Pub Date : 2023-07-06eCollection Date: 2023-10-01DOI: 10.1093/ehjdh/ztad039
Pamela Reissenberger, Peter Serfözö, Diana Piper, Norman Juchler, Sara Glanzmann, Jasmin Gram, Karina Hensler, Hannah Tonidandel, Elena Börlin, Marcus D'Souza, Patrick Badertscher, Jens Eckstein
Aims: Recent studies suggest that atrial fibrillation (AF) burden (time AF is present) is an independent risk factor for stroke. The aim of this trial was to study the feasibility and accuracy to identify AF episodes and quantify AF burden in patients with a known history of paroxysmal AF with a photoplethysmography (PPG)-based wearable.
Methods and results: In this prospective, single-centre trial, the PPG-based estimation of AF burden was compared with measurements of a conventional 48 h Holter electrocardiogram (ECG), which served as the gold standard. An automated algorithm performed PPG analysis, while a cardiologist, blinded for the PPG data, analysed the ECG data. Detected episodes of AF measured by both methods were aligned timewise.Out of 100 patients recruited, 8 had to be excluded due to technical issues. Data from 92 patients were analysed [55.4% male; age 73.3 years (standard deviation, SD: 10.4)]. Twenty-five patients presented AF during the study period. The intraclass correlation coefficient of total AF burden minutes detected by the two measurement methods was 0.88. The percentage of correctly identified AF burden over all patients was 85.1% and the respective parameter for non-AF time was 99.9%.
Conclusion: Our results demonstrate that a PPG-based wearable in combination with an analytical algorithm appears to be suitable for a semiquantitative estimation of AF burden in patients with a known history of paroxysmal AF.
{"title":"Determine atrial fibrillation burden with a photoplethysmographic mobile sensor: the atrial fibrillation burden trial: detection and quantification of episodes of atrial fibrillation using a cloud analytics service connected to a wearable with photoplethysmographic sensor.","authors":"Pamela Reissenberger, Peter Serfözö, Diana Piper, Norman Juchler, Sara Glanzmann, Jasmin Gram, Karina Hensler, Hannah Tonidandel, Elena Börlin, Marcus D'Souza, Patrick Badertscher, Jens Eckstein","doi":"10.1093/ehjdh/ztad039","DOIUrl":"10.1093/ehjdh/ztad039","url":null,"abstract":"<p><strong>Aims: </strong>Recent studies suggest that atrial fibrillation (AF) burden (time AF is present) is an independent risk factor for stroke. The aim of this trial was to study the feasibility and accuracy to identify AF episodes and quantify AF burden in patients with a known history of paroxysmal AF with a photoplethysmography (PPG)-based wearable.</p><p><strong>Methods and results: </strong>In this prospective, single-centre trial, the PPG-based estimation of AF burden was compared with measurements of a conventional 48 h Holter electrocardiogram (ECG), which served as the gold standard. An automated algorithm performed PPG analysis, while a cardiologist, blinded for the PPG data, analysed the ECG data. Detected episodes of AF measured by both methods were aligned timewise.Out of 100 patients recruited, 8 had to be excluded due to technical issues. Data from 92 patients were analysed [55.4% male; age 73.3 years (standard deviation, SD: 10.4)]. Twenty-five patients presented AF during the study period. The intraclass correlation coefficient of total AF burden minutes detected by the two measurement methods was 0.88. The percentage of correctly identified AF burden over all patients was 85.1% and the respective parameter for non-AF time was 99.9%.</p><p><strong>Conclusion: </strong>Our results demonstrate that a PPG-based wearable in combination with an analytical algorithm appears to be suitable for a semiquantitative estimation of AF burden in patients with a known history of paroxysmal AF.</p><p><strong>Trial registration number: </strong>NCT04563572.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/47/42/ztad039.PMC10545505.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41156367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-05eCollection Date: 2023-10-01DOI: 10.1093/ehjdh/ztad038
Lisa Maria Jahre, Julia Lortz, Tienush Rassaf, Christos Rammos, Charlotta Mallien, Eva-Maria Skoda, Martin Teufel, Alexander Bäuerle
Aims: Cardiovascular diseases are one of the main contributors to disability and mortality worldwide. Meanwhile, risk factors can be modified by lifestyle changes. mHealth is an innovative and effective way to deliver cardiac health promotion. This study aims to examine the needs and demands regarding the design and contents of an mHealth intervention for cardiac health promotion among individuals with cardiac diseases. Different clusters were determined and analysed in terms of the intention to use an mHealth intervention.
Methods and results: A cross-sectional study was conducted via a web-based survey. Three hundred and four individuals with coronary artery diseases (CADs) and/or congestive heart failure (CHF) were included in the data analysis. Descriptive statistics were applied to evaluate needs and demands regarding an mHealth intervention. A k-medoids cluster analysis was performed. Individuals with CAD and CHF favoured an mHealth intervention that supports its users permanently and is easily integrated into everyday life. Handheld devices and content formats that involve active user participation and regular updates were preferred. Three clusters were observed and labelled high, moderate, and low burden, according to their psychometric properties. The high burden cluster indicated higher behavioural intention towards use of an mHealth intervention than the other clusters.
Conclusion: The results of the study are a valuable foundation for the development of an mHealth intervention for cardiac health promotion following a user-centred design approach. Individuals with cardiac diseases report positive attitudes in the form of high usage intention regarding mHealth. Highly burdened individuals report a high intention to use such interventions.
{"title":"Needs and demands for mHealth cardiac health promotion among individuals with cardiac diseases: a patient-centred design approach.","authors":"Lisa Maria Jahre, Julia Lortz, Tienush Rassaf, Christos Rammos, Charlotta Mallien, Eva-Maria Skoda, Martin Teufel, Alexander Bäuerle","doi":"10.1093/ehjdh/ztad038","DOIUrl":"10.1093/ehjdh/ztad038","url":null,"abstract":"<p><strong>Aims: </strong>Cardiovascular diseases are one of the main contributors to disability and mortality worldwide. Meanwhile, risk factors can be modified by lifestyle changes. mHealth is an innovative and effective way to deliver cardiac health promotion. This study aims to examine the needs and demands regarding the design and contents of an mHealth intervention for cardiac health promotion among individuals with cardiac diseases. Different clusters were determined and analysed in terms of the intention to use an mHealth intervention.</p><p><strong>Methods and results: </strong>A cross-sectional study was conducted via a web-based survey. Three hundred and four individuals with coronary artery diseases (CADs) and/or congestive heart failure (CHF) were included in the data analysis. Descriptive statistics were applied to evaluate needs and demands regarding an mHealth intervention. A <i>k</i>-medoids cluster analysis was performed. Individuals with CAD and CHF favoured an mHealth intervention that supports its users permanently and is easily integrated into everyday life. Handheld devices and content formats that involve active user participation and regular updates were preferred. Three clusters were observed and labelled high, moderate, and low burden, according to their psychometric properties. The high burden cluster indicated higher behavioural intention towards use of an mHealth intervention than the other clusters.</p><p><strong>Conclusion: </strong>The results of the study are a valuable foundation for the development of an mHealth intervention for cardiac health promotion following a user-centred design approach. Individuals with cardiac diseases report positive attitudes in the form of high usage intention regarding mHealth. Highly burdened individuals report a high intention to use such interventions.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/5c/36/ztad038.PMC10545514.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41159404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-02eCollection Date: 2023-10-01DOI: 10.1093/ehjdh/ztad036
Gerhard Hindricks
{"title":"European Society of Cardiology and Radical Health Festival Helsinki join forces to transform healthcare as we know it.","authors":"Gerhard Hindricks","doi":"10.1093/ehjdh/ztad036","DOIUrl":"10.1093/ehjdh/ztad036","url":null,"abstract":"","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10545521/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41171637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-03eCollection Date: 2023-08-01DOI: 10.1093/ehjdh/ztad030
Henry Seligman, Sapna B Patel, Anissa Alloula, James P Howard, Christopher M Cook, Yousif Ahmad, Guus A de Waard, Mauro Echavarría Pinto, Tim P van de Hoef, Haseeb Rahman, Mihir A Kelshiker, Christopher A Rajkumar, Michael Foley, Alexandra N Nowbar, Samay Mehta, Mathieu Toulemonde, Meng-Xing Tang, Rasha Al-Lamee, Sayan Sen, Graham Cole, Sukhjinder Nijjer, Javier Escaned, Niels Van Royen, Darrel P Francis, Matthew J Shun-Shin, Ricardo Petraco
Aims: Coronary flow reserve (CFR) assessment has proven clinical utility, but Doppler-based methods are sensitive to noise and operator bias, limiting their clinical applicability. The objective of the study is to expand the adoption of invasive Doppler CFR, through the development of artificial intelligence (AI) algorithms to automatically quantify coronary Doppler quality and track flow velocity.
Methods and results: A neural network was trained on images extracted from coronary Doppler flow recordings to score signal quality and derive values for coronary flow velocity and CFR. The outputs were independently validated against expert consensus. Artificial intelligence successfully quantified Doppler signal quality, with high agreement with expert consensus (Spearman's rho: 0.94), and within individual experts. Artificial intelligence automatically tracked flow velocity with superior numerical agreement against experts, when compared with the current console algorithm [AI flow vs. expert flow bias -1.68 cm/s, 95% confidence interval (CI) -2.13 to -1.23 cm/s, P < 0.001 with limits of agreement (LOA) -4.03 to 0.68 cm/s; console flow vs. expert flow bias -2.63 cm/s, 95% CI -3.74 to -1.52, P < 0.001, 95% LOA -8.45 to -3.19 cm/s]. Artificial intelligence yielded more precise CFR values [median absolute difference (MAD) against expert CFR: 4.0% for AI and 7.4% for console]. Artificial intelligence tracked lower-quality Doppler signals with lower variability (MAD against expert CFR 8.3% for AI and 16.7% for console).
Conclusion: An AI-based system, trained by experts and independently validated, could assign a quality score to Doppler traces and derive coronary flow velocity and CFR. By making Doppler CFR more automated, precise, and operator-independent, AI could expand the clinical applicability of coronary microvascular assessment.
{"title":"Development of artificial intelligence tools for invasive Doppler-based coronary microvascular assessment.","authors":"Henry Seligman, Sapna B Patel, Anissa Alloula, James P Howard, Christopher M Cook, Yousif Ahmad, Guus A de Waard, Mauro Echavarría Pinto, Tim P van de Hoef, Haseeb Rahman, Mihir A Kelshiker, Christopher A Rajkumar, Michael Foley, Alexandra N Nowbar, Samay Mehta, Mathieu Toulemonde, Meng-Xing Tang, Rasha Al-Lamee, Sayan Sen, Graham Cole, Sukhjinder Nijjer, Javier Escaned, Niels Van Royen, Darrel P Francis, Matthew J Shun-Shin, Ricardo Petraco","doi":"10.1093/ehjdh/ztad030","DOIUrl":"10.1093/ehjdh/ztad030","url":null,"abstract":"<p><strong>Aims: </strong>Coronary flow reserve (CFR) assessment has proven clinical utility, but Doppler-based methods are sensitive to noise and operator bias, limiting their clinical applicability. The objective of the study is to expand the adoption of invasive Doppler CFR, through the development of artificial intelligence (AI) algorithms to automatically quantify coronary Doppler quality and track flow velocity.</p><p><strong>Methods and results: </strong>A neural network was trained on images extracted from coronary Doppler flow recordings to score signal quality and derive values for coronary flow velocity and CFR. The outputs were independently validated against expert consensus. Artificial intelligence successfully quantified Doppler signal quality, with high agreement with expert consensus (Spearman's rho: 0.94), and within individual experts. Artificial intelligence automatically tracked flow velocity with superior numerical agreement against experts, when compared with the current console algorithm [AI flow vs. expert flow bias -1.68 cm/s, 95% confidence interval (CI) -2.13 to -1.23 cm/s, <i>P</i> < 0.001 with limits of agreement (LOA) -4.03 to 0.68 cm/s; console flow vs. expert flow bias -2.63 cm/s, 95% CI -3.74 to -1.52, <i>P</i> < 0.001, 95% LOA -8.45 to -3.19 cm/s]. Artificial intelligence yielded more precise CFR values [median absolute difference (MAD) against expert CFR: 4.0% for AI and 7.4% for console]. Artificial intelligence tracked lower-quality Doppler signals with lower variability (MAD against expert CFR 8.3% for AI and 16.7% for console).</p><p><strong>Conclusion: </strong>An AI-based system, trained by experts and independently validated, could assign a quality score to Doppler traces and derive coronary flow velocity and CFR. By making Doppler CFR more automated, precise, and operator-independent, AI could expand the clinical applicability of coronary microvascular assessment.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/09/27/ztad030.PMC10393887.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9929220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Philip Düsing, Irina Eckardt, Stephan H Schirmer, Jan-Malte Sinning, Nikos Werner, Florian Bönner, Alexander Krogmann, Sebastian Schäfer, Alexander Sedaghat, Cornelius Müller, Georg Nickenig, Andreas Zietzer
Aims: Coronary artery disease (CAD) remains the leading cause of death worldwide. 'Stable' CAD is a chronic progressive condition, which recent European guidelines recommend referring to as 'chronic coronary syndrome' (CCS). Despite therapeutic advances, morbidity and mortality among patients with CCS remain high. Optimal secondary prevention in patients with CCS includes optimization of modifiable risk factors with behavioural changes and pharmacological therapy. The CHANGE study aims to provide evidence for optimization of secondary prevention in CCS patients by using a smartphone application (app).
Methods and results: The CHANGE study is designed as a prospective, randomized, controlled trial with a 1:1 allocation ratio, which is currently performed in nine centres in Germany in a parallel group design. 210 patients with CCS will be randomly allocated either to the control group (standard-of-care) or to the intervention group, who will be provided the VantisTherapy* app in addition to standard-of-care to incorporate secondary prevention into their daily life. The study will be performed in an open design. Outcomes will be assessed using objective data from three in-person visits (0, 12, and 24 weeks). Primary outcomes will involve adherence to secondary prevention recommendations and quality of life (QoL). The recruitment process started in July 2022.
Conclusion: The CHANGE study will investigate whether a smartphone-guided secondary prevention app, combined with a monitor function compared with standard-of-care, has beneficial effects on overall adherence to secondary prevention guidelines and QoL in patients with CCS.
Trial registration: The study is listed at the German study registry (DRKS) under the registered number DRKS00028081.
{"title":"A prospective, randomized, controlled, multicentre trial for secondary prevention in patients with chronic coronary syndrome using a smartphone application for digital therapy: the CHANGE study protocol.","authors":"Philip Düsing, Irina Eckardt, Stephan H Schirmer, Jan-Malte Sinning, Nikos Werner, Florian Bönner, Alexander Krogmann, Sebastian Schäfer, Alexander Sedaghat, Cornelius Müller, Georg Nickenig, Andreas Zietzer","doi":"10.1093/ehjdh/ztad012","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad012","url":null,"abstract":"<p><strong>Aims: </strong>Coronary artery disease (CAD) remains the leading cause of death worldwide. 'Stable' CAD is a chronic progressive condition, which recent European guidelines recommend referring to as 'chronic coronary syndrome' (CCS). Despite therapeutic advances, morbidity and mortality among patients with CCS remain high. Optimal secondary prevention in patients with CCS includes optimization of modifiable risk factors with behavioural changes and pharmacological therapy. The CHANGE study aims to provide evidence for optimization of secondary prevention in CCS patients by using a smartphone application (app).</p><p><strong>Methods and results: </strong>The CHANGE study is designed as a prospective, randomized, controlled trial with a 1:1 allocation ratio, which is currently performed in nine centres in Germany in a parallel group design. 210 patients with CCS will be randomly allocated either to the control group (standard-of-care) or to the intervention group, who will be provided the VantisTherapy* app in addition to standard-of-care to incorporate secondary prevention into their daily life. The study will be performed in an open design. Outcomes will be assessed using objective data from three in-person visits (0, 12, and 24 weeks). Primary outcomes will involve adherence to secondary prevention recommendations and quality of life (QoL). The recruitment process started in July 2022.</p><p><strong>Conclusion: </strong>The CHANGE study will investigate whether a smartphone-guided secondary prevention app, combined with a monitor function compared with standard-of-care, has beneficial effects on overall adherence to secondary prevention guidelines and QoL in patients with CCS.</p><p><strong>Trial registration: </strong>The study is listed at the German study registry (DRKS) under the registered number DRKS00028081.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/df/f5/ztad012.PMC10232292.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9568844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Peng Zhang, Fan Lin, Fei Ma, Yuting Chen, Siyi Fang, Haiyan Zheng, Zuwen Xiang, Xiaoyun Yang, Qiang Li
Aims: As the demand for atrial fibrillation (AF) screening increases, clinicians spend a significant amount of time identifying AF signals from massive amounts of data obtained during long-term dynamic electrocardiogram (ECG) monitoring. The identification of AF signals is subjective and depends on the experience of clinicians. However, experienced cardiologists are scarce. This study aimed to apply a deep learning-based algorithm to fully automate primary screening of patients with AF using 24-h Holter monitoring.
Methods and results: A deep learning model was developed to automatically detect AF episodes using RR intervals and was trained and evaluated on 23 621 (2297 AF and 21 324 non-AF) 24-h Holter recordings from 23 452 patients. Based on the AF episode detection results, patients with AF were automatically identified using the criterion of at least one AF episode lasting 6 min or longer. Performance was assessed on an independent real-world hospital-scenario test set (19 227 recordings) and a community-scenario test set (1299 recordings). For the two test sets, the model obtained high performance for the identification of patients with AF (sensitivity: 0.995 and 1.000; specificity: 0.985 and 0.997, respectively). Moreover, it obtained good and consistent performance (sensitivity: 1.000; specificity: 0.972) for an external public data set.
Conclusion: Using the criterion of at least one AF episode of 6 min or longer, the deep learning model can fully automatically screen patients for AF with high accuracy from long-term Holter monitoring data. This method may serve as a powerful and cost-effective tool for primary screening for AF.
{"title":"Automatic screening of patients with atrial fibrillation from 24-h Holter recording using deep learning.","authors":"Peng Zhang, Fan Lin, Fei Ma, Yuting Chen, Siyi Fang, Haiyan Zheng, Zuwen Xiang, Xiaoyun Yang, Qiang Li","doi":"10.1093/ehjdh/ztad018","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad018","url":null,"abstract":"<p><strong>Aims: </strong>As the demand for atrial fibrillation (AF) screening increases, clinicians spend a significant amount of time identifying AF signals from massive amounts of data obtained during long-term dynamic electrocardiogram (ECG) monitoring. The identification of AF signals is subjective and depends on the experience of clinicians. However, experienced cardiologists are scarce. This study aimed to apply a deep learning-based algorithm to fully automate primary screening of patients with AF using 24-h Holter monitoring.</p><p><strong>Methods and results: </strong>A deep learning model was developed to automatically detect AF episodes using RR intervals and was trained and evaluated on 23 621 (2297 AF and 21 324 non-AF) 24-h Holter recordings from 23 452 patients. Based on the AF episode detection results, patients with AF were automatically identified using the criterion of at least one AF episode lasting 6 min or longer. Performance was assessed on an independent real-world hospital-scenario test set (19 227 recordings) and a community-scenario test set (1299 recordings). For the two test sets, the model obtained high performance for the identification of patients with AF (sensitivity: 0.995 and 1.000; specificity: 0.985 and 0.997, respectively). Moreover, it obtained good and consistent performance (sensitivity: 1.000; specificity: 0.972) for an external public data set.</p><p><strong>Conclusion: </strong>Using the criterion of at least one AF episode of 6 min or longer, the deep learning model can fully automatically screen patients for AF with high accuracy from long-term Holter monitoring data. This method may serve as a powerful and cost-effective tool for primary screening for AF.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/42/c9/ztad018.PMC10232289.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9568845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chat Generative Pre-trained Transformer (ChatGPT) is currently a trending topic worldwide triggering extensive debate about its predictive power, its potential uses, and its wider implications. Recent publications have demonstrated that ChatGPT can correctly answer questions from undergraduate exams such as the United States Medical Licensing Examination. We challenged it to answer questions from a more demanding, post-graduate exam-the European Exam in Core Cardiology (EECC), the final exam for the completion of specialty training in Cardiology in many countries. Our results demonstrate that ChatGPT succeeds in the EECC.
{"title":"ChatGPT takes on the European Exam in Core Cardiology: an artificial intelligence success story?","authors":"Ioannis Skalidis, Aurelien Cagnina, Wongsakorn Luangphiphat, Thabo Mahendiran, Olivier Muller, Emmanuel Abbe, Stephane Fournier","doi":"10.1093/ehjdh/ztad029","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad029","url":null,"abstract":"<p><p>Chat Generative Pre-trained Transformer (ChatGPT) is currently a trending topic worldwide triggering extensive debate about its predictive power, its potential uses, and its wider implications. Recent publications have demonstrated that ChatGPT can correctly answer questions from undergraduate exams such as the United States Medical Licensing Examination. We challenged it to answer questions from a more demanding, post-graduate exam-the European Exam in Core Cardiology (EECC), the final exam for the completion of specialty training in Cardiology in many countries. Our results demonstrate that ChatGPT succeeds in the EECC.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/48/d5/ztad029.PMC10232281.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9933179","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shantanu Sengupta, Siddharth Biswal, Jitto Titus, Atandra Burman, Keshav Reddy, Mahesh C Fulwani, Aziz Khan, Niteen Deshpande, Smit Shrivastava, Naveena Yanamala, Partho P Sengupta
Aims: Clinical differentiation of acute myocardial infarction (MI) from unstable angina and other presentations mimicking acute coronary syndromes (ACS) is critical for implementing time-sensitive interventions and optimizing outcomes. However, the diagnostic steps are dependent on blood draws and laboratory turnaround times. We tested the clinical feasibility of a wrist-worn transdermal infrared spectrophotometric sensor (transdermal-ISS) in clinical practice and assessed the performance of a machine learning algorithm for identifying elevated high-sensitivity cardiac troponin-I (hs-cTnI) levels in patients hospitalized with ACS.
Methods and results: We enrolled 238 patients hospitalized with ACS at five sites. The final diagnosis of MI (with or without ST elevation) and unstable angina was adjudicated using electrocardiography (ECG), cardiac troponin (cTn) test, echocardiography (regional wall motion abnormality), or coronary angiography. A transdermal-ISS-derived deep learning model was trained (three sites) and externally validated with hs-cTnI (one site) and echocardiography and angiography (two sites), respectively. The transdermal-ISS model predicted elevated hs-cTnI levels with areas under the receiver operator characteristics of 0.90 [95% confidence interval (CI), 0.84-0.94; sensitivity, 0.86; and specificity, 0.82] and 0.92 (95% CI, 0.80-0.98; sensitivity, 0.94; and specificity, 0.64), for internal and external validation cohorts, respectively. In addition, the model predictions were associated with regional wall motion abnormalities [odds ratio (OR), 3.37; CI, 1.02-11.15; P = 0.046] and significant coronary stenosis (OR, 4.69; CI, 1.27-17.26; P = 0.019).
Conclusion: A wrist-worn transdermal-ISS is clinically feasible for rapid, bloodless prediction of elevated hs-cTnI levels in real-world settings. It may have a role in establishing a point-of-care biomarker diagnosis of MI and impact triaging patients with suspected ACS.
{"title":"A novel breakthrough in wrist-worn transdermal troponin-I-sensor assessment for acute myocardial infarction.","authors":"Shantanu Sengupta, Siddharth Biswal, Jitto Titus, Atandra Burman, Keshav Reddy, Mahesh C Fulwani, Aziz Khan, Niteen Deshpande, Smit Shrivastava, Naveena Yanamala, Partho P Sengupta","doi":"10.1093/ehjdh/ztad015","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad015","url":null,"abstract":"<p><strong>Aims: </strong>Clinical differentiation of acute myocardial infarction (MI) from unstable angina and other presentations mimicking acute coronary syndromes (ACS) is critical for implementing time-sensitive interventions and optimizing outcomes. However, the diagnostic steps are dependent on blood draws and laboratory turnaround times. We tested the clinical feasibility of a wrist-worn transdermal infrared spectrophotometric sensor (transdermal-ISS) in clinical practice and assessed the performance of a machine learning algorithm for identifying elevated high-sensitivity cardiac troponin-I (hs-cTnI) levels in patients hospitalized with ACS.</p><p><strong>Methods and results: </strong>We enrolled 238 patients hospitalized with ACS at five sites. The final diagnosis of MI (with or without ST elevation) and unstable angina was adjudicated using electrocardiography (ECG), cardiac troponin (cTn) test, echocardiography (regional wall motion abnormality), or coronary angiography. A transdermal-ISS-derived deep learning model was trained (three sites) and externally validated with hs-cTnI (one site) and echocardiography and angiography (two sites), respectively. The transdermal-ISS model predicted elevated hs-cTnI levels with areas under the receiver operator characteristics of 0.90 [95% confidence interval (CI), 0.84-0.94; sensitivity, 0.86; and specificity, 0.82] and 0.92 (95% CI, 0.80-0.98; sensitivity, 0.94; and specificity, 0.64), for internal and external validation cohorts, respectively. In addition, the model predictions were associated with regional wall motion abnormalities [odds ratio (OR), 3.37; CI, 1.02-11.15; <i>P</i> = 0.046] and significant coronary stenosis (OR, 4.69; CI, 1.27-17.26; <i>P</i> = 0.019).</p><p><strong>Conclusion: </strong>A wrist-worn transdermal-ISS is clinically feasible for rapid, bloodless prediction of elevated hs-cTnI levels in real-world settings. It may have a role in establishing a point-of-care biomarker diagnosis of MI and impact triaging patients with suspected ACS.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/fb/b8/ztad015.PMC10232240.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9566519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Andreas Leha, Cynthia Huber, Tim Friede, Timm Bauer, Andreas Beckmann, Raffi Bekeredjian, Sabine Bleiziffer, Eva Herrmann, Helge Möllmann, Thomas Walther, Friedhelm Beyersdorf, Christian Hamm, Arnaud Künzi, Stephan Windecker, Stefan Stortecky, Ingo Kutschka, Gerd Hasenfuß, Stephan Ensminger, Christian Frerker, Tim Seidler
Aims: Identification of high-risk patients and individualized decision support based on objective criteria for rapid discharge after transcatheter aortic valve implantation (TAVI) are key requirements in the context of contemporary TAVI treatment. This study aimed to predict 30-day mortality following TAVI based on machine learning (ML) using data from the German Aortic Valve Registry.
Methods and results: Mortality risk was determined using a random forest ML model that was condensed in the newly developed TAVI Risk Machine (TRIM) scores, designed to represent clinically meaningful risk modelling before (TRIMpre) and in particular after (TRIMpost) TAVI. Algorithm was trained and cross-validated on data of 22 283 patients (729 died within 30 days post-TAVI) and generalisation was examined on data of 5864 patients (146 died). TRIMpost demonstrated significantly better performance than traditional scores [C-statistics value, 0.79; 95% confidence interval (CI)] [0.74; 0.83] compared to Society of Thoracic Surgeons (STS) with C-statistics value 0.69; 95%-CI [0.65; 0.74]). An abridged (aTRIMpost) score comprising 25 features (calculated using a web interface) exhibited significantly higher performance than traditional scores (C-statistics value, 0.74; 95%-CI [0.70; 0.78]). Validation on external data of 6693 patients (205 died within 30 days post-TAVI) of the Swiss TAVI Registry confirmed significantly better performance for the TRIMpost (C-statistics value 0.75, 95%-CI [0.72; 0.79]) compared to STS (C-statistics value 0.67, CI [0.63; 0.70]).
Conclusion: TRIM scores demonstrate good performance for risk estimation before and after TAVI. Together with clinical judgement, they may support standardised and objective decision-making before and after TAVI.
目的:在当代主动脉瓣置入术(TAVI)治疗的背景下,根据客观标准识别高危患者和个性化决策支持是TAVI治疗的关键要求。本研究旨在利用德国主动脉瓣登记处的数据,基于机器学习(ML)预测TAVI后30天的死亡率。方法和结果:使用随机森林ML模型确定死亡风险,该模型浓缩在新开发的TAVI风险机器(TRIM)评分中,旨在表示在(TRIMpre) TAVI之前,特别是(TRIMpost) TAVI之后有临床意义的风险模型。对22 283例患者(729例tavi后30天内死亡)的数据进行训练和交叉验证,并对5864例患者(146例死亡)的数据进行泛化检验。TRIMpost的表现明显优于传统评分[c统计值,0.79;95%置信区间[0.74;0.83]而胸外科学会(STS)的c统计值为0.69;95% ci 0.65;0.74])。包含25个特征(使用web界面计算)的精简(aTRIMpost)分数表现出比传统分数显著更高的性能(c统计值,0.74;95% ci 0.70;0.78])。瑞士TAVI注册中心6693例患者(其中205例在TAVI后30天内死亡)的外部数据验证证实TRIMpost的疗效显著更好(c -统计值0.75,95% ci [0.72;0.79])与STS相比(c统计值0.67,CI [0.63;0.70])。结论:TRIM评分对TAVI前后的风险评估有较好的效果。与临床判断相结合,可为TAVI前后的规范化、客观决策提供支持。
{"title":"Development and validation of explainable machine learning models for risk of mortality in transcatheter aortic valve implantation: TAVI risk machine scores.","authors":"Andreas Leha, Cynthia Huber, Tim Friede, Timm Bauer, Andreas Beckmann, Raffi Bekeredjian, Sabine Bleiziffer, Eva Herrmann, Helge Möllmann, Thomas Walther, Friedhelm Beyersdorf, Christian Hamm, Arnaud Künzi, Stephan Windecker, Stefan Stortecky, Ingo Kutschka, Gerd Hasenfuß, Stephan Ensminger, Christian Frerker, Tim Seidler","doi":"10.1093/ehjdh/ztad021","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad021","url":null,"abstract":"<p><strong>Aims: </strong>Identification of high-risk patients and individualized decision support based on objective criteria for rapid discharge after transcatheter aortic valve implantation (TAVI) are key requirements in the context of contemporary TAVI treatment. This study aimed to predict 30-day mortality following TAVI based on machine learning (ML) using data from the German Aortic Valve Registry.</p><p><strong>Methods and results: </strong>Mortality risk was determined using a random forest ML model that was condensed in the newly developed TAVI Risk Machine (TRIM) scores, designed to represent clinically meaningful risk modelling before (TRIMpre) and in particular after (TRIMpost) TAVI. Algorithm was trained and cross-validated on data of 22 283 patients (729 died within 30 days post-TAVI) and generalisation was examined on data of 5864 patients (146 died). TRIMpost demonstrated significantly better performance than traditional scores [<i>C</i>-statistics value, 0.79; 95% confidence interval (CI)] [0.74; 0.83] compared to Society of Thoracic Surgeons (STS) with <i>C</i>-statistics value 0.69; 95%-CI [0.65; 0.74]). An abridged (aTRIMpost) score comprising 25 features (calculated using a web interface) exhibited significantly higher performance than traditional scores (<i>C</i>-statistics value, 0.74; 95%-CI [0.70; 0.78]). Validation on external data of 6693 patients (205 died within 30 days post-TAVI) of the Swiss TAVI Registry confirmed significantly better performance for the TRIMpost (<i>C</i>-statistics value 0.75, 95%-CI [0.72; 0.79]) compared to STS (<i>C</i>-statistics value 0.67, CI [0.63; 0.70]).</p><p><strong>Conclusion: </strong>TRIM scores demonstrate good performance for risk estimation before and after TAVI. Together with clinical judgement, they may support standardised and objective decision-making before and after TAVI.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/61/90/ztad021.PMC10232286.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9568848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Aims: The black box nature of artificial intelligence (AI) hinders the development of interpretable AI models that are applicable in clinical practice. We aimed to develop an AI model for classifying patients of reduced left ventricular ejection fraction (LVEF) from 12-lead electrocardiograms (ECG) with the decision-interpretability.
Methods and results: We acquired paired ECG and echocardiography datasets from the central and co-operative institutions. For the central institution dataset, a random forest model was trained to identify patients with reduced LVEF among 29 907 ECGs. Shapley additive explanations were applied to 7196 ECGs. To extract the model's decision criteria, the calculated Shapley additive explanations values were clustered for 192 non-paced rhythm patients in which reduced LVEF was predicted. Although the extracted criteria were different for each cluster, these criteria generally comprised a combination of six ECG findings: negative T-wave inversion in I/V5-6 leads, low voltage in I/II/V4-6 leads, Q wave in V3-6 leads, ventricular activation time prolongation in I/V5-6 leads, S-wave prolongation in V2-3 leads, and corrected QT interval prolongation. Similarly, for the co-operative institution dataset, the extracted criteria comprised a combination of the same six ECG findings. Furthermore, the accuracy of seven cardiologists' ECG readings improved significantly after watching a video explaining the interpretation of these criteria (before, 62.9% ± 3.9% vs. after, 73.9% ± 2.4%; P = 0.02).
Conclusion: We visually interpreted the model's decision criteria to evaluate its validity, thereby developing a model that provided the decision-interpretability required for clinical application.
{"title":"An explainable artificial intelligence-enabled electrocardiogram analysis model for the classification of reduced left ventricular function.","authors":"Susumu Katsushika, Satoshi Kodera, Shinnosuke Sawano, Hiroki Shinohara, Naoto Setoguchi, Kengo Tanabe, Yasutomi Higashikuni, Norifumi Takeda, Katsuhito Fujiu, Masao Daimon, Hiroshi Akazawa, Hiroyuki Morita, Issei Komuro","doi":"10.1093/ehjdh/ztad027","DOIUrl":"https://doi.org/10.1093/ehjdh/ztad027","url":null,"abstract":"<p><strong>Aims: </strong>The black box nature of artificial intelligence (AI) hinders the development of interpretable AI models that are applicable in clinical practice. We aimed to develop an AI model for classifying patients of reduced left ventricular ejection fraction (LVEF) from 12-lead electrocardiograms (ECG) with the decision-interpretability.</p><p><strong>Methods and results: </strong>We acquired paired ECG and echocardiography datasets from the central and co-operative institutions. For the central institution dataset, a random forest model was trained to identify patients with reduced LVEF among 29 907 ECGs. Shapley additive explanations were applied to 7196 ECGs. To extract the model's decision criteria, the calculated Shapley additive explanations values were clustered for 192 non-paced rhythm patients in which reduced LVEF was predicted. Although the extracted criteria were different for each cluster, these criteria generally comprised a combination of six ECG findings: negative T-wave inversion in I/V5-6 leads, low voltage in I/II/V4-6 leads, Q wave in V3-6 leads, ventricular activation time prolongation in I/V5-6 leads, S-wave prolongation in V2-3 leads, and corrected QT interval prolongation. Similarly, for the co-operative institution dataset, the extracted criteria comprised a combination of the same six ECG findings. Furthermore, the accuracy of seven cardiologists' ECG readings improved significantly after watching a video explaining the interpretation of these criteria (before, 62.9% ± 3.9% vs. after, 73.9% ± 2.4%; <i>P</i> = 0.02).</p><p><strong>Conclusion: </strong>We visually interpreted the model's decision criteria to evaluate its validity, thereby developing a model that provided the decision-interpretability required for clinical application.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10232279/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9568843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}