European heart journal. Digital health最新文献

{"title":"Interpretable predictive value of including HDL-2b and HDL-3 in an explainable boosting machine model for multiclass classification of coronary artery stenosis severity in acute myocardial infarction patients.","authors":"Bin Wang, Dong Li, Yu Geng, Feifei Jin, Yujie Wang, Changhua Lv, Tingting Lv, Yajun Xue, Ping Zhang","doi":"10.1093/ehjdh/ztae100","DOIUrl":"10.1093/ehjdh/ztae100","url":null,"abstract":"<p><strong>Aims: </strong>The aim of this study was to use explainable boosting machine (EBM) to evaluate the predictive value of HDL-2b and HDL-3 levels in comparison with traditional lipid parameters in three-class classification of coronary artery stenosis severity in acute myocardial infarction (AMI) patients.</p><p><strong>Methods and results: </strong>In this cross-sectional study, 1200 AMI patients were evaluated. HDL subtypes were quantified via microfluidic chip detection, and stenosis severity was assessed via the Gensini scoring system. The Gensini scores were divided into three groups: low group (<36.5), moderate group (36.5-72), and high group (>72). Explainable boosting machine, an interpretable machine learning technique, was employed to assess the predictive value of HDL-2b and HDL-3 compared with traditional lipid markers. Explainable boosting machine was used as the main model in this study, whereas logistic regression, XGBoost, and Random Forest were selected as reference models for predictive performance. Model performance was evaluated using receiver operating characteristic curves. The HDL-3 (%) values were divided into three risk categories: low (>43), moderate (30-43), and high (<30). The incorporation of HDL-2b and HDL-3 levels into lipid profiling significantly increased the group importance scores. The macro-average area under the curve values for the four models were as follows: 0.56 for the logistic model, 0.54 for the EBM model, 0.50 for the Random Forest model, and 0.49 for the XGBoost model.</p><p><strong>Conclusion: </strong>HDL-3 provides superior predictive value for evaluating coronary artery stenosis severity in AMI patients compared to HDL-2b and other conventional lipid markers.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 2","pages":"228-239"},"PeriodicalIF":3.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11914717/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665531","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 score to predict in-hospital outcomes in patients hospitalized in cardiac intensive care unit.","authors":"Orianne Weizman, Kenza Hamzi, Patrick Henry, Guillaume Schurtz, Marie Hauguel-Moreau, Antonin Trimaille, Marc Bedossa, Jean Claude Dib, Sabir Attou, Tanissia Boukertouta, Franck Boccara, Thibaut Pommier, Pascal Lim, Thomas Bochaton, Damien Millischer, Benoit Merat, Fabien Picard, Nissim Grinberg, David Sulman, Bastien Pasdeloup, Yassine El Ouahidi, Treçy Gonçalves, Eric Vicaut, Jean-Guillaume Dillinger, Solenn Toupin, Théo Pezel","doi":"10.1093/ehjdh/ztae098","DOIUrl":"10.1093/ehjdh/ztae098","url":null,"abstract":"<p><strong>Aims: </strong>Although some scores based on traditional statistical methods are available for risk stratification in patients hospitalized in cardiac intensive care units (CICUs), the interest of machine learning (ML) methods for risk stratification in this field is not well established. We aimed to build an ML model to predict in-hospital major adverse events (MAE) in patients hospitalized in CICU.</p><p><strong>Methods and results: </strong>In April 2021, a French national prospective multicentre study involving 39 centres included all consecutive patients admitted to CICU. The primary outcome was in-hospital MAE, including death, resuscitated cardiac arrest, or cardiogenic shock. Using 31 randomly assigned centres as an index cohort (divided into training and testing sets), several ML models were evaluated to predict in-hospital MAE. The eight remaining centres were used as an external validation cohort. Among 1499 consecutive patients included (aged 64 ± 15 years, 70% male), 67 had in-hospital MAE (4.3%). Out of 28 clinical, biological, ECG, and echocardiographic variables, seven were selected to predict MAE in the training set (<i>n</i> = 844). Boosted cost-sensitive C5.0 technique showed the best performance compared with other ML methods [receiver operating characteristic area under the curve (AUROC) = 0.90, precision-recall AUC = 0.57, <i>F</i>1 score = 0.5]. Our ML score showed a better performance than existing scores (AUROC: ML score = 0.90 vs. Thrombolysis In Myocardial Infarction (TIMI) score: 0.56, Global Registry of Acute Coronary Events (GRACE) score: 0.52, Acute Heart Failure (ACUTE-HF) score: 0.65; all <i>P</i> < 0.05). Machine learning score also showed excellent performance in the external cohort (AUROC = 0.88).</p><p><strong>Conclusion: </strong>This new ML score is the first to demonstrate improved performance in predicting in-hospital outcomes over existing scores in patients admitted to the intensive care unit based on seven simple and rapid clinical and echocardiographic variables.</p><p><strong>Trial registration: </strong>ClinicalTrials.gov Identifier: NCT05063097.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 2","pages":"218-227"},"PeriodicalIF":3.9,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11914730/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665534","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":"Efficacy of an Internet of Things-based system for cardiac rehabilitation monitoring: insights from the IntellIoT pivotal trial in heart failure patients.","authors":"Maria E Marketou, Ioannis Anastasiou, Alexis Fourlis, Aphrodite Alevizaki, George Kochiadakis","doi":"10.1093/ehjdh/ztae093","DOIUrl":"10.1093/ehjdh/ztae093","url":null,"abstract":"<p><strong>Aims: </strong>Digital health solutions targeted to remote clinical monitoring are constantly gaining ground in cardiovascular care. However, evidence regarding their impact on cardiac rehabilitation efficiency in heart failure (HF) patients is relatively limited. In this study, conducted in the context of the IntellIoT project, we evaluated the effect of a purpose-designed Internet of Things (IoT)-based patient monitoring system on cardiac rehabilitation outcomes in a cohort of HF patients.</p><p><strong>Methods and results: </strong>Nineteen clinically stable HF patients were enrolled in the study, which consisted of a 12-month standard-of-care run-in phase and a remote follow-up phase of equal duration, whereby an IoT-based e-health system was provided to study subjects. Device-derived data transmission was facilitated by a mobile phone application, coupled with a web-based platform accessible to study physicians. Study endpoints were (i) patient adherence rates to e-health system use and their associations to key clinical parameters, (ii) the degree of change in physical activity, and (iii) total time dedicated by physicians to enrolled patients' care with and without the aid of the e-health system. Baseline-to-peak increase in daily step count was calculated at 23.34%. System use was associated with a decrease in time dedicated by physicians to enrolled patients' care. A significant negative correlation was observed between age and progressive drop-in adherence rate to system use (<i>r</i> = -0.5722, <i>P</i> = 0.02).</p><p><strong>Conclusion: </strong>Internet of Things-based healthcare constitutes a promising approach in HF patients' rehabilitation, whereas elderly patients might constitute the population most likely to benefit. However, larger, randomized studies are required to confirm our findings.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 2","pages":"293-297"},"PeriodicalIF":3.9,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11914715/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665594","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":"Supervised machine learning including environmental factors to predict in-hospital outcomes in acute heart failure patients.","authors":"Benjamin Sibilia, Solenn Toupin, Nabil Bouali, Jean-Baptiste Brette, Arthur Ramonatxo, Guillaume Schurtz, Kenza Hamzi, Antonin Trimaille, Emmanuel Gall, Nicolas Piliero, Alexandre Unger, Stéphane Andrieu, Trecy Gonçalves, Fabien Picard, Vincent Roule, François Roubille, Sonia Houssany-Pissot, Océane Bouchot, Victor Aboyans, Reza Rossanaly Vasram, Thomas Bochaton, Damien Logeart, Alain Cohen Solal, Jérôme Cartailler, Alexandre Mebazaa, Jean-Guillaume Dillinger, Patrick Henry, Théo Pezel","doi":"10.1093/ehjdh/ztae094","DOIUrl":"10.1093/ehjdh/ztae094","url":null,"abstract":"<p><strong>Aims: </strong>While few traditional scores are available for risk stratification of patients hospitalized for acute heart failure (AHF), the potential benefit of machine learning (ML) is not well established. We aimed to assess the feasibility and accuracy of a supervised ML model including environmental factors to predict in-hospital major adverse events (MAEs) in patients hospitalized for AHF.</p><p><strong>Methods and results: </strong>In April 2021, a French national prospective multicentre study included all consecutive patients hospitalized in intensive cardiac care unit. Patients admitted for AHF were included in the analyses. A ML model involving automated feature selection by least absolute shrinkage and selection operator (LASSO) and model building with a random forest (RF) algorithm was developed. The primary composite outcome was in-hospital MAE defined by death, resuscitated cardiac arrest, or cardiogenic shock requiring assistance. Among 459 patients included (age 68 ± 14 years, 68% male), 47 experienced in-hospital MAE (10.2%). Seven variables were selected by LASSO for predicting MAE in the training data set (<i>n</i> = 322): mean arterial pressure, ischaemic aetiology, sub-aortic velocity time integral, E/e', tricuspid annular plane systolic excursion, recreational drug use, and exhaled carbon monoxide level. The RF model showed the best performance compared with other evaluated models [area under the receiver operating curve (AUROC) = 0.82, 95% confidence interval (CI) (0.78-0.86); precision-recall area under the curve = 0.48, 95% CI (0.42-0.5), <i>F</i>1 score = 0.56). Our ML model exhibited a higher AUROC compared with an existing score for the prediction of MAE (AUROC for our ML model: 0.82 vs. ACUTE HF score: 0.57; <i>P</i> < 0.001).</p><p><strong>Conclusion: </strong>Our ML model including in particular environmental variables exhibited a better performance than traditional statistical methods to predict in-hospital outcomes in patients admitted for AHF.</p><p><strong>Study registration: </strong>ClinicalTrials.gov identifier: NCT05063097.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 2","pages":"190-199"},"PeriodicalIF":3.9,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11914725/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143664048","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":"Multi-modal artificial intelligence algorithm for the prediction of left atrial low-voltage areas in atrial fibrillation patient based on sinus rhythm electrocardiogram and clinical characteristics: a retrospective, multicentre study.","authors":"Yirao Tao, Deyun Zhang, Naidong Pang, Shijia Geng, Chen Tan, Ying Tian, Shenda Hong, XingPeng Liu","doi":"10.1093/ehjdh/ztae095","DOIUrl":"10.1093/ehjdh/ztae095","url":null,"abstract":"<p><strong>Aims: </strong>We aimed to develop an artificial intelligence (AI) algorithm capable of accurately predicting the presence of left atrial low-voltage areas (LVAs) based on sinus rhythm electrocardiograms (ECGs) in patients with atrial fibrillation (AF).</p><p><strong>Methods and results: </strong>The study included 1133 patients with AF who underwent catheter ablation procedures, with a total of 1787 12-lead ECG images analysed. Artificial intelligence-based algorithms were used to construct models for predicting the presence of LVAs. The DR-FLASH and APPLE clinical scores for LVAs prediction were calculated. A receiver operating characteristic (ROC) curve and a calibration curve were used to evaluate model performance. Multicentre validation included 92 AF patients from five centres, with a total of 174 ECGs. The data obtained from the participants were split into training (<i>n</i> = 906), validation (<i>n</i> = 113), and test sets (<i>n</i> = 114). Low-voltage areas were detected in 47.4% of all participants. Using ECG alone, the convolutional neural network (CNN) model achieved an area under the ROC curve (AUROC) of 0.704, outperforming both the DR-FLASH score (AUROC = 0.601) and the APPLE score (AUROC = 0.589). Two multimodal AI models, which integrated ECG images and clinical features, demonstrated higher diagnostic accuracy (AUROC 0.816 and 0.796 for the CNN-Multimodal and CNN-Random Forest-Multimodal models, respectively). Our models also performed well in the multicentre validation dataset (AUROC 0.711, 0.785, and 0.879 for the ECG alone, CNN-Multimodal, and CNN-Random Forest-Multimodal models, respectively).</p><p><strong>Conclusion: </strong>The multimodal AI algorithm, which integrated ECG images and clinical features, predicted the presence of LVAs with a higher degree of accuracy than ECG alone and the clinical LVA scores.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 2","pages":"200-208"},"PeriodicalIF":3.9,"publicationDate":"2024-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11914728/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665551","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":"Research on atrial fibrillation diagnosis in electrocardiograms based on CLA-AF model.","authors":"Jiajia Si, Yiliang Bao, Fengling Chen, Yue Wang, Meimei Zeng, Nongyue He, Zhu Chen, Yuan Guo","doi":"10.1093/ehjdh/ztae092","DOIUrl":"10.1093/ehjdh/ztae092","url":null,"abstract":"<p><strong>Aims: </strong>The electrocardiogram (ECG) is the primary method for diagnosing atrial fibrillation (AF), but interpreting ECGs can be time-consuming and labour-intensive, which deserves more exploration.</p><p><strong>Methods and results: </strong>We collected ECG data from 6590 patients as YY2023, classified as Normal, AF, and Other. Convolutional Neural Network (CNN), bidirectional Long Short-Term Memory (BiLSTM), and Attention construct the AF recognition model CNN BiLSTM Attention-Atrial Fibrillation (CLA-AF). The generalization ability of the model is validated on public datasets CPSC2018, PhysioNet2017, and PTB-XL, and we explored the performance of oversampling, resampling, and hybrid datasets. Finally, additional PhysioNet2021 was added to validate the robustness and applicability in different clinical settings. We employed the SHapley Additive exPlanations (SHAP) method to interpret the model's predictions. The F1-score, Precision, and area under the ROC curve (AUC) of the CLA-AF model on YY2023 are 0.956, 0.970, and 1.00, respectively. Similarly, the AUC on CPSC2018, PhysioNet2017, and PTB-XL reached above 0.95, demonstrating its strong generalization ability. After oversampling PhysioNet2017, F1-score and Recall improved by 0.156 and 0.260. Generalization ability varied with sampling frequency. The model trained from the hybrid dataset has the most robust generalization ability, achieving an AUC of 0.96 or more. The AUC of PhysioNet2021 is 1.00, which proves the applicability of CLA-AF. The SHAP values visualization results demonstrate that the model's interpretation of AF aligns with the diagnostic criteria of AF.</p><p><strong>Conclusion: </strong>The CLA-AF model demonstrates a high accuracy in recognizing AF from ECG, exhibiting remarkable applicability and robustness in diverse clinical settings.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 1","pages":"82-95"},"PeriodicalIF":3.9,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11750197/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143025878","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":"Wristwatch pulse wave monitoring: assessing daily activity post-catheter ablation for atrial fibrillation.","authors":"Noriko Matsushita Nonoguchi, Kyoko Soejima, Yumi Katsume, Kyoko Hoshida, Ikuko Togashi, Ayumi Goda, Akiko Ueda, Seiichiro Matsuo, Toshiaki Sato, Yuichi Takano, Fumio Koyama, Shin Fujita, Kunihiro Nishimura, Takashi Kohno","doi":"10.1093/ehjdh/ztae091","DOIUrl":"10.1093/ehjdh/ztae091","url":null,"abstract":"<p><strong>Aims: </strong>Atrial fibrillation (AF) leads to impaired exercise capacity, and catheter ablation (CA) for AF improves exercise capacity. However, the precise changes in daily activities after CA for AF remain unclear. The authors aimed to evaluate the changes in daily activities following CA for AF using a wristwatch-type pulse wave monitor (PWM), which tracks steps and exercise time, estimates burnt daily calories, and records sleep duration, in addition to establishing the rhythm diagnosis of AF or non-AF.</p><p><strong>Methods and results: </strong>One hundred and twenty-three patients with AF (97 paroxysmal, 26 persistent) wore a wristwatch-type PWM for 1 week duration at three time points: before, 1 month after, and 3 months after ablation. Daily activity data were compared. Steps did not change in both groups, and the number of burnt daily calories and total exercise time increased after CA in patients with paroxysmal AF (burnt daily calories: before, 1591 kcal/day; 1 month, 1688 kcal/day; and 3 months, 1624 kcal/day; <i>P</i> < 0.001 and exercise time: before, 45.8 min; 1 month, 51.2 min; and 3 months, 56.3 min; <i>P</i> = 0.023). Sleep hours significantly increased (paroxysmal AF: before, 6.8 h; 1 month, 7.1 h; and 3 months, 7.1 h; <i>P</i> = 0.039 and persistent AF: before, 6.0 h; 1 month, 7.0 h; and 3 months, 7.0 h; <i>P</i> = 0.007).</p><p><strong>Conclusion: </strong>Using a wristwatch-type PWM, we demonstrated changes in daily activities after CA in patients with AF.</p><p><strong>Trial registration number: </strong>jRCT1030210022.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 1","pages":"96-103"},"PeriodicalIF":3.9,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11750189/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143025895","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":"Anonymize or synthesize? Privacy-preserving methods for heart failure score analytics.","authors":"Tim I Johann, Karen Otte, Fabian Prasser, Christoph Dieterich","doi":"10.1093/ehjdh/ztae083","DOIUrl":"10.1093/ehjdh/ztae083","url":null,"abstract":"<p><strong>Aims: </strong>Data availability remains a critical challenge in modern, data-driven medical research. Due to the sensitive nature of patient health records, they are rightfully subject to stringent privacy protection measures. One way to overcome these restrictions is to preserve patient privacy by using anonymization and synthetization strategies. In this work, we investigate the effectiveness of these methods for protecting patient privacy using real-world cardiology health records.</p><p><strong>Methods and results: </strong>We implemented anonymization and synthetization techniques for a structure data set, which was collected during the HiGHmed Use Case Cardiology study. We employed the data anonymization tool ARX and the data synthetization framework ASyH individually and in combination. We evaluated the utility and shortcomings of the different approaches by statistical analyses and privacy risk assessments. Data utility was assessed by computing two heart failure risk scores on the protected data sets. We observed only minimal deviations to scores from the original data set. Additionally, we performed a re-identification risk analysis and found only minor residual risks for common types of privacy threats.</p><p><strong>Conclusion: </strong>We could demonstrate that anonymization and synthetization methods protect privacy while retaining data utility for heart failure risk assessment. Both approaches and a combination thereof introduce only minimal deviations from the original data set over all features. While data synthesis techniques produce any number of new records, data anonymization techniques offer more formal privacy guarantees. Consequently, data synthesis on anonymized data further enhances privacy protection with little impacting data utility. We share all generated data sets with the scientific community through a use and access agreement.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 1","pages":"147-154"},"PeriodicalIF":3.9,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11750188/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143025791","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":"Early discharge programme after transcatheter aortic valve implantation based on close follow-up supported by telemonitoring using artificial intelligence: the TeleTAVI study.","authors":"Marta Herrero-Brocal, Raquel Samper, Jorge Riquelme, Javier Pineda, Pascual Bordes, Fernando Torres-Mezcua, José Valencia, Francisco Torres-Saura, María González Manso, Raquel Ajo, Juan Arenas, Eloísa Feliu, Juan Gabriel Martínez, Juan Miguel Ruiz-Nodar","doi":"10.1093/ehjdh/ztae089","DOIUrl":"10.1093/ehjdh/ztae089","url":null,"abstract":"<p><strong>Aims: </strong>Evidence regarding the safety of early discharge following transcatheter aortic valve implantation (TAVI) is limited. The aim of this study was to evaluate the safety of very early (<24) and early discharge (24-48 h) as compared to standard discharge (>48 h), supported by the implementation of a voice-based virtual assistant using artificial intelligence (AI) and natural language processing.</p><p><strong>Methods and results: </strong>Single-arm prospective observational study that included consecutive patients who underwent TAVI in a tertiary hospital in 2023 and were discharged under an AI follow-up programme. Primary endpoint was a composite of death, pacemaker implantation, readmission for heart failure, stroke, acute myocardial infarction, major vascular complications, or major bleeding, at 30-day follow-up. A total of 274 patients were included. 110 (40.1%) patients were discharged very early (<24 h), 90 (32.9%) early (24-48 h), and 74 (27.0%) were discharged after 48 h. At 30-day follow-up, no significant differences were found among patients discharged very early, early, and those discharged after 48 h for the primary endpoint (very early 9.1% vs. early 11.1% vs. standard 9.5%; <i>P</i> = 0.88). The AI platform detected complications that could be effectively addressed. The implementation of this follow-up system was simple and satisfactory for TAVI patients.</p><p><strong>Conclusion: </strong>Early and very early discharge in patients undergoing TAVI, supported by close follow-up using AI, were shown to be safe. Patients with early and very early discharge had similar 30-day event rates compared to those with longer hospital stays. The AI system contributed to the early detection and resolution of complications.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 1","pages":"73-81"},"PeriodicalIF":3.9,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11750190/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143025839","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":"A comparison of artificial intelligence-enhanced electrocardiography approaches for the prediction of time to mortality using electrocardiogram images.","authors":"Arunashis Sau, Boroumand Zeidaabadi, Konstantinos Patlatzoglou, Libor Pastika, Antônio H Ribeiro, Ester Sabino, Nicholas S Peters, Antonio Luiz P Ribeiro, Daniel B Kramer, Jonathan W Waks, Fu Siong Ng","doi":"10.1093/ehjdh/ztae090","DOIUrl":"10.1093/ehjdh/ztae090","url":null,"abstract":"<p><strong>Aims: </strong>Most artificial intelligence-enhanced electrocardiogram (AI-ECG) models used to predict adverse events including death require that the ECGs be stored digitally. However, the majority of clinical facilities worldwide store ECGs as images.</p><p><strong>Methods and results: </strong>A total of 1 163 401 ECGs (189 539 patients) from a secondary care data set were available as both natively digital traces and PDF images. A digitization pipeline extracted signals from PDFs. Separate 1D convolutional neural network (CNN) models were trained on natively digital or digitized ECGs, with a discrete-time survival loss function to predict <i>time to mortality</i>. A 2D CNN model was trained on 310 × 868 px ECG images. External validation was performed in 958 954 ECGs (645 373 patients) from a Brazilian primary care cohort and 1022 ECGs (1022 patients) from a Chagas disease cohort. The image 2D CNN model and digitized 1D CNN model performed comparably to natively digital 1D CNN model in internal [C-index 0.780 (0.779-0.781), 0.772 (0.771-0.774), and 0.775 (0.774-0.776), respectively] and external validation. Models trained on natively digital 1D ECGs had comparable performance when applied to digitized 1D ECGs [C-index 0.773 (0.771-0.774)].</p><p><strong>Conclusion: </strong>Both the image 2D CNN and digitized 1D CNN enable mortality prediction from ECG images, with comparable performance to natively digital 1D CNN. Models trained on natively digital 1D ECGs can also be applied to digitized 1D ECGs, without any significant loss in performance. This work allows AI-ECG mortality prediction to be applied in diverse global settings lacking digital ECG infrastructure.</p>","PeriodicalId":72965,"journal":{"name":"European heart journal. Digital health","volume":"6 2","pages":"180-189"},"PeriodicalIF":3.9,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11914724/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665489","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}