European heart journal. Digital health最新文献

{"title":"Machine learning for prediction of key haemodynamic parameters in pulmonary arterial hypertension.","authors":"Tilmann Kramer, Henning Weis, Mira Kramer, Stephan Baldus, Stephan Rosenkranz, Stefan Spinler","doi":"10.1093/ehjdh/ztaf074","DOIUrl":"10.1093/ehjdh/ztaf074","url":null,"abstract":"<p><strong>Aims: </strong>Machine learning (ML) is increasingly recognized for its ability to identify and structure variables for predictive tasks. Pulmonary arterial hypertension (PAH) is a progressive disease characterized by elevated mean pulmonary arterial pressure (mPAP) and pulmonary vascular resistance (PVR) with normal pulmonary arterial wedge pressure (PAWP), as assessed by right heart catheterization (RHC). Despite increased awareness, delays between onset of non-specific symptoms and diagnosis continue to hinder early initiation of targeted therapies, leading to poorer outcomes. To develop and evaluate ML models for predicting key haemodynamic parameters in PAH, based on routinely available non-invasive data collected within 8 weeks prior to RHC, as a proof of concept.</p><p><strong>Methods and results: </strong>We analysed data from 181 patients with invasively confirmed PAH, incorporating 56 variables, including demographics, echocardiography, blood gas analyses, 6-min walk distances, laboratory tests, and WHO functional class. An 80/20 train-test split and fivefold cross-validation were applied across multiple ML models, including least absolute shrinkage and selection operator (lasso) regression, ridge regression, k-nearest neighbours, decision trees, random forest, and gradient boosting machine. Lasso achieved best performance for predicting mPAP (<i>r</i> = 0.80, <i>R</i>² = 0.64, RMSE = 8.49). For PVR, ridge performed best (<i>r</i> = 0.71, <i>R</i>² = 0.51, RMSE = 3.60). Random forest and gradient boosting machines achieved modest but consistent performance for cardiac index (<i>r</i> = 0.38 and 0.37), while PAWP prediction remained limited across all models.</p><p><strong>Conclusion: </strong>Machine learning models can estimate mPAP and PVR from routine clinical data obtained prior to RHC in patients with confirmed PAH. External validation is required to confirm generalizability and clinical applicability.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztaf074"},"PeriodicalIF":4.4,"publicationDate":"2026-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13276432/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148297489","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":"ECG-GPT: considerations for advancing toward clinical implementation.","authors":"Zekai Yu, Feiwei Qin","doi":"10.1093/ehjdh/ztag089","DOIUrl":"10.1093/ehjdh/ztag089","url":null,"abstract":"","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag089"},"PeriodicalIF":4.4,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13278622/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148297472","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":"Correction to: The inclusion of a Holter Reading software in the clinical practice of cardiology shows a multi-level high positive impact in healthcare: a real-world implementation study in three Spanish hospitals.","authors":"","doi":"10.1093/ehjdh/ztag085","DOIUrl":"https://doi.org/10.1093/ehjdh/ztag085","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.1093/ehjdh/ztaf058.].</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag085"},"PeriodicalIF":4.4,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13267979/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148267273","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 ethical application of artificial intelligence in digital health: patients' knowledge, pulmonary hypertension, and the problem of misrecognition.","authors":"Iain Armstrong, Peter Winter","doi":"10.1093/ehjdh/ztag087","DOIUrl":"10.1093/ehjdh/ztag087","url":null,"abstract":"<p><p>Artificial intelligence (AI) is increasingly being introduced into healthcare to improve efficiency, accuracy, and personalisation. Debate has centred on concerns such as bias, safety, and transparency. We argue that another problem deserves much more attention: misrecognition. By this we mean the risk that digital systems know patients mainly through what is easiest to measure and record, while overlooking what is hardest to code but most central to living with illness. Drawing on the concept of epistemic injustice, we suggest that patients may be disbelieved, misunderstood or required to translate complex, embodied, and relational experiences into clinical categories that fit poorly. Our point is not that patients' accounts fall outside data, but that all data require interpretation, and patient and caregiver inputs are often treated as lower-status knowledge in decisions about burden, benefit, and value. These risks do not arise uniformly across computational tools: they take different forms in task-specific machine learning systems used for classification or prediction, and in generative AI systems, including large language models, used to process or generate text. AI may deepen the problem by relying on proxies such as cost, utilization, and adherence thereby hardening narrow understandings of illness into technical systems. Using pulmonary hypertension (PH) as a case, we reflect on patient-led outcome measures such as emPHasis-10 to show that measurement is never neutral: it shapes what counts as legitimate knowledge, meaningful change, and good care. The key question is not only whether AI is accurate or fair, but what patient experiences become visible within it.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag087"},"PeriodicalIF":4.4,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13331348/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148392634","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":"Editorial: listening for congestion in heart failure.","authors":"Fabian Kerwagen, Maximilian Bauser, Stefan Störk","doi":"10.1093/ehjdh/ztag096","DOIUrl":"10.1093/ehjdh/ztag096","url":null,"abstract":"<p><p>Graphical AbstractFor image description, please refer to the figure legend and surrounding text.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag096"},"PeriodicalIF":4.4,"publicationDate":"2026-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13326629/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148378285","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":"AI-augmented ECG for pre-echocardiography triage: a tool to optimize cardiac imaging utilization.","authors":"Abhyuday Kumara Swamy, Deepak Krishnan, Pranay Narhari Umredkar, Aditya Hn, Santhosh Rathnam Palani, Vivek Rajagopal, Pradeep Narayan, Deepak Padmanabhan","doi":"10.1093/ehjdh/ztag082","DOIUrl":"10.1093/ehjdh/ztag082","url":null,"abstract":"<p><strong>Aims: </strong>Echocardiography is a key diagnostic modality for cardiac dysfunction but is often over utilized due to variability in pre-test clinical assessment. There is a need for a scalable, cost-effective screening tool that can reduce unnecessary referrals without compromising diagnostic accuracy. To develop and validate an AI tool that uses standard 12-lead ECG images to predict the presence of major echocardiographic abnormalities, including reduced ejection fraction (EF ≤35%), valvular heart disease, and elevated pulmonary artery pressure, as a triage tool prior to echocardiography.</p><p><strong>Methods and results: </strong>51,055 patients aged ≥15 years from a tertiary cardiac care centre, which underwent ECG and echocardiography on the same day between January 2021, and February 2024 were identified. ECGs were stored as images and pre-processed for model input. Echocardiographic findings were extracted using structured reports and regular expression-based keyword searches. The final dataset (<i>n</i> = 52,817) was split into training (40,796), epoch-monitoring (2,148), and testing (9,873) sets. An ensemble of 3 deep learning models was trained. Model performance was assessed using AUROC, PRAUC, sensitivity, specificity, positive predictive value, and negative predictive value. The internal test set demonstrated an AUROC of 0.87 (95% CI: 0.86-0.88) and PRAUC of 0.66 (95% CI: 0.65-0.69). At the Youden threshold (0.27), sensitivity, specificity, PPV, and NPV were 0.80, 0.80, 0.46, and 0.95, respectively. External validation was performed on 20,053 patients. It yielded an AUROC of 0.84 and PRAUC of 0.50.</p><p><strong>Conclusion: </strong>The proposed AI model accurately identifies major echocardiographic abnormalities from ECG images, achieving high NPV and demonstrating strong generalizability.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag082"},"PeriodicalIF":4.4,"publicationDate":"2026-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13267140/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148267308","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":"Interpretable machine learning models for predicting perioperative myocardial injury in non-cardiac surgery.","authors":"Benjamin Sailer, Sibel Sari-Yavuz, Stephanie Biergans, Raphael Verbücheln, Lars-Christian Achauer, Peter Rosenberger, Michaela Hardt, Michael Koeppen","doi":"10.1093/ehjdh/ztag093","DOIUrl":"10.1093/ehjdh/ztag093","url":null,"abstract":"<p><strong>Aims: </strong>Perioperative myocardial injury (PMI) is a frequent and often asymptomatic complication after non-cardiac surgery and is associated with increased short- and long-term mortality. Conventional risk scores, such as the Revised Cardiac Risk Index (RCRI), have limited predictive accuracy and are infrequently used in clinical practice. We aimed to develop and temporally validate an interpretable machine learning model using Explainable Boosting Machines (EBMs) to predict PMI from routine pre-operative data.</p><p><strong>Methods and results: </strong>In this retrospective cohort study at a tertiary care centre in Germany, we included 9323 adult patients undergoing 9824 non-cardiac surgical procedures between 2014 and 2023 who received post-operative high-sensitivity cardiac troponin testing as part of routine care. PMI was defined as a post-operative elevation of high-sensitivity cardiac troponin above the upper reference limit. An EBM was trained on structured pre-operative data from 2014 to 2021 and evaluated in a temporally independent test cohort from 2022 to 2023, with performance compared with logistic regression, random forest, XGBoost, and a modified RCRI. Model discrimination, calibration, and Brier scores were assessed. Feature contributions were examined using internal shape functions and SHAP values. PMI occurred in 2804 procedures (28.5%). The EBM achieved the highest predictive performance (AUROC 0.730, 95% CI 0.720-0.740), outperforming all comparators. Calibration was robust across clinically relevant risk ranges. Key predictors included age, leukocyte count, renal function, potassium, and platelet count. The EBM identified high-risk patients more efficiently than the modified RCRI and ESC guideline-based strategies (Number Needed to Evaluate 3.0 vs. 3.5) and reduced troponin assays by 18.2% in the temporally independent cohort.</p><p><strong>Conclusion: </strong>An interpretable machine learning model trained on routine clinical data can accurately predict PMI and outperform existing risk scores. The EBM supports individualized risk stratification and may enhance perioperative decision-making and resource allocation within a guideline-directed testing population. Prospective and external validation is required before clinical implementation.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag093"},"PeriodicalIF":4.4,"publicationDate":"2026-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13310016/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148347679","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":"An artificial intelligence model to detect abnormal ejection fraction from non-contrast chest computed tomography: the CT-LVEF study.","authors":"Jayant Raikhelkar, Zilong Bai, Ashley N Beecy, Ilan Richter, Fengbei Liu, Nusrat Binta Nizam, Varsha Kishore, Chris Kelsey, David vanMaanen, Jeffrey Ruhl, Naomi Tesfuzigta, Erica Lancet, Jay Leb, Alan Legasto, Pierre Elias, Timothy Poterucha, Deepa Kumaraiah, Martin Prince, Fei Wang, Gabriel Sayer, Deborah Estrin, Mert Sabuncu, Nir Uriel","doi":"10.1093/ehjdh/ztag088","DOIUrl":"10.1093/ehjdh/ztag088","url":null,"abstract":"<p><strong>Aims: </strong>Heart failure (HF), a major global health challenge, affects millions worldwide and poses substantial healthcare and economic burdens. It is estimated that a large proportion of those with early systolic dysfunction remain asymptomatic at a stage when guideline-directed medical therapies have been shown to prevent disease progression. To develop an artificial intelligence (AI) model capable of predicting abnormal left ventricular ejection fraction (EF) directly from static, non-gated, non-contrast chest computed tomography (CT) scans as a form of opportunistic screening.</p><p><strong>Methods and results: </strong>Using a multi-institutional dataset of 34 058 paired non- contrast CT images and echocardiogram reports from two academic centres, we trained our model of classification for predicting left-ventricle ejection fraction (LVEF) categories: abnormal EF (EF < 50%) vs. normal on 25 948 studies. We validated the model on 8110 paired chest CT and echocardiogram results from a separate institution. The model achieved an area under the receiver operating characteristic (AUROC) curve of 0.786 on the hold-out test set and 0.762 on external validation to detect an abnormal EF (<50%). Beyond strong predictive performance, the AI model surpassed expert radiologists in both accuracy and efficiency and provided interpretable visualizations highlighting imaging features linked to reduced LVEF.</p><p><strong>Conclusion: </strong>In this study, we developed and validated an AI model capable of predicting abnormal LVEF directly from static, non-gated, non-contrast chest CT scans, a novel application for an imaging modality typically used for unrelated indications as a form of opportunistic screening. This technology holds significant promise for early detection of systolic HF, reducing the diagnostic gap, and improving outcomes in asymptomatic HF patients.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag088"},"PeriodicalIF":4.4,"publicationDate":"2026-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13322390/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148371078","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":"From signal to bedside: the missing link between ECG-AI explainability and clinical trust.","authors":"Akinchan Bhardwaj, Ishita Vasudev, Pushkala Jayaraman","doi":"10.1093/ehjdh/ztag092","DOIUrl":"10.1093/ehjdh/ztag092","url":null,"abstract":"","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag092"},"PeriodicalIF":4.4,"publicationDate":"2026-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13390571/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148563935","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":"Comparative analysis of 6-lead and single-lead consumer-grade electrocardiograms: diagnostic accuracy, numerical agreement, and inter-rater reliability.","authors":"Myoung Jung Kim, Sujeong Eom, Juwon Kim, Seung-Jung Park, Kyoung-Min Park, Young Keun On, Sunghoon Joo, Mineok Chang, Yeongyeon Na, Minje Park, Jeonghwa Lim, Taehyung Yu, Hyun Jin Ahn, Ju Youn Kim","doi":"10.1093/ehjdh/ztag086","DOIUrl":"10.1093/ehjdh/ztag086","url":null,"abstract":"<p><strong>Aims: </strong>Consumer-grade electrocardiogram (ECG) devices enable accessible rhythm monitoring. Recently developed six-lead handheld ECGs promise improved signal quality and diagnostic performance; however, comparative validation remains limited. Using 12-lead ECGs as reference, we compared diagnostic performance and quantitative agreement of a 6-lead and single-lead consumer ECGs.</p><p><strong>Methods and results: </strong>In this prospective single-centre study, 194 arrhythmia-clinic patients (498 paired recordings) underwent simultaneous 10-s 12-lead and 30-s 6-lead ECG acquisition, followed by sequential 30-s single-lead recording. Two blinded electrophysiologists interpreted ECGs. Diagnostic performance for predefined rhythm categories (sinus rhythm, atrial/ventricular premature complex, atrial fibrillation/flutter/tachycardia, atrioventricular block, others) was assessed by sensitivity, specificity, and accuracy. Waveform agreement was evaluated using Bland-Altman analysis and intraclass correlation coefficients (ICC) accounting for repeated measures. Sensitivity was higher for the 6-lead ECG for ectopic beats, atrial flutter, and first-degree atrioventricular block, with comparable specificity across categories. Overall diagnostic accuracy was 98.6% (95% confidence interval, 97.4-99.6) for the 6-lead and 96.9% (94.7-98.7) for the single-lead ECG. Agreement with the 12-lead ECG was higher for the 6-lead device for PR interval (ICC 0.89; 0.85-0.92) and QRS amplitude (0.96; 0.94-0.97) than for the single-lead device (0.57; 0.48-0.63 and 0.04; -0.13-0.21, respectively). Bland-Altman analysis demonstrated generally narrower limits of agreement for the 6-lead ECG. Findings were consistent in patient-level sensitivity analyses.</p><p><strong>Conclusion: </strong>The 6-lead handheld ECG demonstrated a higher point estimate for diagnostic accuracy and closer agreement with the 12-lead ECG than the single-lead smartwatch ECG, supporting its use for arrhythmia assessment and interval measurement.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag086"},"PeriodicalIF":4.4,"publicationDate":"2026-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13403557/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148610711","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}