European heart journal. Digital health最新文献

{"title":"Deep learning for atrioventricular regurgitation diagnosis: an external validation study.","authors":"Ido Cohen, Jeffrey G Malins, Michal Cohen-Shelly, Yossi Asaf, Michael Fiman, Kobi Faierstein, Lior Fisher, Karin Sudri, Ehud Raanani, Ehud Schwammenthal, Robert Klempfner, Elad Maor","doi":"10.1093/ehjdh/ztaf078","DOIUrl":"10.1093/ehjdh/ztaf078","url":null,"abstract":"<p><strong>Aims: </strong>Mitral and tricuspid regurgitation (MR and TR) are common in older adults and associated with substantial morbidity and mortality. While transthoracic echocardiography (TTE) is the diagnostic gold standard, access remains limited in many care settings. Artificial intelligence (AI)-based echocardiographic analysis may help address this diagnostic gap.</p><p><strong>Methods and results: </strong>We externally validated a deep learning algorithm developed by Aisap.ai using TTE studies from the Mayo Clinic Health System (2013-23). The model analyses echocardiographic images to classify atrioventricular regurgitation severity and was evaluated against cardiologist interpretations. Performance was assessed using binary (normal-mild vs. moderate-severe) and ordinal (normal, mild, moderate, severe) classification schemes. Among 1541 eligible TTEs, the model returned predictions for 578 studies (38%). Performance analysis was limited to these cases. The MR cohort included 280 studies and the TR cohort 298. For MR, the model achieved an area under the receiver operating characteristic curve (AUC) of 0.98 [95% confidence interval (CI): 0.97-0.99], with 91% accuracy, 95% sensitivity, and 89% specificity. For TR, the AUC was 0.96 (95% CI: 0.94-0.98), with 84% accuracy, 91% sensitivity, and 80% specificity.</p><p><strong>Conclusion: </strong>In cases where a prediction was generated, the model demonstrated high diagnostic performance in identifying clinically significant atrioventricular regurgitation. These findings support the feasibility of AI-assisted echocardiography in diverse populations, while underscoring the need for technical alignment between model requirements and local acquisition practices to ensure real-world applicability.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"949-958"},"PeriodicalIF":4.4,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450501/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126757","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}

{"title":"Identifying congestion phenotypes using unsupervised machine learning in acute heart failure.","authors":"Tripti Rastogi, Olivier Hutin, Jozine M Ter Maaten, Guillaume Baudry, Luca Monzo, Emmanuel Bresso, Kevin Duarte, Jasper Tromp, Adriaan A Voors, Nicolas Girerd","doi":"10.1093/ehjdh/ztaf065","DOIUrl":"10.1093/ehjdh/ztaf065","url":null,"abstract":"<p><strong>Aims: </strong>Data-driven clustering techniques may improve heart failure (HF) categorisation and provide prognostic insights. The present study aimed to elucidate the underlying pathophysiology of acute HF phenotypes based on pulmonary and systemic congestion at both the tissue (PTC, pulmonary tissue congestion; STC, systemic tissue congestion) and intravascular (PIVC, pulmonary intravascular congestion; SIVC, systemic intravascular congestion) level and to assess the association of identified phenotypes with a composite outcome of HF hospitalisation and death.</p><p><strong>Methods and results: </strong>Nineteen clinical, laboratory, and echocardiographic congestion markers were analyzed using clustering techniques to identify phenotypes in patients with worsening HF in the Nancy-HF cohort (<i>n</i> = 741), followed by validation of the clustering model in the BIOSTAT-CHF cohort (<i>n</i> = 4254). Network analysis was conducted using 363 proteins to identify underlying biological pathways. Five congestion phenotypes were identified: (1) PTC-dilated left ventricle (LV), (2) PTC-HFpEF, (3) PTC, STC-atrial fibrillation (AF), (4) PIVC-dilated left atrium (LA) and LV and (5) Global congestion. Compared with the 'PTC-dilated LV' phenotype, the risk of composite outcome was higher in 'PTC, STC-AF' and 'Global' congestion phenotypes [adjusted HR: 1.74 (1.13-2.67) and 2.41 (1.60-3.63), respectively]. In BIOSTAT-CHF, 'Global' congestion phenotype was associated with significantly higher risk [HR: 1.64 (1.04-2.58)]. In network analysis, the immune response pathway was linked to all phenotypes. 'PTC-HFpEF' was related to lipid, protein and angiotensin metabolism, 'PTC, STC-AF' was related to kinase-mediated signalling, extracellular matrix organisation and TNF-regulated cell death, while 'PIVC-dilated LA & LV' was related to kinase-mediated signalling and hemostasis.</p><p><strong>Conclusion: </strong>In worsening HF, clustering techniques identified clinical congestion profiles associated with both long-term clinical risk and differences in biomarkers, suggesting potential different underlying pathophysiologies. These clusters can be applied using the available online model to identify phenotypes as well as associated risks (https://cic-p-nancy.fr/ai-cong-hf/).</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"907-918"},"PeriodicalIF":4.4,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450512/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126709","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}

{"title":"Using telecommunication dialogue and nursing documentation to predict the risk of emergency room visit in a web-based telehealth programme.","authors":"Hui-Wen Wu, Chi-Sheng Hung, Ying-Hsien Chen, Ching-Chang Huang, Jen-Kuang Lee, Shin-Tsyr Hwang, Yi-Lwun Ho","doi":"10.1093/ehjdh/ztaf076","DOIUrl":"10.1093/ehjdh/ztaf076","url":null,"abstract":"<p><strong>Aims: </strong>The effectiveness of telehealth care programmes in reducing mortality among patients with chronic conditions has been well established. Valuable insights into patients' conditions can be gleaned through daily telecommunication between patients and nurse case managers. We hypothesized that using natural language processing can predict acute deterioration in patients with chronic conditions in telehealth care programme based on the nursing records and speech dialogues occurring during daily telecommunication.</p><p><strong>Methods and results: </strong>We conducted a retrospective study utilizing audio recording transcripts from telecommunication sessions between patients and nurse case managers at our telehealth care centre, along with nursing notes as input data. Pre-trained transformer-based neural network models were constructed to predict emergency room (ER) visits within a 2-week timeframe. The case group included 94 patients with 585 speech recordings and nursing records, while the control group included 36 patients with 396 speech recordings and nursing records. Our results showed that employing transcripts and a bidirectional encoder representations from transformers (BERT)-base model with a sliding window for predicting ER visits yielded moderate accuracy 0.75 (interquartile range: 0.742, 0.773). The inclusion of long short-term memory in the model did not significantly enhance accuracy. Notably, combining nursing records and transcripts as inputs exhibited superior performance, achieving an overall accuracy of 0.892 (interquartile range: 0.891, 0.893) by the six models.</p><p><strong>Conclusion: </strong>Our study demonstrates the feasibility of predicting ER visits using telehealth dialogue transcripts and nursing notes with pre-trained transformer models. The incorporation of nursing notes significantly enhances the model's performance, providing a valuable method for improving predictive accuracy in telehealth care.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"1036-1045"},"PeriodicalIF":4.4,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450502/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126249","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}

{"title":"External assessment of an artificial intelligence-enabled electrocardiogram for aortic stenosis detection.","authors":"Darae Kim, Eunjung Lee, Jihoon Kim, Eun Kyoung Kim, Sung-A Chang, Sung-Ji Park, Jin-Oh Choi, Young Keun On, Zachi Attia, Paul Friedman, Kyoung-Min Park, Jae K Oh","doi":"10.1093/ehjdh/ztaf067","DOIUrl":"10.1093/ehjdh/ztaf067","url":null,"abstract":"<p><strong>Aims: </strong>To assess the performance of an artificial intelligence-enabled electrocardiogram (AI-ECG) algorithm in identifying patients with moderate to severe aortic stenosis (AS) in an Asian cohort from a tertiary care centre.</p><p><strong>Methods and results: </strong>We identified a randomly selected patients ≥60 years old who underwent echocardiography and ECG within in 31 days between 2012 and 2021 at the Samsung Medical Center in Korea. Patients with previous cardiac surgery, prosthetic valves, or pacemakers were excluded. The AI-ECG model, originally developed and validated by Mayo Clinic in the USA, was applied without fine-tuning. Performance metrics, including the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy, were calculated to compare AI-ECG predictions with TTE-confirmed AS status. Among 5425 patients, 1095 had moderate to severe AS, and 4330 age- and sex-matched patients without AS were included as controls. The AI-ECG model achieved an AUC of 0.85 (95% CI: 0.84-0.87) in detecting moderate to severe AS. Sensitivity, specificity, PPV, NPV, and accuracy were 0.83, 0.65, 0.37, 0.94, and 68.29%, respectively. The model's performance was consistent across various age and sex subgroups, with sensitivity increasing in older patients.</p><p><strong>Conclusion: </strong>The AI-ECG model developed in the USA demonstrated comparable performance in detecting moderate to severe AS in an Asian cohort compared with its original validation population. These findings highlight the potential utility of AI-ECG as a non-invasive screening tool for AS across diverse patient populations.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 4","pages":"656-664"},"PeriodicalIF":3.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12282354/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144700490","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}

{"title":"Artificial intelligence-based accurate myocardial infarction mapping using 12-lead electrocardiography.","authors":"Hui Wang, Zhifan Gao, Heye Zhang, Yuzhen Zhu, Shichang Lian, Kairui Bo, Shuang Li, Yifeng Gao, Baiyan Zhuang, Zhen Zhou, Xinwei Zhang, Cuiyan Wang, Koen Nieman, Lei Xu","doi":"10.1093/ehjdh/ztaf077","DOIUrl":"10.1093/ehjdh/ztaf077","url":null,"abstract":"<p><strong>Aims: </strong>Assessing myocardial fibrosis (MF) in patients with prior myocardial infarction (MI) is crucial for prognosis. Artificial intelligence-assisted electrocardiography (AI-ECG) has a great potential to detect MF. However, training a precise AI-ECG model requires voluminous ECGs. A biosimulation model may be an efficient substitution. This study aimed to develop and validate a novel artificial intelligence-assisted method using 12-lead electrocardiography (AI-MI-12ECG).</p><p><strong>Methods and results: </strong>The AI-MI-12ECG was trained by a biosimulation model to visualize the presence, location, and size of MF in post-MI patients. A total of 182 post-MI patients were included in this prospective study. The MF detected by AI-MI-12ECG and the cardiologist were compared with the late gadolinium-enhanced (LGE) area of cardiac magnetic resonance (CMR). The results show that AI-MI-12ECG exhibited strong correlation with LGE in identifying the MI location (<i>R</i> = 0.955). Compared with CMR-LGE, AI-MI-12ECG achieved receiver operating characteristic curves of 0.95, 0.95, and 0.89 for left anterior descending coronary artery (LAD), right coronary artery (RCA), and left circumflex coronary artery (LCX) territories, respectively, with high accuracies for LAD (0.95), RCA (0.97), and LCX (0.91).</p><p><strong>Conclusion: </strong>The AI-MI-12ECG trained using the biosimulation model in post-MI patients was adequately aligned with CMR-LGE. This highlights its potential for accurate detection of fibrosis and identification of individuals with significant infarct burdens.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"939-948"},"PeriodicalIF":4.4,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450511/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126580","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}

{"title":"The 'Advancing Cardiovascular Risk Identification with Structured Clinical Documentation and Biosignal Derived Phenotypes Synthesis' project: conceptual design, project planning, and first implementation experiences.","authors":"Dominik Felbel, Merten Prüser, Constanze Schmidt, Björn Schreiweis, Nicolai Spicher, Wolfgang Rottbauer, Julian Varghese, Andreas Zietzer, Stefan Störk, Christoph Dieterich, Dagmar Krefting, Eimo Martens, Martin Sedlmayr, Dario Bongiovanni, Christoph B Olivier, Hendrik Lapp, Hannes H J G Schmidt, Julius L Katzmann, Felix Nensa, Norbert Frey, Gudrun S Ulrich-Merzenich, Carina A Peter, Peter Heuschmann, Udo Bavendiek, Sven Zenker","doi":"10.1093/ehjdh/ztaf075","DOIUrl":"10.1093/ehjdh/ztaf075","url":null,"abstract":"<p><strong>Aims: </strong>Personalized risk assessment tools (PRTs) are recommended by cardiovascular guidelines to tailor prevention, diagnosis, and treatment. However, PRT implementation in clinical routine is poor. ACRIBiS (Advancing Cardiovascular Risk Identification with Structured Clinical Documentation and Biosignal Derived Phenotypes Synthesis) aims to establish interoperable infrastructures for standardized documentation of routine data and integration of high-resolution biosignals (HRBs) enabling data-based risk assessment.</p><p><strong>Methods and results: </strong>Established cardiovascular risk scores were selected by their predictive performance and served as basis for building a core cardiovascular dataset with risk-relevant clinical routine information. Data items not yet represented in the Medical Informatics Inititative (MII) Core Dataset (CDS) FHIR profiles will be added to an extension module 'Cardiology' allowing for maximum interoperability. HRB integration will be implemented at each site through a modular infrastructure for electrocardiography (ECG) processing. Predictive performance of PRTs and their dynamic recalibration through HRB integration will be evaluated within the ACRIBiS cohort consisting of 5250 prospectively recruited patients at 15 German academic cardiology departments with 12-month follow-up. The potential of visualising these risks to improve patient education will also be assessed and supported by the development of a self-assessment app.</p><p><strong>Discussion: </strong>The ACRIBiS project presents an innovative concept to harmonize clinical data documentation and integrate ECG data, ultimately facilitating personalized risk assessment to improve patient empowerment and prognosis. Importantly, the consensus-based documentation and interoperability specifications developed will support the standardisation of routine patient data collection at the national and international levels, while the ACRIBiS cohort dataset will be available for broad secondary use.</p><p><strong>Trial registration: </strong>The study is registered at the German study registry (DRKS): #DRKS00034792.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"1084-1093"},"PeriodicalIF":4.4,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450505/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126712","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}

{"title":"The hope and the hype of artificial intelligence for syncope management.","authors":"Samuel L Johnston, E John Barsotti, Constantinos Bakogiannis, Artur Fedorowski, Fabrizio Ricci, Eric G Heller, Robert S Sheldon, Richard Sutton, Win-Kuang Shen, Venkatesh Thiruganasambandamoorthy, Mehul Adhaduk, William H Parker, Arwa Aburizik, Corey R Haselton, Alex J Cuskey, Sangil Lee, Madeleine Johansson, Donald Macfarlane, Paari Dominic, Haruhiko Abe, B Hygriv Rao, Avinash Mudireddy, Milan Sonka, Roopinder K Sandhu, Rose Anne Kenny, Giselle M Statz, Rakesh Gopinathannair, David Benditt, Franca Dipaola, Mauro Gatti, Roberto Menè, Alessandro Giaj Levra, Dana Shiffer, Giorgio Costantino, Raffaello Furlan, Martin H Ruwald, Vassilios Vassilikos, Milena A Gebska, Brian Olshansky","doi":"10.1093/ehjdh/ztaf061","DOIUrl":"10.1093/ehjdh/ztaf061","url":null,"abstract":"<p><strong>Aims: </strong>Syncope remains a diagnostic challenge despite advancements in testing and treatment. Cardiac syncope is an independent predictor of mortality and can be difficult to distinguish from other causes of transient loss of consciousness (TLOC). This paper explores whether artificial intelligence (AI) can improve the evaluation and management of patients with syncope.</p><p><strong>Methods and results: </strong>We conducted a literature review and incorporated the opinions of experts in the fields of syncope and AI. The cause of TLOC is often unclear, hospitalization criteria are ambiguous, diagnostic tests are frequently non-informative, and assessments are costly. Patients are left with unanswered questions and limited guidance. Artificial intelligence (AI) has the potential to optimize syncope evaluation by processing large data sets, detecting imperceptible patterns, and assisting clinicians. However, AI has limitations, including errors, lack of human empathy, and uncertain clinical utility. Liability issues further complicate its integration. We present three viewpoints: (i) AI is crucial for advancing syncope management; (ii) AI can enhance the patient experience; and (iii) AI in syncope care is inevitable.</p><p><strong>Conclusion: </strong>Artificial intelligence may improve syncope diagnosis and management, particularly through machine learning-based test interpretation and wearable device data. However, it has yet to surpass human clinical judgment in complex decision-making. Current challenges include gaps in understanding syncope mechanisms, AI interpretability, generalizability, and clinical integration. Standardized diagnostic approaches, real-world validation, and curated data sets are essential for progress. Artificial intelligence may enhance efficiency and communication but raises concerns regarding confidentiality, bias, inequities, and legal implications.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"1046-1054"},"PeriodicalIF":4.4,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450521/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126720","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}

{"title":"Catheterization laboratories open the doors for Extended Realities-review of clinical applications in cardiology.","authors":"Maria Kundzierewicz, Katarzyna Kołodziej, Arif Khokhar, Tsai Tsung-Ying, Artur Leśniak, Pawel Zakrzewski, Hubert Borecki, Ewelina Bohn, Jan Hecko, Jaroslav Januska, Daniel Precek, Maciej Stanuch, Andrzej Skalski, Yoshinobu Onuma, Patrick Serruys, Nico Bruining, Adriana Złahoda-Huzior, Dariusz Dudek","doi":"10.1093/ehjdh/ztaf072","DOIUrl":"10.1093/ehjdh/ztaf072","url":null,"abstract":"<p><p>The complexity and spatial relationships between vascular and cardiac structures, as well as anatomical diversity, pose a challenge for planning and performing cardiac interventions. Medical imaging, especially precise three-dimensional imaging techniques, plays a key role in the decision-making process. While traditional imaging methods like angiography, echocardiography, computed tomography, and magnetic resonance imaging remain gold standards, they have limitations in representing spatial relationships effectively. To overcome these limitations, advanced techniques such as three-dimensional printing, three-dimensional modelling, and Extended Realities are needed. Focusing on Extended Realities, their main advantages are direct spatial visualization based on medical data, interaction with objects, and immersion in cardiac anatomy. These benefits impact procedural planning and intra-procedural navigation. The following publication presents current applications, benefits, drawbacks, and limitations of Virtual, Augmented, and Mixed Reality technologies in cardiac interventions. The aim of this review is to improve understanding and utilization of the entire spectrum of these innovative tools in clinical practice.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"1055-1068"},"PeriodicalIF":4.4,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450514/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126701","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}

{"title":"Machine learning electrocardiography model to differentiate takotsubo syndrome from myocardial infarction.","authors":"Felicia H K Hakansson, Erik Bodin, Vincent Dutordoir, Axel Gemvik, Thomas Olsson, Isabelle Nilsson, Mikael Andersson Franko, Jonas Spaak, Christina Ekenbäck, Loghman Henareh, Carl Henrik Ek, Per Tornvall","doi":"10.1093/ehjdh/ztaf073","DOIUrl":"10.1093/ehjdh/ztaf073","url":null,"abstract":"<p><strong>Aims: </strong>Machine learning (ML) algorithms applied to the electrocardiography (ECG) have been successful in several cardiac diagnoses, however, rarely been used for the diagnostics of takotsubo syndrome (TTS). Our aim was to develop ML-based ECG-models to differentiate TTS from patients with myocardial infarction (MI).</p><p><strong>Methods and results: </strong>Cross-sectional study in Stockholm. A neural network with UNet architecture was trained and validated on 507 TTS cases and 14 978 controls with suspected and verified MI, identified from the Swedish coronary angiography and angioplasty register. Cross-validation was performed. The models were compared with cardiologists using previously proposed ECG criteria. Receiver operating characteristics (ROC) area under the curve (AUC) for discriminating TTS from patients with ST-elevation and non-ST-elevation MI ROC AUC 0.88 (cross-validation: 0.85-0.92) and 0.86 (cross-validation: 0.82-0.91), respectively. ROC AUC for discriminating TTS from verified MI [non-ST-elevation MI (NSTEMI) and ST-elevation MI (STEMI)] was 0.87 (cross-validation: 0.83-0.91) with sensitivity (0.75) and specificity (0.83) with low positive predictive value (PPV) and high negative predictive value (NPV). Results for suspected MI was ROC AUC 0.85 (cross validation: 0.81-0.91) with sensitivity (0.75) and specificity (0.79) with low PPV (0.11) and high NPV (0.99). The committee of two cardiologists using a combination of ECG criteria achieved an ROC AUC of 0.71.</p><p><strong>Conclusion: </strong>Machine learning models could discriminate TTS from MI (NSTEMI and STEMI) and suspected MI with high sensitivity and NPV, outperforming cardiologists using conventional criteria. The models require further refinement to increase PPV, precision-recall and external validation, but it holds promise for TTS screening aiding the clinician in ruling out TTS.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"929-938"},"PeriodicalIF":4.4,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450510/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126738","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}

{"title":"Identification of clinical phenotypes and heterogeneous treatment effects of surgical revascularization in ischaemic cardiomyopathy: a machine learning consensus clustering analysis.","authors":"Tongxin Chu, Zhuoming Zhou, Huayang Li, Han Hu, Pengning Fan, Suiqing Huang, Jiatang Xu, Qiushi Ren, Qingyang Song, Gang Li, Mengya Liang, Zhongkai Wu","doi":"10.1093/ehjdh/ztaf066","DOIUrl":"10.1093/ehjdh/ztaf066","url":null,"abstract":"<p><strong>Aims: </strong>To identify ischaemic cardiomyopathy (ICM) patients with different phenotypes for evaluating their outcomes and heterogeneous treatment effects (HTEs) of coronary artery bypass grafting (CABG).</p><p><strong>Methods and results: </strong>We applied a machine learning-based consensus, K-Medoids clustering analysis to the Surgical Treatment for Ischemic Heart Failure trial. We compared the risk of all-cause mortality and cardiovascular mortality among different phenotypes. The survival benefits of CABG compared with medical therapy alone were assessed in the identified phenotypes for evaluating HTEs. The consensus clustering analysis identified three distinct clinical phenotypes among 1212 ICM patients based on 19 variables. Specifically, phenotype 1 (<i>n</i> = 371) was characterized by younger ages, higher left ventricular ejection fraction (LVEF), and lower left ventricular end-systolic volume index (<i>n</i> = 371). Phenotype 2 had higher angina grades and more left main/left anterior descending artery stenosis (<i>n</i> = 520). Phenotype 3 had lower LVEF, higher New York Heart Association (NYHA) grades, more diabetes, and less hypertension (<i>n</i> = 321). After a median of 9.8 follow-up years, phenotype 3 had the highest risk of all-cause mortality [hazard ratio (HR), 1.96; 95% confidence intervals (CI), 1.62-2.37] and cardiovascular mortality (HR, 2.46; 95% CI, 1.95-3.10) compared to phenotype 1. Among phenotype 3, CABG provided significant survival benefits in all-cause mortality (HR, 0.75; 95% CI, 0.58-0.96) and cardiovascular mortality (HR, 0.67; 95% CI, 0.50-0.90) compared with medical therapy alone.</p><p><strong>Conclusion: </strong>We identified three phenotypes with distinct outcomes and HTEs among ICM patients. Patients with lower LVEF, higher NYHA grades, and diabetes had the poorest clinical outcomes but were more likely to derive greater survival benefits from CABG.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 5","pages":"919-928"},"PeriodicalIF":4.4,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12450507/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145126725","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}