M. van Vliet, S. Monnink, M. Kuiper, J. Constandse, D. Hoftijzer, E. Ronner
Elevated blood pressure is a key risk factor in cardiovascular diseases. However, obtaining reliable and reproducible blood pressure remains a challenge. This study, therefore, aimed to evaluate a novel cuffless wristband, based on photoplethysmography, for continuous blood pressure monitoring. Predictions by a photoplethysmography-guided algorithm were compared to arterial blood pressure measurements (in the subclavian artery), obtained during cardiac catheterisation. Eligible patients were included and screened based on AAMI/ESH/ISO Universal Standard requirements. The machine learning-based blood pressure algorithm required three cuff-based initialisation measurements in combination with approximately 100 features (signal-derived and patient demographic-based). 97 patients and 420 samples were included. Mean age, weight, and height were 67.1 years (SD 11.1), 83.4 kg (SD 16.1), and 174 cm (SD 10), respectively. Systolic blood pressure was ≤100 mmHg in 48 samples (11%) and ≥160 mmHg in 106 samples (25%). Diastolic blood pressure was ≤70 mmHg in 222 samples (53%) and ≥85 mmHg in 99 samples (24%). The algorithm showed mean errors of ±3.7 mmHg (SD 4.4 mmHg) and ±2.5 mmHg (SD 3.7 mmHg) for systolic and diastolic blood pressure, respectively. Similar results were observed across all genders and skin colours (Fitzpatrick I-VI). This study provides initial evidence for the accuracy of a photoplethysmography-based blood pressure algorithm in combination with a cuffless wristband across a range of blood pressure distributions. This research complies with the AAMI/ESH/ISO Universal Standard, however, further research is required to evaluate the algorithms performance in light of the remaining European Society of Hypertension recommendations. Trial registration: www.clinicaltrials.gov, NCT05566886.
{"title":"Evaluation of a novel cuffless photoplethysmography-based wristband for measuring blood pressure according to the regulatory standards","authors":"M. van Vliet, S. Monnink, M. Kuiper, J. Constandse, D. Hoftijzer, E. Ronner","doi":"10.1093/ehjdh/ztae006","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae006","url":null,"abstract":"\u0000 \u0000 \u0000 Elevated blood pressure is a key risk factor in cardiovascular diseases. However, obtaining reliable and reproducible blood pressure remains a challenge. This study, therefore, aimed to evaluate a novel cuffless wristband, based on photoplethysmography, for continuous blood pressure monitoring.\u0000 \u0000 \u0000 \u0000 Predictions by a photoplethysmography-guided algorithm were compared to arterial blood pressure measurements (in the subclavian artery), obtained during cardiac catheterisation. Eligible patients were included and screened based on AAMI/ESH/ISO Universal Standard requirements. The machine learning-based blood pressure algorithm required three cuff-based initialisation measurements in combination with approximately 100 features (signal-derived and patient demographic-based).\u0000 \u0000 \u0000 \u0000 97 patients and 420 samples were included. Mean age, weight, and height were 67.1 years (SD 11.1), 83.4 kg (SD 16.1), and 174 cm (SD 10), respectively. Systolic blood pressure was ≤100 mmHg in 48 samples (11%) and ≥160 mmHg in 106 samples (25%). Diastolic blood pressure was ≤70 mmHg in 222 samples (53%) and ≥85 mmHg in 99 samples (24%). The algorithm showed mean errors of ±3.7 mmHg (SD 4.4 mmHg) and ±2.5 mmHg (SD 3.7 mmHg) for systolic and diastolic blood pressure, respectively. Similar results were observed across all genders and skin colours (Fitzpatrick I-VI).\u0000 \u0000 \u0000 \u0000 This study provides initial evidence for the accuracy of a photoplethysmography-based blood pressure algorithm in combination with a cuffless wristband across a range of blood pressure distributions. This research complies with the AAMI/ESH/ISO Universal Standard, however, further research is required to evaluate the algorithms performance in light of the remaining European Society of Hypertension recommendations.\u0000 Trial registration: www.clinicaltrials.gov, NCT05566886.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":" 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139792608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Permanent pacemaker implantation and left bundle branch block are common complications after transcatheter aortic valve replacement (TAVR) and are associated with impaired prognosis. This study aimed to develop an artificial intelligence (AI) model for predicting conduction disturbances after TAVR using preprocedural 12-lead electrocardiogram (ECG) data. We collected preprocedural 12-lead ECGs of patients who underwent TAVR at West China Hospital between March 2016 and March 2022. A hold-out testing set comprising 20% of the sample was randomly selected. We developed an AI model using a convolutional neural network, trained it using fivefold cross validation, and tested it on the hold-out testing cohort. We also developed and validated an enhanced model that included additional clinical features. After applying exclusion criteria, we included 1354 ECGs of 718 patients in the study. The AI model predicted conduction disturbances in the hold-out testing cohort with an AUC of 0.764, accuracy of 0.743, F1 score of 0.752, sensitivity of 0.876, and specificity of 0.624, based solely on pre-procedural ECG data. The performance was better than the Emory score (AUC = 0.704), as well as the Logistic (AUC = 0.574) and XGboost (AUC = 0.520) models built with previously identified high-risk ECG patterns. After adding clinical features, there was an increase in the overall performance with an AUC of 0.779, accuracy of 0.774, F1 score of 0.776, sensitivity of 0.794, and specificity of 0.752. AI-enhanced ECGs may offer better predictive value than traditionally defined high-risk ECG patterns.
{"title":"ECG-based Prediction of Conduction Disturbances after Transcatheter Aortic Valve Replacement with Convolutional Neural Network","authors":"Yuheng Jia, Yiming Li, Gaden Luosang, Jianyong Wang, Gang Peng, Xingzhou Pu, Weili Jiang, Wenjian Li, Zhengang Zhao, Yong Peng, Yuan Feng, Jiafu Wei, Yuanning Xu, Xingbin Liu, Zhang Yi, Mao Chen","doi":"10.1093/ehjdh/ztae007","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae007","url":null,"abstract":"\u0000 \u0000 \u0000 Permanent pacemaker implantation and left bundle branch block are common complications after transcatheter aortic valve replacement (TAVR) and are associated with impaired prognosis.\u0000 \u0000 \u0000 \u0000 This study aimed to develop an artificial intelligence (AI) model for predicting conduction disturbances after TAVR using preprocedural 12-lead electrocardiogram (ECG) data.\u0000 \u0000 \u0000 \u0000 We collected preprocedural 12-lead ECGs of patients who underwent TAVR at West China Hospital between March 2016 and March 2022. A hold-out testing set comprising 20% of the sample was randomly selected. We developed an AI model using a convolutional neural network, trained it using fivefold cross validation, and tested it on the hold-out testing cohort. We also developed and validated an enhanced model that included additional clinical features.\u0000 \u0000 \u0000 \u0000 After applying exclusion criteria, we included 1354 ECGs of 718 patients in the study. The AI model predicted conduction disturbances in the hold-out testing cohort with an AUC of 0.764, accuracy of 0.743, F1 score of 0.752, sensitivity of 0.876, and specificity of 0.624, based solely on pre-procedural ECG data. The performance was better than the Emory score (AUC = 0.704), as well as the Logistic (AUC = 0.574) and XGboost (AUC = 0.520) models built with previously identified high-risk ECG patterns. After adding clinical features, there was an increase in the overall performance with an AUC of 0.779, accuracy of 0.774, F1 score of 0.776, sensitivity of 0.794, and specificity of 0.752.\u0000 \u0000 \u0000 \u0000 AI-enhanced ECGs may offer better predictive value than traditionally defined high-risk ECG patterns.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"82 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139851138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Heather J Ross, Mohammad Peikari, J. K. Vishram-Nielsen, C. Fan, Jason Hearn, M. Walker, Edgar Crowdy, A. C. Alba, Cedric Manlhiot
Mathematical models previously developed to predict outcomes in patients with heart failure (HF) generally have limited performance, and have yet to integrate complex data derived from cardiopulmonary exercise testing (CPET), including breath-by-breath data. We aimed to develop and validate a time-to-event prediction model using a deep learning framework using the DeepSurv algorithm to predict outcomes of HF. Inception cohort of 2,490 adult patients with heart failure underwent CPET with breath-by-breath measurements. Potential predictive features included known clinical indicators, standard summary statistics from CPETs and mathematical features extracted from the breath-by-breath time series of 13 measurements. The primary outcome was a composite of death, heart transplant or mechanical circulatory support treated as a time-to-event outcomes. Predictive features ranked as most important included many of the features engineered from the breath-by-breath data in addition to traditional clinical risk factors. The prediction model showed excellent performance in predicting the composite outcome with an AUROC of 0.93 in the training and 0.87 in the validation datasets. Both the predicted vs. actual freedom from the composite outcome and the calibration of the prediction model were excellent. Model performance remained stable in multiple subgroups of patients. Using a combined deep learning and survival algorithm, integrating breath-by-breath data from CPETs, resulted in improved predictive accuracy for long term (up to 10 years) outcomes in HF. DeepSurv opens the door for future prediction models that are both highly performing and can more fully use the large and complex quantity of data generated during the care of patients with heart failure.
{"title":"Predicting heart failure outcomes by integrating breath-by-breath measurements from cardiopulmonary exercise testing and clinical data through a deep learning survival neural network","authors":"Heather J Ross, Mohammad Peikari, J. K. Vishram-Nielsen, C. Fan, Jason Hearn, M. Walker, Edgar Crowdy, A. C. Alba, Cedric Manlhiot","doi":"10.1093/ehjdh/ztae005","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae005","url":null,"abstract":"\u0000 \u0000 \u0000 Mathematical models previously developed to predict outcomes in patients with heart failure (HF) generally have limited performance, and have yet to integrate complex data derived from cardiopulmonary exercise testing (CPET), including breath-by-breath data. We aimed to develop and validate a time-to-event prediction model using a deep learning framework using the DeepSurv algorithm to predict outcomes of HF.\u0000 \u0000 \u0000 \u0000 Inception cohort of 2,490 adult patients with heart failure underwent CPET with breath-by-breath measurements. Potential predictive features included known clinical indicators, standard summary statistics from CPETs and mathematical features extracted from the breath-by-breath time series of 13 measurements. The primary outcome was a composite of death, heart transplant or mechanical circulatory support treated as a time-to-event outcomes.\u0000 \u0000 \u0000 \u0000 Predictive features ranked as most important included many of the features engineered from the breath-by-breath data in addition to traditional clinical risk factors. The prediction model showed excellent performance in predicting the composite outcome with an AUROC of 0.93 in the training and 0.87 in the validation datasets. Both the predicted vs. actual freedom from the composite outcome and the calibration of the prediction model were excellent. Model performance remained stable in multiple subgroups of patients.\u0000 \u0000 \u0000 \u0000 Using a combined deep learning and survival algorithm, integrating breath-by-breath data from CPETs, resulted in improved predictive accuracy for long term (up to 10 years) outcomes in HF. DeepSurv opens the door for future prediction models that are both highly performing and can more fully use the large and complex quantity of data generated during the care of patients with heart failure.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"13 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140478651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mitchel A. Molenaar, B. Bouma, F. Asselbergs, Niels J Verouden, J. Selder, Steven A J Chamuleau, Mark J Schuuring
The European Society of Cardiology guidelines recommend risk stratification with limited clinical parameters such as left ventricular (LV) function in patients with chronic coronary syndrome (CCS). Machine learning (ML) methods enable analysis of complex datasets including transthoracic echocardiography (TTE) studies. We aimed to evaluate the accuracy of ML using clinical and TTE data to predict all-cause five-year mortality in patients with CCS and to compare its performance with traditional risk stratification scores. Data of consecutive patients with CCS were retrospectively collected if they attended the outpatient clinic of Amsterdam UMC location AMC between 2015 and 2017 and had TTE assessment of the LV function. An eXtreme Gradient Boosting (XGBoost) model was trained to predict all-cause five-year mortality. The performance of this ML model was evaluated using data of the Amsterdam UMC location VUmc and compared to the reference standard of traditional risk scores. A total of 1253 patients (775 training set, 478 testing set) were included, of which 176 patients (105 training set, 71 testing set) died during the five-year follow-up period. The ML model demonstrated a superior performance (area under the curve [AUC] 0.79) compared to traditional risk stratification tools (AUC 0.62-0.76), and showed good external performance. The most important TTE risk predictors included in the ML model were LV dysfunction and significant tricuspid regurgitation. This study demonstrates that an explainable ML model using TTE and clinical data can accurately identify high-risk CCS patients, with a prognostic value superior to traditional risk scores.
{"title":"Explainable machine learning using echocardiography to improve risk prediction in patients with chronic coronary syndrome","authors":"Mitchel A. Molenaar, B. Bouma, F. Asselbergs, Niels J Verouden, J. Selder, Steven A J Chamuleau, Mark J Schuuring","doi":"10.1093/ehjdh/ztae001","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae001","url":null,"abstract":"\u0000 \u0000 \u0000 The European Society of Cardiology guidelines recommend risk stratification with limited clinical parameters such as left ventricular (LV) function in patients with chronic coronary syndrome (CCS). Machine learning (ML) methods enable analysis of complex datasets including transthoracic echocardiography (TTE) studies. We aimed to evaluate the accuracy of ML using clinical and TTE data to predict all-cause five-year mortality in patients with CCS and to compare its performance with traditional risk stratification scores.\u0000 \u0000 \u0000 \u0000 Data of consecutive patients with CCS were retrospectively collected if they attended the outpatient clinic of Amsterdam UMC location AMC between 2015 and 2017 and had TTE assessment of the LV function. An eXtreme Gradient Boosting (XGBoost) model was trained to predict all-cause five-year mortality. The performance of this ML model was evaluated using data of the Amsterdam UMC location VUmc and compared to the reference standard of traditional risk scores. A total of 1253 patients (775 training set, 478 testing set) were included, of which 176 patients (105 training set, 71 testing set) died during the five-year follow-up period. The ML model demonstrated a superior performance (area under the curve [AUC] 0.79) compared to traditional risk stratification tools (AUC 0.62-0.76), and showed good external performance. The most important TTE risk predictors included in the ML model were LV dysfunction and significant tricuspid regurgitation.\u0000 \u0000 \u0000 \u0000 This study demonstrates that an explainable ML model using TTE and clinical data can accurately identify high-risk CCS patients, with a prognostic value superior to traditional risk scores.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"53 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139608379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alejandra Zepeda-Echavarria, R. R. van de Leur, Melle Vessies, Nynke M. de Vries, Meike van Sleuwen, R. Hassink, T. Wildbergh, J. L. van Doorn, Rien van der Zee, Pieter A. Doevendans, J. Jaspers, R. van Es
Many portable ECG devices have been developed to monitor patients at home, but the majority of these devices are single lead, and only intended for rhythm disorders. We developed the miniECG, a smartphone sized portable device with four dry electrodes capable of recording a high-quality multi-lead ECG by placing the device on the chest. The aim of our study was to investigate the ability of the miniECG to detect occlusive myocardial infarction (OMI) in patients with chest pain. Patients presenting with acute chest pain at the emergency department of the University Medical Center Utrecht or Meander Medical Center, between May 2021 and February 2022 were included in the study. The clinical 12-lead ECG and the miniECG before coronary intervention were recorded. The recordings were evaluated by cardiologists and compared the outcome of the coronary angiography, if performed. A total of 369 patients were measured with the miniECG, 46 of whom had OMI. The miniECG detected OMI with a sensitivity and specificity of 65% and 92%, compared to 83% and 90% for the 12-lead ECG. Sensitivity of the miniECG was similar for different culprit vessels. The miniECG can record a multi-lead ECG and rule-in ST-segment deviation in patients with occluded or near occluded coronary arteries from different culprit vessels without many false alarms. Further research is required to add automated analysis to the recordings and to show feasibility to use the miniECG by patients at home.
{"title":"Detection of Acute Coronary Occlusion with A Novel Mobile ECG Device: A Pilot Study","authors":"Alejandra Zepeda-Echavarria, R. R. van de Leur, Melle Vessies, Nynke M. de Vries, Meike van Sleuwen, R. Hassink, T. Wildbergh, J. L. van Doorn, Rien van der Zee, Pieter A. Doevendans, J. Jaspers, R. van Es","doi":"10.1093/ehjdh/ztae002","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae002","url":null,"abstract":"\u0000 \u0000 \u0000 Many portable ECG devices have been developed to monitor patients at home, but the majority of these devices are single lead, and only intended for rhythm disorders. We developed the miniECG, a smartphone sized portable device with four dry electrodes capable of recording a high-quality multi-lead ECG by placing the device on the chest.\u0000 \u0000 \u0000 \u0000 The aim of our study was to investigate the ability of the miniECG to detect occlusive myocardial infarction (OMI) in patients with chest pain.\u0000 \u0000 \u0000 \u0000 Patients presenting with acute chest pain at the emergency department of the University Medical Center Utrecht or Meander Medical Center, between May 2021 and February 2022 were included in the study. The clinical 12-lead ECG and the miniECG before coronary intervention were recorded. The recordings were evaluated by cardiologists and compared the outcome of the coronary angiography, if performed.\u0000 \u0000 \u0000 \u0000 A total of 369 patients were measured with the miniECG, 46 of whom had OMI. The miniECG detected OMI with a sensitivity and specificity of 65% and 92%, compared to 83% and 90% for the 12-lead ECG. Sensitivity of the miniECG was similar for different culprit vessels.\u0000 \u0000 \u0000 \u0000 The miniECG can record a multi-lead ECG and rule-in ST-segment deviation in patients with occluded or near occluded coronary arteries from different culprit vessels without many false alarms. Further research is required to add automated analysis to the recordings and to show feasibility to use the miniECG by patients at home.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"12 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139527455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Deepak Balamurali, M. Preda, S. Ben-Aicha, Fabiana Martino, Dimitra Palioura, Jordy M M Kocken, C. Emanueli, Yvan Devaux
Journal clubs have been a staple in scientific communities, facilitating discussions on recent publications. However, the overwhelming volume of biomedical information poses a challenge in literature selection. This article provides an overview of journal club types and their efficacy in training potential peer reviewers, enhancing communication skills, and critical thinking. Originating in the 19th century, journal clubs have evolved from traditional in-person meetings to virtual or hybrid formats, accelerated by the COVID-19 pandemic. Face-to-face interactions offer personal connections, while virtual events ensure wider participation and accessibility. Organizing journal clubs demands effort, but it has several benefits, including promoting new publications and providing a platform for meaningful discussions. The virtual CardioRNA J-club experience exemplifies successful multidisciplinary collaboration, fostering international connections and inspiring new research. Journal clubs remain a vital component of academic research, equipping senior researchers with the latest developments and nurturing the next generation of scientists. As millennial and Gen Z researchers join the scientific field, journal clubs continue to evolve as a fertile ground for education and collaborative learning in an ever-changing scientific landscape.
{"title":"Evolution of Journal Clubs: Fostering Collaborative Learning in Modern Research","authors":"Deepak Balamurali, M. Preda, S. Ben-Aicha, Fabiana Martino, Dimitra Palioura, Jordy M M Kocken, C. Emanueli, Yvan Devaux","doi":"10.1093/ehjdh/ztae003","DOIUrl":"https://doi.org/10.1093/ehjdh/ztae003","url":null,"abstract":"\u0000 Journal clubs have been a staple in scientific communities, facilitating discussions on recent publications. However, the overwhelming volume of biomedical information poses a challenge in literature selection. This article provides an overview of journal club types and their efficacy in training potential peer reviewers, enhancing communication skills, and critical thinking. Originating in the 19th century, journal clubs have evolved from traditional in-person meetings to virtual or hybrid formats, accelerated by the COVID-19 pandemic. Face-to-face interactions offer personal connections, while virtual events ensure wider participation and accessibility. Organizing journal clubs demands effort, but it has several benefits, including promoting new publications and providing a platform for meaningful discussions. The virtual CardioRNA J-club experience exemplifies successful multidisciplinary collaboration, fostering international connections and inspiring new research. Journal clubs remain a vital component of academic research, equipping senior researchers with the latest developments and nurturing the next generation of scientists. As millennial and Gen Z researchers join the scientific field, journal clubs continue to evolve as a fertile ground for education and collaborative learning in an ever-changing scientific landscape.","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"49 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139526664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Victor M Gallegos-Rejas, Johnathan Rawstorn, Robyn Gallagher, Ray Mahoney, E. Thomas
Telehealth-delivered cardiac rehabilitation (CR) programmes can potentially increase participation rates while delivering equivalent outcomes to facility-based programmes. However, key components of these interventions that reduce cardiovascular risk factors are not yet distinguished. This study aims to identify features of telehealth-delivered CR that improve secondary prevention outcomes, exercise capacity, participation, and participant satisfaction; and develop recommendations for future telehealth-delivered CR. The protocol for our review was registered with the Prospective Register of Systematic Reviews (#CRD42021236471). We systematically searched four databases (PubMed, Scopus, EMBASE and Cochrane Database) for randomised controlled trials comparing telehealth-delivered CR programmes to facility-based interventions or usual care. Two independent reviewers screened the abstracts and then full-texts. Using a qualitative review methodology (realist synthesis), included articles were evaluated to determine contextual factors and potential mechanisms that impacted cardiovascular risk factors, exercise capacity, participation in the intervention, and increased satisfaction. We included 37 reports describing 26 randomised controlled trials published from 2010 to 2022. Studies were primarily conducted in Europe and Australia/Asia. Identified contextual factors and mechanisms were synthesised into four theories required to enhance participant outcomes and participation. These theories are: 1) Early and regular engagement; 2) Personalised interventions and shared goals; 3) Usable, accessible, and supported interventions; and 4) Exercise that is measured and monitored. Providing a personalised approach with frequent opportunities for bi-directional interaction were critical features for success across telehealth-delivered CR trials. Real-world effectiveness studies are now needed to complement our findings.
{"title":"Key features in telehealth-delivered cardiac rehabilitation required to optimise cardiovascular health in coronary heart disease: A systematic review and realist synthesis","authors":"Victor M Gallegos-Rejas, Johnathan Rawstorn, Robyn Gallagher, Ray Mahoney, E. Thomas","doi":"10.1093/ehjdh/ztad080","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad080","url":null,"abstract":"\u0000 \u0000 \u0000 Telehealth-delivered cardiac rehabilitation (CR) programmes can potentially increase participation rates while delivering equivalent outcomes to facility-based programmes. However, key components of these interventions that reduce cardiovascular risk factors are not yet distinguished. This study aims to identify features of telehealth-delivered CR that improve secondary prevention outcomes, exercise capacity, participation, and participant satisfaction; and develop recommendations for future telehealth-delivered CR.\u0000 \u0000 \u0000 \u0000 The protocol for our review was registered with the Prospective Register of Systematic Reviews (#CRD42021236471). We systematically searched four databases (PubMed, Scopus, EMBASE and Cochrane Database) for randomised controlled trials comparing telehealth-delivered CR programmes to facility-based interventions or usual care. Two independent reviewers screened the abstracts and then full-texts. Using a qualitative review methodology (realist synthesis), included articles were evaluated to determine contextual factors and potential mechanisms that impacted cardiovascular risk factors, exercise capacity, participation in the intervention, and increased satisfaction.\u0000 \u0000 \u0000 \u0000 We included 37 reports describing 26 randomised controlled trials published from 2010 to 2022. Studies were primarily conducted in Europe and Australia/Asia. Identified contextual factors and mechanisms were synthesised into four theories required to enhance participant outcomes and participation. These theories are: 1) Early and regular engagement; 2) Personalised interventions and shared goals; 3) Usable, accessible, and supported interventions; and 4) Exercise that is measured and monitored.\u0000 \u0000 \u0000 \u0000 Providing a personalised approach with frequent opportunities for bi-directional interaction were critical features for success across telehealth-delivered CR trials. Real-world effectiveness studies are now needed to complement our findings.\u0000","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"33 47","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139382727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sebastian König, Sven Hohenstein, A. Nitsche, V. Pellissier, J. Leiner, Lars Stellmacher, G. Hindricks, Andreas Bollmann
The diagnostic application of artificial intelligence (AI)-based models to detect cardiovascular diseases from electrocardiograms (ECG) evolves and promising results were reported. However, external validation is not available for all published algorithms. Aim of this study was to validate an existing algorithm for the detection of left ventricular systolic dysfunction (LVSD) from 12-lead ECGs. Patients with digitalized data pairs of 12-lead ECGs and echocardiography (at intervals of ≤7 days) were retrospectively selected from the Heart Center Leipzig ECG and electronic medical records databases. A previously developed AI-based model was applied to ECGs and calculated probabilities for LVSD. The area under the receiver operating characteristic curve (AUROC) was computed overall and in cohorts stratified for baseline and ECG characteristics. Repeated echocardiography studies recorded ≥3 months after index diagnostics were used for follow-up (FU) analysis. At baseline, 42,291 ECG-echocardiography pairs were analyzed and AUROC for LVSD detection was 0.88. Sensitivity and specificity were 82% and 77% for the optimal LVSD-probability cutoff based on Youden’s J. AUROCs were lower in ECG-subgroups with tachycardia, atrial fibrillation and wide QRS complexes. In patients without LVSD at baseline and available FU, model-generated high probability for LVSD was associated with a 4-fold increased risk of developing LVSD during FU. We provide the external validation of an existing AI-based ECG-analyzing model for the detection of LVSD with robust performance metrics. The association of false positive LVSD screenings at baseline with a deterioration of ventricular function during FU deserves a further evaluation in prospective trials.
{"title":"Artificial intelligence-based identification of left ventricular systolic dysfunction from 12-lead electrocardiograms: External validation and advanced application of an existing model","authors":"Sebastian König, Sven Hohenstein, A. Nitsche, V. Pellissier, J. Leiner, Lars Stellmacher, G. Hindricks, Andreas Bollmann","doi":"10.1093/ehjdh/ztad081","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad081","url":null,"abstract":"The diagnostic application of artificial intelligence (AI)-based models to detect cardiovascular diseases from electrocardiograms (ECG) evolves and promising results were reported. However, external validation is not available for all published algorithms. Aim of this study was to validate an existing algorithm for the detection of left ventricular systolic dysfunction (LVSD) from 12-lead ECGs. Patients with digitalized data pairs of 12-lead ECGs and echocardiography (at intervals of ≤7 days) were retrospectively selected from the Heart Center Leipzig ECG and electronic medical records databases. A previously developed AI-based model was applied to ECGs and calculated probabilities for LVSD. The area under the receiver operating characteristic curve (AUROC) was computed overall and in cohorts stratified for baseline and ECG characteristics. Repeated echocardiography studies recorded ≥3 months after index diagnostics were used for follow-up (FU) analysis. At baseline, 42,291 ECG-echocardiography pairs were analyzed and AUROC for LVSD detection was 0.88. Sensitivity and specificity were 82% and 77% for the optimal LVSD-probability cutoff based on Youden’s J. AUROCs were lower in ECG-subgroups with tachycardia, atrial fibrillation and wide QRS complexes. In patients without LVSD at baseline and available FU, model-generated high probability for LVSD was associated with a 4-fold increased risk of developing LVSD during FU. We provide the external validation of an existing AI-based ECG-analyzing model for the detection of LVSD with robust performance metrics. The association of false positive LVSD screenings at baseline with a deterioration of ventricular function during FU deserves a further evaluation in prospective trials.","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"923 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139169990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Y. Mizuguchi, M. Nakao, T. Nagai, Y. Takahashi, Takahiro Abe, Shigeo Kakinoki, S. Imagawa, Kenichi Matsutani, Takahiko Saito, Masashige Takahashi, Yoshiya Kato, Hirokazu Komoriyama, H. Hagiwara, Kenji Hirata, Takahiro Ogawa, Takuto Shimizu, Manabu Otsu, Kunihiro Chiyo, Toshihisa Anzai
Although frailty assessment is recommended for guiding treatment strategies and outcome prediction in elderly patients with heart failure (HF), most frailty scales are subjective and the scores vary among raters. We sought to develop a machine learning-based automatic rating method/system/model of the clinical frailty scale (CFS) for patients with HF. We prospectively examined 417 elderly (≥75 years) with symptomatic chronic HF patients from seven centers between January 2019 and October 2023. The patients were divided into derivation (n = 194) and validation (n = 223) cohorts. We obtained body-tracking motion data using a deep learning-based pose estimation library, on a smartphone camera. Predicted CFS was calculated from 128 key features, including gait parameters, using the Light Gradient Boosting Machine (LightGBM) model. To evaluate the performance of this model, we calculated Cohen’s weighted kappa (CWK) and intraclass correlation coefficient (ICC) between the predicted and actual CFSs. In the derivation and validation datasets, the LightGBM models showed excellent agreements between the actual and predicted CFSs (CWK 0.866, 95% CI 0.807-0.911; ICC 0.866, 95% CI 0.827-0.898; CWK 0.812, 95% CI 0.752-0.868; ICC 0.813, 95% CI 0.761-0.854, respectively). During a median follow-up period of 391 (IQR 273-617) days, the higher predicted CFS was independently associated with a higher risk of all-cause death (HR 1.60, 95% CI 1.02-2.50) after adjusting for significant prognostic covariates. Machine learning-based algorithms of automatically CFS rating are feasible, and the predicted CFS is associated with the risk of all-cause death in elderly patients with HF.
虽然虚弱程度评估被推荐用于指导老年心力衰竭（HF）患者的治疗策略和预后预测，但大多数虚弱程度量表都是主观的，而且不同评分者的评分也不尽相同。我们试图为心衰患者开发一种基于机器学习的临床虚弱量表（CFS）自动评分方法/系统/模型。 我们在 2019 年 1 月至 2023 年 10 月期间对来自七个中心的 417 名老年（≥75 岁）有症状的慢性心房颤动患者进行了前瞻性研究。患者被分为推导组（194 人）和验证组（223 人）。我们使用基于深度学习的姿势估计库，在智能手机摄像头上获取了身体追踪运动数据。利用光梯度提升机（LightGBM）模型，通过包括步态参数在内的 128 个关键特征计算出预测的 CFS。为了评估该模型的性能，我们计算了预测和实际 CFS 之间的科恩加权卡帕（CWK）和类内相关系数（ICC）。在推导数据集和验证数据集中，LightGBM 模型在实际 CFS 与预测 CFS 之间显示出极佳的一致性（CWK 0.866，95% CI 0.807-0.911；ICC 0.866，95% CI 0.827-0.898；CWK 0.812，95% CI 0.752-0.868；ICC 0.813，95% CI 0.761-0.854）。中位随访期为 391 天（IQR 273-617 天），在调整重要的预后协变量后，预测 CFS 越高，全因死亡风险越高（HR 1.60，95% CI 1.02-2.50）。 基于机器学习的CFS自动评级算法是可行的，预测的CFS与老年心房颤动患者的全因死亡风险相关。
{"title":"Machine Learning-based Gait Analysis to Predict Clinical Frailty Scale in Elderly Patients with Heart Failure","authors":"Y. Mizuguchi, M. Nakao, T. Nagai, Y. Takahashi, Takahiro Abe, Shigeo Kakinoki, S. Imagawa, Kenichi Matsutani, Takahiko Saito, Masashige Takahashi, Yoshiya Kato, Hirokazu Komoriyama, H. Hagiwara, Kenji Hirata, Takahiro Ogawa, Takuto Shimizu, Manabu Otsu, Kunihiro Chiyo, Toshihisa Anzai","doi":"10.1093/ehjdh/ztad082","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad082","url":null,"abstract":"Although frailty assessment is recommended for guiding treatment strategies and outcome prediction in elderly patients with heart failure (HF), most frailty scales are subjective and the scores vary among raters. We sought to develop a machine learning-based automatic rating method/system/model of the clinical frailty scale (CFS) for patients with HF. We prospectively examined 417 elderly (≥75 years) with symptomatic chronic HF patients from seven centers between January 2019 and October 2023. The patients were divided into derivation (n = 194) and validation (n = 223) cohorts. We obtained body-tracking motion data using a deep learning-based pose estimation library, on a smartphone camera. Predicted CFS was calculated from 128 key features, including gait parameters, using the Light Gradient Boosting Machine (LightGBM) model. To evaluate the performance of this model, we calculated Cohen’s weighted kappa (CWK) and intraclass correlation coefficient (ICC) between the predicted and actual CFSs. In the derivation and validation datasets, the LightGBM models showed excellent agreements between the actual and predicted CFSs (CWK 0.866, 95% CI 0.807-0.911; ICC 0.866, 95% CI 0.827-0.898; CWK 0.812, 95% CI 0.752-0.868; ICC 0.813, 95% CI 0.761-0.854, respectively). During a median follow-up period of 391 (IQR 273-617) days, the higher predicted CFS was independently associated with a higher risk of all-cause death (HR 1.60, 95% CI 1.02-2.50) after adjusting for significant prognostic covariates. Machine learning-based algorithms of automatically CFS rating are feasible, and the predicted CFS is associated with the risk of all-cause death in elderly patients with HF.","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"49 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139169473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Tromp, Chenik Sarra, Bouchahda Nidhal, Ben Messaoud Mejdi, Fourat Zouari, Yoran Hummel, K. Mzoughi, S. Kraiem, Wafa Fehri, Habib Gamra, Carolyn S P Lam, Alexandre Mebazaa, Faouzi Addad
Access to echocardiography is a significant barrier to heart failure (HF) care in many low- and middle-income countries. We hypothesised that an artificial intelligence (AI) enhanced point-of-care ultrasound (POCUS) device could enable the detection of cardiac dysfunction by nurses in Tunisia. The CUMIN study was a prospective feasibility pilot assessing the diagnostic accuracy of home-based AI-POCUS for HF conducted by novice nurses compared to conventional clinic-based transthoracic echocardiography (TTE). Seven nurses underwent a one-day training program in AI-POCUS. 94 patients without a previous HF diagnosis received home-based AI-POCUS, POC NTproBNP testing, and clinic-based TTE. The primary outcome was the sensitivity of AI-POCUS in detecting left ventricular ejection fraction (LVEF) <50% or left atrial volume index (LAVI) >34 mL/m2, using clinic-based TTE as the reference. Out of 7 nurses, 5 achieved a minimum standard to participate in the study. Out of 94 patients (60% women, median age 67), 16 (17%) had LVEF<50% or LAVI >34 mL/m2. AI-POCUS provided interpretable LVEF in 75 (80%) patients and LAVI in 64 (68%). The only significant predictor of an interpretable LVEF or LAVI proportion was the nurse operator. The sensitivity for the primary outcome was 92% (95% CI 62-99) for AI-POCUS compared to 87% (95% CI 60-98) for NT-proBNP>125 pg/mL, with AI-POCUS having a significantly higher AUC (P=0.040). The study demonstrated the feasibility of novice nurse-led home-based detection of cardiac dysfunction using AI-POCUS in HF patients, which could alleviate the burden on under-resourced healthcare systems
{"title":"Nurse-led home-based detection of cardiac dysfunction by ultrasound: Results of the CUMIN pilot study","authors":"J. Tromp, Chenik Sarra, Bouchahda Nidhal, Ben Messaoud Mejdi, Fourat Zouari, Yoran Hummel, K. Mzoughi, S. Kraiem, Wafa Fehri, Habib Gamra, Carolyn S P Lam, Alexandre Mebazaa, Faouzi Addad","doi":"10.1093/ehjdh/ztad079","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad079","url":null,"abstract":"Access to echocardiography is a significant barrier to heart failure (HF) care in many low- and middle-income countries. We hypothesised that an artificial intelligence (AI) enhanced point-of-care ultrasound (POCUS) device could enable the detection of cardiac dysfunction by nurses in Tunisia. The CUMIN study was a prospective feasibility pilot assessing the diagnostic accuracy of home-based AI-POCUS for HF conducted by novice nurses compared to conventional clinic-based transthoracic echocardiography (TTE). Seven nurses underwent a one-day training program in AI-POCUS. 94 patients without a previous HF diagnosis received home-based AI-POCUS, POC NTproBNP testing, and clinic-based TTE. The primary outcome was the sensitivity of AI-POCUS in detecting left ventricular ejection fraction (LVEF) <50% or left atrial volume index (LAVI) >34 mL/m2, using clinic-based TTE as the reference. Out of 7 nurses, 5 achieved a minimum standard to participate in the study. Out of 94 patients (60% women, median age 67), 16 (17%) had LVEF<50% or LAVI >34 mL/m2. AI-POCUS provided interpretable LVEF in 75 (80%) patients and LAVI in 64 (68%). The only significant predictor of an interpretable LVEF or LAVI proportion was the nurse operator. The sensitivity for the primary outcome was 92% (95% CI 62-99) for AI-POCUS compared to 87% (95% CI 60-98) for NT-proBNP>125 pg/mL, with AI-POCUS having a significantly higher AUC (P=0.040). The study demonstrated the feasibility of novice nurse-led home-based detection of cardiac dysfunction using AI-POCUS in HF patients, which could alleviate the burden on under-resourced healthcare systems","PeriodicalId":508387,"journal":{"name":"European Heart Journal - Digital Health","volume":"59 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139182957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：