ML models improved the prediction of long-term all-cause mortality in patients with coronary artery disease before PCI. […] The performance of the RF model was better than that of the other models, providing a meaningful stratification. […] The RF-PCI score can be used to predict the long-term prognosis of PCI patients. […] Diastolic blood pressure, age, and brain natriuretic peptide are the top three risk factors for 5-year mortality in patients undergoing PCI. […] This study provides useful information for a predictive model for PCI recipients.

• #35 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Recent studies have proposed advanced machine learning (ML) techniques using electronic health record (EHR) data to assess risk and predict mortality in AMI cases. […] RIAS generates an optimized model for a given dataset and ensures both reliability and interpretability in outcomes. […] The mortality rates for each clinical event are 3%, 5%, and 6% for in-hospital, 6-month, and 12-month mortality, respectively. […] Our study introduces the RIASan end-to-end framework that is designed with the above principles at its core. […] The system gained the most benefit in predicting in-hospital mortality with an AUROC of 0.990 and an F1 score of 0.833. […] RIAS has resolved these issues by providing global (overall) and local (individual patient) explanations based on SHAP values and reliable predictions which can be interpreted as the actual likelihood through confidence calibration. […] Predicting mortality after AMI is critical for timely and personalized interventions to reduce AMI mortality and hinder heart failure progression.

• #36 Machine learning-based long-term outcome prediction in patients undergoing percutaneous coronary intervention – Liu – Cardiovascular Diagnosis and Therapy

  https://cdt.amegroups.org/article/view/69917/html

  During the 5-year follow-up, 467 (4.82%) patients died. […] ML models improved the prediction of long-term all-cause mortality in patients with coronary artery disease before PCI. […] The performance of the RF model was better than that of the other models, providing a meaningful stratification. […] The RF-PCI score can be used to predict the long-term prognosis of PCI patients. […] Diastolic blood pressure, age, and brain natriuretic peptide are the top three risk factors for 5-year mortality in patients undergoing PCI. […] This study provides useful information for a predictive model for PCI recipients.

• #37 Machine learning for diagnosis of myocardial infarction using cardiac troponin concentrations | Nature Medicine

  https://www.nature.com/articles/s41591-023-02325-4

  Although guidelines recommend fixed cardiac troponin thresholds for the diagnosis of myocardial infarction, troponin concentrations are influenced by age, sex, comorbidities and time from symptom onset. […] CoDE-ACS had excellent discrimination for myocardial infarction (area under curve, 0.953; 95% confidence interval, 0.9470.958), performed well across subgroups and identified more patients at presentation as low probability of having myocardial infarction than fixed cardiac troponin thresholds (61 versus 27%) with a similar negative predictive value and fewer as high probability of having myocardial infarction (10 versus 16%) with a greater positive predictive value. […] Patients identified as having a low probability of myocardial infarction had a lower rate of cardiac death than those with intermediate or high probability 30days (0.1 versus 0.5 and 1.8%) and 1year (0.3 versus 2.8 and 4.2%; P0.001 for both) from patient presentation.

• #38 Machine learning for diagnosis of myocardial infarction using cardiac troponin concentrations | Nature Medicine

  https://www.nature.com/articles/s41591-023-02325-4

  CoDE-ACS used as a clinical decision support system has the potential to reduce hospital admissions and have major benefits for patients and health care providers. […] In patients presenting with possible acute myocardial infarction, we developed and validated the CoDE-ACS clinical decision support system using machine learning with single or serial high-sensitivity cardiac troponin measurements to inform the probability of acute myocardial infarction. […] Compared with guideline-recommended pathways using cardiac troponin thresholds and risk scores, CoDE-ACS identified twice as many patients as low probability of myocardial infarction at presentation with a similar negative predictive value and fewer patients as high probability with an improved positive predictive value. […] While our models were trained to estimate the probability of myocardial infarction during the index hospital admission, patients who were identified as low probability of myocardial infarction were also at low risk of death following discharge, with fewer than 1 in 300 having a cardiac death at 1year.

• #39 Interest of Lung Ultrasound in the Management of Acute Heart Failure in Post-Emergency Service

  https://www.mdpi.com/2075-1729/15/5/752

  Lung ultrasound (LUS) has emerged as a simple, rapid, and non-invasive method for the dynamic assessment of pulmonary congestion, a major prognostic factor and a therapeutic target in acute heart failure (AHF). […] Pulmonary congestion appears as a major prognostic factor in AHF, and therefore represents an important therapeutic target. […] The goal of this prospective observational study is to evaluate whether changes in B-lines observed on LUSs can reliably reflect pulmonary congestion status and correlate with diuretic dosing, as well as clinical and biological signs, in AHF patients hospitalized in a polyvalent medicine unit. […] The presence of B-lines at discharge is significantly associated with a higher risk of mortality at 30-days. […] The relationship between the number of B-lines and the vital status at 30-days discharge is statistically significant with a p-value < 0.001.

• #40 Machine learning for diagnosis of myocardial infarction using cardiac troponin concentrations | Nature Medicine

  https://www.nature.com/articles/s41591-023-02325-4

  CoDE-ACS used as a clinical decision support system has the potential to reduce hospital admissions and have major benefits for patients and health care providers. […] In patients presenting with possible acute myocardial infarction, we developed and validated the CoDE-ACS clinical decision support system using machine learning with single or serial high-sensitivity cardiac troponin measurements to inform the probability of acute myocardial infarction. […] Compared with guideline-recommended pathways using cardiac troponin thresholds and risk scores, CoDE-ACS identified twice as many patients as low probability of myocardial infarction at presentation with a similar negative predictive value and fewer patients as high probability with an improved positive predictive value. […] While our models were trained to estimate the probability of myocardial infarction during the index hospital admission, patients who were identified as low probability of myocardial infarction were also at low risk of death following discharge, with fewer than 1 in 300 having a cardiac death at 1year.

• #41 Interest of Lung Ultrasound in the Management of Acute Heart Failure in Post-Emergency Service

  https://www.mdpi.com/2075-1729/15/5/752

  Lung ultrasound (LUS) has emerged as a simple, rapid, and non-invasive method for the dynamic assessment of pulmonary congestion, a major prognostic factor and a therapeutic target in acute heart failure (AHF). […] Pulmonary congestion appears as a major prognostic factor in AHF, and therefore represents an important therapeutic target. […] The goal of this prospective observational study is to evaluate whether changes in B-lines observed on LUSs can reliably reflect pulmonary congestion status and correlate with diuretic dosing, as well as clinical and biological signs, in AHF patients hospitalized in a polyvalent medicine unit. […] The presence of B-lines at discharge is significantly associated with a higher risk of mortality at 30-days. […] The relationship between the number of B-lines and the vital status at 30-days discharge is statistically significant with a p-value < 0.001.

• #42 Interest of Lung Ultrasound in the Management of Acute Heart Failure in Post-Emergency Service

  https://www.mdpi.com/2075-1729/15/5/752

  The lack of clearance of B-lines at discharge seems to be associated with a high risk of adverse events, such as ED consultation, readmission for AHF, and all-cause mortality. […] We found a significant relationship between the number of B-lines at discharge and risk of death at 30 days. […] In the absence of a gold standard for the quantitative assessment of cardiac overload, LUS is a valuable tool that could help optimize the management of congestion, particularly subclinical congestion, thereby improving the 30-day prognoses.

• #43 A deep learning-based electrocardiogram risk score for long term cardiovascular death and disease | npj Digital Medicine

  https://www.nature.com/articles/s41746-023-00916-6

  The electrocardiogram (ECG) is the most frequently performed cardiovascular diagnostic test, but it is unclear how much information resting ECGs contain about long term cardiovascular risk. Here we report that a deep convolutional neural network can accurately predict the long-term risk of cardiovascular mortality and disease based on a resting ECG alone. SEER predicts 5-year cardiovascular mortality with an area under the receiver operator characteristic curve (AUC) of 0.83 in a held-out test set at Stanford, and with AUCs of 0.78 and 0.83 respectively when independently evaluated at Cedars-Sinai Medical Center and Columbia University Irving Medical Center. SEER can also predict several other cardiovascular conditions such as heart failure and atrial fibrillation. SEER stratifies patients in five different evaluation cohorts from three different medical centers and two different ECG vendors, and out-performs predictions based on traditional risk factors. SEER uncovers 16% of patients misclassified as low-risk by the PCE, highlighting a new cohort of patients who may benefit from statins and would be missed following current practices. SEER also stratifies patients with intermediate risk according to the PCE score, showing potential to perform a similar role to what the CAC score currently plays. The relative low-cost and ubiquity of the ECG makes common application possible.

• #44 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Depending on the patient’s cardiac forecast, their physician may begin appropriate interventions with medications to manage cholesterol, diabetes and high blood pressure as well as recommendations and plans to change damaging habits. […] In trials, the system predicted 7.6 percent more events than traditional methods with 1.6 percent fewer false alarms. […] The European Society of Cardiology found that measuring levels of a molecule called trimethylamine N-oxide (TMAO), which is produced by gut bacteria, could help them reliably assess a patient’s risk of heart problems. […] Quyyumi is interested in looking at plaques in the bloodstream as predictors. […] “Can we try to detect activation of plaque by looking for blood markers?” […] Today, though, he’s more optimistic. “I think it is possible. We can diagnose the condition much better. It’s possible this will become less of a problem. We’re moving in the right direction.”

• #45 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Depending on the patient’s cardiac forecast, their physician may begin appropriate interventions with medications to manage cholesterol, diabetes and high blood pressure as well as recommendations and plans to change damaging habits. […] In trials, the system predicted 7.6 percent more events than traditional methods with 1.6 percent fewer false alarms. […] The European Society of Cardiology found that measuring levels of a molecule called trimethylamine N-oxide (TMAO), which is produced by gut bacteria, could help them reliably assess a patient’s risk of heart problems. […] Quyyumi is interested in looking at plaques in the bloodstream as predictors. […] “Can we try to detect activation of plaque by looking for blood markers?” […] Today, though, he’s more optimistic. “I think it is possible. We can diagnose the condition much better. It’s possible this will become less of a problem. We’re moving in the right direction.”

• #46 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Depending on the patient’s cardiac forecast, their physician may begin appropriate interventions with medications to manage cholesterol, diabetes and high blood pressure as well as recommendations and plans to change damaging habits. […] In trials, the system predicted 7.6 percent more events than traditional methods with 1.6 percent fewer false alarms. […] The European Society of Cardiology found that measuring levels of a molecule called trimethylamine N-oxide (TMAO), which is produced by gut bacteria, could help them reliably assess a patient’s risk of heart problems. […] Quyyumi is interested in looking at plaques in the bloodstream as predictors. […] “Can we try to detect activation of plaque by looking for blood markers?” […] Today, though, he’s more optimistic. “I think it is possible. We can diagnose the condition much better. It’s possible this will become less of a problem. We’re moving in the right direction.”

• #47 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  The case fatality rate following acute heart failure declined modestly over the 25-year study period but remained higher in socially deprived groups and there was a slight rise in the incidence of rehospitalisation with heart failure. […] The reduction in mortality rate over time in older patients was similar to that observed in younger patients. […] However, the overall negative predictive value (NPV) thresholds masks major heterogeneity across important subgroups, such as older age, history of ischaemic heart disease, atrial fibrillation, renal impairment and, most of all, prior heart failure. […] Most notably, patients without prior heart failure had a positive predictive value (PPV) of only around 50% with a single guideline-recommended threshold (300pg/mL) approach with NT-proBNP.

• #48 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  Therefore, deriving distinct optimal thresholds for patients with and without a prior diagnosis of heart failure – and potentially other key subgroups – has the potential to improve the practical clinical utility of NT-proBNP testing. […] Overall, my thesis findings suggest that improvements in the prevention and treatment of heart failure have had progressive and sustained effects on the incident AHF hospitalisation and outcomes at the population and subgroup level although the overall prognosis remains poor. […] Our analysis of the sub-population of patients with and without heart failure history with NT-proBNP suggests that there is scope to optimise the rule-out and rule-in thresholds in patients without a prior heart failure history.

• #49 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  Acute heart failure (AHF) is common and is a major cause of morbidity and mortality in the UK. […] Available data from several countries suggests there may have been a decline in both the incidence of and mortality from AHF in recent years. […] Timely diagnosis of acute heart failure is essential to improving outcomes and natriuretic peptides (NPs) are the recommended biomarkers for diagnosis of AHF. […] The principal aims of my thesis were three-fold. First, to examine the national temporal trends in the incidence and outcomes for AHF hospitalisations in Scotland, including important sub-groups of interest. […] Our analysis of temporal trends (1990-2014) of the Scottish population for incident hospitalised AHF using a national individual-level linkage approach showed that incident acute hospitalised heart failure has reduced over the study period.

• #50 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  The case fatality rate following acute heart failure declined modestly over the 25-year study period but remained higher in socially deprived groups and there was a slight rise in the incidence of rehospitalisation with heart failure. […] The reduction in mortality rate over time in older patients was similar to that observed in younger patients. […] However, the overall negative predictive value (NPV) thresholds masks major heterogeneity across important subgroups, such as older age, history of ischaemic heart disease, atrial fibrillation, renal impairment and, most of all, prior heart failure. […] Most notably, patients without prior heart failure had a positive predictive value (PPV) of only around 50% with a single guideline-recommended threshold (300pg/mL) approach with NT-proBNP.

• #51 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  Acute heart failure (AHF) is common and is a major cause of morbidity and mortality in the UK. […] Available data from several countries suggests there may have been a decline in both the incidence of and mortality from AHF in recent years. […] Timely diagnosis of acute heart failure is essential to improving outcomes and natriuretic peptides (NPs) are the recommended biomarkers for diagnosis of AHF. […] The principal aims of my thesis were three-fold. First, to examine the national temporal trends in the incidence and outcomes for AHF hospitalisations in Scotland, including important sub-groups of interest. […] Our analysis of temporal trends (1990-2014) of the Scottish population for incident hospitalised AHF using a national individual-level linkage approach showed that incident acute hospitalised heart failure has reduced over the study period.

• #52 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  The case fatality rate following acute heart failure declined modestly over the 25-year study period but remained higher in socially deprived groups and there was a slight rise in the incidence of rehospitalisation with heart failure. […] The reduction in mortality rate over time in older patients was similar to that observed in younger patients. […] However, the overall negative predictive value (NPV) thresholds masks major heterogeneity across important subgroups, such as older age, history of ischaemic heart disease, atrial fibrillation, renal impairment and, most of all, prior heart failure. […] Most notably, patients without prior heart failure had a positive predictive value (PPV) of only around 50% with a single guideline-recommended threshold (300pg/mL) approach with NT-proBNP.

• #53 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  The case fatality rate following acute heart failure declined modestly over the 25-year study period but remained higher in socially deprived groups and there was a slight rise in the incidence of rehospitalisation with heart failure. […] The reduction in mortality rate over time in older patients was similar to that observed in younger patients. […] However, the overall negative predictive value (NPV) thresholds masks major heterogeneity across important subgroups, such as older age, history of ischaemic heart disease, atrial fibrillation, renal impairment and, most of all, prior heart failure. […] Most notably, patients without prior heart failure had a positive predictive value (PPV) of only around 50% with a single guideline-recommended threshold (300pg/mL) approach with NT-proBNP.

• #54 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  The case fatality rate following acute heart failure declined modestly over the 25-year study period but remained higher in socially deprived groups and there was a slight rise in the incidence of rehospitalisation with heart failure. […] The reduction in mortality rate over time in older patients was similar to that observed in younger patients. […] However, the overall negative predictive value (NPV) thresholds masks major heterogeneity across important subgroups, such as older age, history of ischaemic heart disease, atrial fibrillation, renal impairment and, most of all, prior heart failure. […] Most notably, patients without prior heart failure had a positive predictive value (PPV) of only around 50% with a single guideline-recommended threshold (300pg/mL) approach with NT-proBNP.

• #55 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Predicting mortality after acute myocardial infarction (AMI) is crucial for timely prescription and treatment of AMI patients, but there are no appropriate AI systems for clinicians. Our primary goal is to develop a reliable and interpretable AI system and provide some valuable insights regarding short, and long-term mortality. […] RIAS reveals the significance of statin-based medications, beta-blockers, and age on mortality regardless of time period. […] RIAS addresses the black-box issue in AI by providing both global and local explanations based on SHAP values and reliable predictions, interpretable as actual likelihoods. […] The proposed framework provides reliable and interpretable predictions along with counterfactual examples. […] To deliver timely intervention and personalized therapy for AMI patients, it is crucial to assess prognosis and predict short-term mortality.

• #56 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Depending on the patient’s cardiac forecast, their physician may begin appropriate interventions with medications to manage cholesterol, diabetes and high blood pressure as well as recommendations and plans to change damaging habits. […] In trials, the system predicted 7.6 percent more events than traditional methods with 1.6 percent fewer false alarms. […] The European Society of Cardiology found that measuring levels of a molecule called trimethylamine N-oxide (TMAO), which is produced by gut bacteria, could help them reliably assess a patient’s risk of heart problems. […] Quyyumi is interested in looking at plaques in the bloodstream as predictors. […] “Can we try to detect activation of plaque by looking for blood markers?” […] Today, though, he’s more optimistic. “I think it is possible. We can diagnose the condition much better. It’s possible this will become less of a problem. We’re moving in the right direction.”

• #57 Acute myocardial infarction prognosis prediction with reliable and interpretable artificial intelligence system

  https://pmc.ncbi.nlm.nih.gov/articles/PMC11187491/

  Predicting mortality after acute myocardial infarction (AMI) is crucial for timely prescription and treatment of AMI patients, but there are no appropriate AI systems for clinicians. Our primary goal is to develop a reliable and interpretable AI system and provide some valuable insights regarding short, and long-term mortality. […] RIAS reveals the significance of statin-based medications, beta-blockers, and age on mortality regardless of time period. […] RIAS addresses the black-box issue in AI by providing both global and local explanations based on SHAP values and reliable predictions, interpretable as actual likelihoods. […] The proposed framework provides reliable and interpretable predictions along with counterfactual examples. […] To deliver timely intervention and personalized therapy for AMI patients, it is crucial to assess prognosis and predict short-term mortality.

• #58 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Fortunately, some doctors and researchers are putting that forward momentum to good use when it comes to heart attack prediction. These scientists are working tirelessly to find ways to predict heart attacks years before any symptoms arise — because early prediction means early intervention, and early intervention can save lives. […] By gathering data, test results and patient information, cardiologists like Quyyumi can generate a score that indicates a patient’s heart attack risk. […] “It’s not an exact prediction,” says Quyyumi. “We use the scores to reduce risk and to prevent disease, heart attack or sudden cardiac death. They’re now developing new technologies which might get us to better predict [cardiovascular disease].” […] His system can predict heart attacks 10 years before they occur with 76 percent accuracy.

• #59 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Depending on the patient’s cardiac forecast, their physician may begin appropriate interventions with medications to manage cholesterol, diabetes and high blood pressure as well as recommendations and plans to change damaging habits. […] In trials, the system predicted 7.6 percent more events than traditional methods with 1.6 percent fewer false alarms. […] The European Society of Cardiology found that measuring levels of a molecule called trimethylamine N-oxide (TMAO), which is produced by gut bacteria, could help them reliably assess a patient’s risk of heart problems. […] Quyyumi is interested in looking at plaques in the bloodstream as predictors. […] “Can we try to detect activation of plaque by looking for blood markers?” […] Today, though, he’s more optimistic. “I think it is possible. We can diagnose the condition much better. It’s possible this will become less of a problem. We’re moving in the right direction.”

• #60 Cardiovascular Event Prediction by Machine Learning: The Multi-Ethnic Study of Atherosclerosis

  https://pmc.ncbi.nlm.nih.gov/articles/PMC5640485/

  Machine learning may be useful to characterize cardiovascular risk, predict outcomes and identify biomarkers in population studies. […] To test the ability of random survival forests (RF), a machine learning technique, to predict six cardiovascular outcomes in comparison to standard cardiovascular risk scores. […] The RF technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 1025%). […] Machine learning in conjunction with deep phenotyping improve prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. […] Increasing age, perhaps reflecting duration of risk exposure, was the most important predictor of all-cause death. […] Inflammation and immune response measured by increased interleukin-6, fibrinogen, homocysteine, TNF- SR and IL2 SR levels; and abnormal hemostasis measured by increased D-Dimer, plasmin-antiplasmin complex and factor VIII levels were among the top 20 markers of all-cause death underlining the role of inflammation and thrombosis as common pathways for chronic diseases leading to death.

• #61

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #62 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  Therefore, deriving distinct optimal thresholds for patients with and without a prior diagnosis of heart failure – and potentially other key subgroups – has the potential to improve the practical clinical utility of NT-proBNP testing. […] Overall, my thesis findings suggest that improvements in the prevention and treatment of heart failure have had progressive and sustained effects on the incident AHF hospitalisation and outcomes at the population and subgroup level although the overall prognosis remains poor. […] Our analysis of the sub-population of patients with and without heart failure history with NT-proBNP suggests that there is scope to optimise the rule-out and rule-in thresholds in patients without a prior heart failure history.

• #63

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #64

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #65

  https://link.springer.com/article/10.1007/s11845-020-02477-z

  Heart failure (HF) is the only cardiovascular disease with an ever increasing incidence. […] The main challenge in the treatment of HF is the availability of reliable prognostic models that would allow patients and doctors to develop realistic expectations about the prognosis and to choose the appropriate therapy and monitoring method. […] Prognosis assessment plays a special role in patients qualified for implantable device therapy or surgical treatment (including heart transplantation). […] Not only does the predictor allow one to identify a high-risk patient in advance but it also allows one to monitor and implement individual preventive therapy. […] However, no single risk factor is sufficient to predict prognosis in HF. Results of a few markers must be interpreted together. […] Despite the identification of many markers and models of poor prognosis, clinical decisions and guidelines in HF are still based mainly on several basic parameters, such as the presence of HF symptoms (NYHA class), LVEF, and the duration and morphology of the QRS complex. […] All of the presented models have shown only moderate probability in predicting death in HF. […] Nevertheless, there is no possibility at the moment to assess and monitor HF with a single parameter or a simple scale that would apply to the whole population of patients.

• #66 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  Therefore, deriving distinct optimal thresholds for patients with and without a prior diagnosis of heart failure – and potentially other key subgroups – has the potential to improve the practical clinical utility of NT-proBNP testing. […] Overall, my thesis findings suggest that improvements in the prevention and treatment of heart failure have had progressive and sustained effects on the incident AHF hospitalisation and outcomes at the population and subgroup level although the overall prognosis remains poor. […] Our analysis of the sub-population of patients with and without heart failure history with NT-proBNP suggests that there is scope to optimise the rule-out and rule-in thresholds in patients without a prior heart failure history.

• #67 The Future of Heart Attack Prediction – Mended Hearts

  https://mendedhearts.org/story/the-future-of-heart-attack-prediction/

  Depending on the patient’s cardiac forecast, their physician may begin appropriate interventions with medications to manage cholesterol, diabetes and high blood pressure as well as recommendations and plans to change damaging habits. […] In trials, the system predicted 7.6 percent more events than traditional methods with 1.6 percent fewer false alarms. […] The European Society of Cardiology found that measuring levels of a molecule called trimethylamine N-oxide (TMAO), which is produced by gut bacteria, could help them reliably assess a patient’s risk of heart problems. […] Quyyumi is interested in looking at plaques in the bloodstream as predictors. […] “Can we try to detect activation of plaque by looking for blood markers?” […] Today, though, he’s more optimistic. “I think it is possible. We can diagnose the condition much better. It’s possible this will become less of a problem. We’re moving in the right direction.”

• #68 Acute heart failure: Scottish temporal trends in hospitalisation, outcomes and clinical application of natriuretic peptides

  https://era.ed.ac.uk/handle/1842/43408

  Therefore, deriving distinct optimal thresholds for patients with and without a prior diagnosis of heart failure – and potentially other key subgroups – has the potential to improve the practical clinical utility of NT-proBNP testing. […] Overall, my thesis findings suggest that improvements in the prevention and treatment of heart failure have had progressive and sustained effects on the incident AHF hospitalisation and outcomes at the population and subgroup level although the overall prognosis remains poor. […] Our analysis of the sub-population of patients with and without heart failure history with NT-proBNP suggests that there is scope to optimise the rule-out and rule-in thresholds in patients without a prior heart failure history.

Zobacz więcej