European heart journal. Digital health最新文献

{"title":"Machine learning-based prediction of mortality and hospitalization in diabetic patients with heart failure with preserved ejection fraction: the GUARDIAN-P risk score.","authors":"Zheng-Wei Chen, Jen-Fang Cheng, Chen-Yu Huang, Tin-Tse Lin, Ting-Chuan Wang, Yen-Yun Yang, Shu-Lin Chuang, Chia-Ti Tsai, Lian-Yu Lin, Chung-Lieh Hung, Cho-Kai Wu","doi":"10.1093/ehjdh/ztag083","DOIUrl":"10.1093/ehjdh/ztag083","url":null,"abstract":"<p><strong>Aims: </strong>Diabetes mellitus (DM) is a major contributor to adverse outcomes in patients with heart failure with preserved ejection fraction (HFpEF). We aim to develop and externally validate a machine learning-based model using a random survival forest (RSF) approach for predicting the composite outcome of hospitalization for heart failure (HHF) and cardiovascular (CV) death in patients with DM and HFpEF.</p><p><strong>Methods and results: </strong>This retrospective cohort study included 1450 adult patients with coexisting DM and HFpEF identified from the National Taiwan University Hospital-Integrated Medical Database. An initial RSF model was trained using 27 clinical variables, and the top 9 predictors were selected to construct a parsimonious final model. Predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC), and external validation was conducted in an independent cohort (<i>n</i> = 729) from MacKay Memorial Hospital. Over a mean follow-up of 3.6 ± 3.0 years, 327 patients (22.6%) experienced the composite outcome. The final RSF model achieved an AUC of 88.2% in the training cohort and 79.8% in the validation cohort. The nine selected predictors were age, N-terminal pro-brain natriuretic peptide, serum albumin, fasting glucose, estimated glomerular filtration rate, uric acid, left atrial diameter, peripheral artery disease, and left ventricular ejection fraction. Risk increased progressively with the number of risk factors present.</p><p><strong>Conclusion: </strong>The RSF-based model incorporating nine routinely available variables accurately predicts HHF and CV death in patients with DM and HFpEF. This tool may support personalized risk assessment and guide clinical decision-making.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag083"},"PeriodicalIF":4.4,"publicationDate":"2026-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13293260/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148335076","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":"Generalizability and validation considerations in AI-guided novice focused cardiac ultrasound.","authors":"Liam Munir, Adele Cai, April Cai","doi":"10.1093/ehjdh/ztag081","DOIUrl":"10.1093/ehjdh/ztag081","url":null,"abstract":"","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 6","pages":"ztag081"},"PeriodicalIF":4.4,"publicationDate":"2026-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13289811/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148321050","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 electrocardiography for left ventricular systolic dysfunction demonstrates preserved performance across demographic training imbalances.","authors":"P Nelson Hsieh, Parth Agrawal, Aman Alok, Sathis Kumar, Charu Ramanathan, Venkatesh L Murthy, Niraj Varma, Venkat Nagarajan, Andrew P Ambrosy, Mattheus Ramsis, Antonis A Armoundas, Jagmeet P Singh","doi":"10.1093/ehjdh/ztag080","DOIUrl":"10.1093/ehjdh/ztag080","url":null,"abstract":"<p><strong>Aims: </strong>Artificial intelligence (AI)-enabled electrocardiograms (AI-ECG) can detect left ventricular systolic dysfunction (LVSD), but demographic imbalance in training datasets may introduce bias. Foundational models, pretrained on large and diverse datasets, may mitigate such concerns. We aimed to assess the impact of demographic composition in training datasets on the performance of an ECG Foundational Model (ECGFM) for diagnosing LVSD.</p><p><strong>Methods and results: </strong>We developed an ECG foundational model (ECGFM) using transformer architecture and self-supervised pretraining on 983 200 ECGs. Using 44 815 paired ECG-echocardiogram datasets, we trained the model under three biased scenarios: (1) sex-skewed (male-only or female-only), (2) race-skewed (White-only or non-White), and (3) balanced. Models were evaluated on a test cohort consisting of 4663 male patients (52%) and 4300 female patients (48%) for the sex configuration and 4440 (49.5%) White, 558 (6.2%) Black, 925 Asian (10.3%), and 3040 other (33.9%) patients, for the race-based configuration using area under the receiver operating characteristic curve (AUROC). The ECGFM demonstrated consistent performance across all demographic configurations. Training on male-only or female-only cohorts yielded comparable AUROC scores of 0.85-0.90 for both sexes in the test set in predicting LVSD. Similarly, training on White-only or non-White cohorts resulted in robust AUROC scores (≥0.90) across all racial groups, including Asian, Black, Hispanic/Latino, and American Indian/Native Alaskan subgroups. Balanced and imbalanced training produced comparable accuracy, sensitivity, and specificity. The performance of the model was externally tested in EchoNext, revealing AUROC scores 0.823-0.917 for sex and 0.822-0.917 for race.</p><p><strong>Conclusion: </strong>Our transformer-based ECG foundational model pretrained using self-supervised learning demonstrated preserved diagnostic accuracy for LVSD across diverse demographic groups, even when trained on demographically imbalanced datasets.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag080"},"PeriodicalIF":4.4,"publicationDate":"2026-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13259591/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148254610","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":"Blood pressure telemonitoring and the incidence of cardiovascular events: a records based, matched patient analysis.","authors":"Janet Hanley, Mary Paterson, Richard Parker, Alice Pearsons, Roz Pollock, Lis Neubeck, Iain Atherton, Brian McKinstry","doi":"10.1093/ehjdh/ztag069","DOIUrl":"10.1093/ehjdh/ztag069","url":null,"abstract":"<p><strong>Aims: </strong>It is well established from trials that blood pressure (BP) telemonitoring leads to improved BP control. However, there is little data available on the impact of BP telemonitoring on the incidence of cardiovascular events when it is used as the routine mode of long-term BP monitoring. The objective of this study was to assess the impact of BP telemonitoring on cardiovascular outcomes.</p><p><strong>Methods and results: </strong>Records were analysed for 454 180 adults with hypertension who had a prescription for a first-line anti-hypertensive drug at any time from 1 March 2019 to 28 February 2021. Follow-up was until 1st March 2022. Women pregnant during that time were excluded. The primary outcome was emergency hospital admission or mortality for Acute Coronary Syndrome (ACS), Stroke, or uncontrolled Heart Failure (HF) within 12 months. Outcomes were compared between people who had used BP telemonitoring for at least 1 year and a group who had never used it, matched for age, sex, ethnicity, social deprivation, number of anti-hypertensive drugs, diabetes, and having a BP assessment in the same year. Ninety percent of the cohort had been diagnosed with hypertension before March 2019. The onset of the COVID-19 pandemic in March 2020 was associated with a rapid increase in the uptake of BP telemonitoring but a reduction in the number of new diagnoses of hypertension. Those who used telemonitoring were significantly younger, less likely to have diabetes, and took less antihypertensive medication. For those who used telemonitoring for over 1 year, a mean reduction in systolic BP was seen by 3 months which was maintained for at least the remainder of the year.</p><p><strong>Conclusion: </strong>In the matched cohort analysis, people who used telemonitoring were less likely to be admitted to hospital with or die from ACS, stroke, or uncontrolled HF telemonitoring [adjusted OR 0.665 (0.501 to 0.884), <i>P</i> = 0.0049] than those who were not using telemonitoring.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag069"},"PeriodicalIF":4.4,"publicationDate":"2026-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13208916/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148044767","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":"Developing and validating an artificial intelligence-based electronic triage model for predicting clinical outcomes among cardiac-suspected patients in the emergency department.","authors":"Ahmad Bavali-Gazik, Elahe Allahyari, Amir Sezavar, Seyed Ali Moezi-Bady, Seyyed Abolfazl Vagharseyyedin","doi":"10.1093/ehjdh/ztag023","DOIUrl":"10.1093/ehjdh/ztag023","url":null,"abstract":"<p><strong>Aims: </strong>Emergency department overcrowding, especially in cardiac units, delays care and raises mortality. Conventional triage is error-prone. We developed an AI-based model integrating routine data and automated ECGs to improve early risk classification.</p><p><strong>Methods and results: </strong>This retrospective cross-sectional study involved 600 medical records of patients presenting with suspected cardiac symptoms. Model development was conducted in three phases: Designing a triage model using routine triage data, designing a triage model based on ECG images, and combining the ECG-based model and triage data. Model performance was evaluated regarding standard clinical outcomes within the first 24 h and compared against the Emergency Severity Index. The best-performing model based on triage data alone (i.e. multilayer perceptron neural network) yielded an accuracy of 89.42%, <i>F</i>-score of 84.51, and area under the curve between 0.815 and 0.858. The best-performing model based on ECG interpretation alone (i.e. convolutional neural network) yielded an accuracy of 93.83%, <i>F</i>-score of 91.08, and area under the curve ranging from 0.852 to 0.914. The fusion model demonstrated superior performance, with an accuracy of 97.22%, <i>F</i>-score of 94.60, and area under the curve between 0.881 and 0.938-significantly outperforming the conventional Emergency Severity Index. In the fusion model, the key predictive variables included ECG interpretation, heart rate, and mode of entry to emergency department.</p><p><strong>Conclusion: </strong>Given its advantages over models using only routine data, ECG, or conventional triage, the fused AI-based triage model may effectively prioritize and predict cardiac emergency outcomes, providing a foundation for developing reliable, intelligent support systems in acute care.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag023"},"PeriodicalIF":4.4,"publicationDate":"2026-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13195029/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148019134","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":"Large language models approach clinician performance in ESC cardiovascular risk stratification: a vignette-based benchmark study.","authors":"José Ferreira Santos, Regina de Brito Duarte, Inês Mota, Rita Carvalheira Santos, José Maria Moreira, Joana Campos, Nuno André Silva, Bernardo Neves, Ricardo Ladeiras-Lopes, Francisca Leite, Hélder Dores","doi":"10.1093/ehjdh/ztag073","DOIUrl":"10.1093/ehjdh/ztag073","url":null,"abstract":"<p><strong>Aims: </strong>Guideline-based cardiovascular risk stratification requires three distinct competencies: extracting risk factor data from clinical text, computing a validated risk score, and applying guideline-defined thresholds to assign a final risk category. We evaluated contemporary large language models (LLMs) on each of these tasks within the European Society of Cardiology (ESC) SCORE2 framework and compared LLM performance against a pooled individual clinician benchmark to contextualize findings against real-world human reproducibility.</p><p><strong>Methods and results: </strong>Eleven LLMs were evaluated using 30 simulated outpatient clinical vignettes presented in both Portuguese and English. For each vignette, models extracted cardiovascular risk factors, determined SCORE2 applicability, generated 10-year risk estimates where appropriate, and assigned a final three-class ESC risk category. A committee of three cardiologists established the reference standard; eight independent clinicians provided an individual-level human benchmark. Traditional risk-factor extraction was near-perfect across all models (micro-F1 0.97-0.99). Agreement with expert-assigned final risk categories was moderate and variable (best: GPT-4o, quadratic-weighted κw 0.69, 95% CI 0.44-0.84), with 10 of 11 models more often underestimating than overestimating risk. To isolate the source of classification error, <i>post hoc</i> deterministic recalculation of SCORE2 was performed using model-extracted variables in eligible vignettes; this markedly improved agreement across all models (κw 0.85-0.90), demonstrating that extraction was largely intact and computational execution was the primary failure mode. The pooled individual clinician benchmark showed moderate agreement with the reference standard (κw 0.52, 95% CI 0.28-0.67), indicating that the best-performing LLMs matched or exceeded the average individual clinician on this guideline-based task. Performance was broadly consistent across Portuguese and English.</p><p><strong>Conclusion: </strong>Contemporary LLMs reliably extract cardiovascular risk information from clinical text, and the best-performing systems achieved agreement within the range of average individual clinicians on this structured task. Their principal limitation lies in downstream computation and rule application.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag073"},"PeriodicalIF":4.4,"publicationDate":"2026-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13215473/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148058718","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 based model for predicting long-term all-cause mortality after acute Myocardial Infarction (the AIMI model).","authors":"Linghan Xue, Wenmiao Wang, Qianli Zhao, Wentao Li, Wenhao Dong, Shaodi Yan, Xiaoxiao Zhao, Jiannan Li, Runzhen Chen, Nan Li, Shuai He, Chen Liu, Peng Zhou, Yi Chen, Li Song, Hongbing Yan, Zhi Liu, Hanjun Zhao","doi":"10.1093/ehjdh/ztag078","DOIUrl":"10.1093/ehjdh/ztag078","url":null,"abstract":"<p><strong>Aims: </strong>Predicting long-term mortality after acute myocardial infarction (AMI) remains challenging. We aimed to establish an Artificial Intelligence-based model for predicting long-term all-cause mortality after AMI (the AIMI model).</p><p><strong>Methods and results: </strong>AIMI model was employed by RF (Random forest). Individual predictions were visualized by SHAP plots. AIMI model was compared against existing clinical risk scores using time-dependent ROC (receiver operating characteristic) curves, and Kaplan-Meier (K-M) analyses. External validation was also performed at the same way. Brier scores were calculated in validation cohorts. We consecutively enrolled 4825 AMI patients underwent emergent coronary angiography or PCI procedures within 24 h of symptom onset to train and test the AIMI model and 723 AMI patients for external validation. Model incorporated 15 variables achieved robust performance (C-index = 0.81). As indicated by AUCs in the test set, AIMI model outperformed GRACE and TIMI risk scores across short-, mid- and long-term periods, especially for long-term prediction (1, 3 and 5 years). K-M curves confirmed precise discrimination between low-, median-, and high-risk groups (all <i>P</i> < 0.05). External validation confirmed good generalization and robustness for AIMI model (AUCs: 0.88, 0.91, 0.83, 0.75, 0.78 and 0.77 during hospitalization, at 30 days, half-year, 1-year, 2-years and 3-years follow-up; comparisons of K-M curves across three risk groups, all <i>P</i> < 0.05). Brier scores demonstrated good individual level performance (internal validation cohort: 0.022; external validation: 0.021).</p><p><strong>Conclusion: </strong>The AIMI model surpassed traditional methods for long-term all-cause death prediction after AMI. AI-based model demonstrated potential to enhance risk stratification and guide post-discharge management.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag078"},"PeriodicalIF":4.4,"publicationDate":"2026-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13270488/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148273912","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":"Monitoring cardiac symptoms during teleprehabilitation prior to coronary artery bypass grafting: a digital cardiac counselling trial subanalysis.","authors":"Lieke van Susante, Rutger van Mierlo, Bart Scheenstra, Arnoud van 't Hof, Peyman Sardari Nia","doi":"10.1093/ehjdh/ztag076","DOIUrl":"10.1093/ehjdh/ztag076","url":null,"abstract":"<p><strong>Aims: </strong>The purpose of this subanalysis was to explore whether or not teleprehabilitation leads to worsening cardiac symptoms in patients accepted for coronary artery bypass grafting (CABG).</p><p><strong>Methods and results: </strong>A subgroup of patients accepted for CABG (including off-pump and minimally invasive) who were randomized between teleprehabilitation and standard care in DCC trial were included. Teleprehabilitation patients had access to blended care supported by online modules, including nutritional support, functional exercise training, smoking cessation, inspiratory muscle training, and psychological support. All patients were weekly telemonitored for cardiac symptoms using a customized digital platform. This cohort included 155 patients. Baseline characteristics, including NYHA and CCS class, were balanced, except for shorter preoperative waiting times in weeks (15.0 [9.8-22.2] vs. 12.5 [7.0-16.0]) in the teleprehabilitation group. There was no significant difference in incidence of major adverse cardiac events in the preoperative period. There were 141 patients (77 control; 64 teleprehabilitation) with at least one symptom monitoring entry. A total of 373/1786 (21%) entries indicated worsening of cardiac symptoms. The distribution was right-skewed and did not significantly differ between the two groups. There was no statistically significant difference in time to event between control and teleprehabilitation for any cardiac symptom (<i>P</i> = 0.529), angina (<i>P</i> = 0.167), or dyspnoea (<i>P</i> = 0.334).</p><p><strong>Conclusion: </strong>Teleprehabilitation does not lead to worsening cardiac symptoms before CABG. We postulate that teleprehabilitation, which includes low-intensity physical training, can be safely offered to patients undergoing elective CABG.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag076"},"PeriodicalIF":4.4,"publicationDate":"2026-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13228138/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148159239","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":"Prior cardiac events and the interpretation of ECG-based artificial intelligence for inducible myocardial ischaemia.","authors":"Liam Munir, Adele Cai","doi":"10.1093/ehjdh/ztag077","DOIUrl":"10.1093/ehjdh/ztag077","url":null,"abstract":"","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag077"},"PeriodicalIF":4.4,"publicationDate":"2026-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13228141/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148159251","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":"Smarter FoCUS: AI-guided focused cardiac ultrasound enables novice detection of left ventricular dysfunction.","authors":"Conor J Kane, Jordan Borgeson, Christie Greason, Jeffrey G Malins, D M Anisuzzaman, Gal Tsaban, Daniel Schonfeld, John I Jackson, Julia Wood, Eunjung Lee, Jwan A Naser, Paul A Friedman, Francisco Lopez-Jimenez, Garvan C Kane, Jae K Oh, Patricia A Pellikka, Timothy J Poterucha, Andrew Rosenbaum, Zachi I Attia, Jeremy J Thaden, Sorin V Pislaru, Jared G Bird","doi":"10.1093/ehjdh/ztag066","DOIUrl":"10.1093/ehjdh/ztag066","url":null,"abstract":"<p><strong>Aims: </strong>Heart failure is prevalent; however, there is no cost-effective screening option. Against formal echocardiography, we assessed the diagnostic performance of focused cardiac ultrasound (FoCUS), performed by novice users guided and interpreted by artificial intelligence (AI) for the assessment of left ventricular ejection fraction (LVEF).</p><p><strong>Methods and results: </strong>We prospectively enrolled 496 adults referred for diagnostic echocardiography. Novice operators (without clinical or imaging experience) underwent a 4-h imaging workshop, then performed FoCUS using a point-of-care device (Philips Lumify) utilizing real-time AI-image guidance (UltraSight). Images were assessed by AI-image interpretation software (Mayo Clinic), and two expert blinded echocardiologists. Median age was 67, 39% female, and body mass index range 17-56 kg/m². Forty-one subjects (8.3%) exhibited moderate or greater LV dilatation and 28 (5.6%) had LVEF <40%. The median scan time was 4 min (IQR 3-5). Adequate views were achieved in 95.0% and 97.4% of subjects for AI interpretation and expert analysis, respectively. A two-step diagnostic screening process for low EF in which only abnormal AI reads or uninterpretable scans were reviewed by experts (15.1%) had a sensitivity 96.2%, specificity 95.4%, PPV 54.3%, and NPV 99.8%. A subsequent prospective validation study of 344 subjects demonstrated similar results (sensitivity 100%, specificity 99.4%).</p><p><strong>Conclusion: </strong>In this early feasibility investigation, AI-guidance technology embedded on a handheld device, enabled novice users without clinical or imaging experience to acquire sufficient quality images to accurately assess LVEF with minimal training. The technology evaluated in this study may hold promise for AI-guidance and interpretation to facilitate low-cost screening for cardiac dysfunction.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"7 5","pages":"ztag066"},"PeriodicalIF":4.4,"publicationDate":"2026-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13188223/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147990163","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}