European heart journal. Digital health最新文献

Aims: Rapid myocardial revascularization in patients with acute myocardial infarction (AMI) is essential to improve clinical outcomes. There is still room for improvement in the timely diagnosis of AMI. This study aimed to develop an artificial intelligence (AI) model using electrocardiograms (ECGs) for detecting AMI needing revascularization.

Methods and results: A total of 723 389 ECGs from 300 627 patients in the derivation cohort at a single centre between 2013 and 2020, including 5872 patients with AMI (1.95%) who underwent revascularization, were used for model training and internal testing. A transformer-based deep learning model, initially trained on about one million unlabelled ECGs through self-supervised learning, was fine-tuned for AMI detection. The model's final performance was evaluated in the internal test and the external validation set. The external validation was conducted at an independent centre between 2002 and 2020 using 261 429 ECGs from 259 454 patients, including 1095 patients with AMI (0.42%). By integrating self-supervised learning to train the AI model, we enhanced the AMI detection performance, as demonstrated by an increase in the area under the receiver operating characteristic curve (AUROC) from 0.910 (95% CI, 0.904-0.915) to 0.968 (95% CI, 0.965-0.971) in the external validation set. For ST-elevation myocardial infarction and non-ST-elevation myocardial infarction detection, the AUROCs were 0.991 (95% CI, 0.989-0.993) and 0.947 (95% CI, 0.942-0.952) in the external validation set, respectively.

Conclusion: This novel ECG-based AI model may be beneficial for the timely identification of patients with AMI needing revascularization.

Aims: Deep learning methods have shown impressive performance in detecting a range of diseases from electrocardiogram (ECG) waveforms, but the breadth of diseases that can be detected with high accuracy remains unknown, and in many cases the changes to the ECG allowing these classifications are also opaque. In this study, we aim to determine the full set of cardiac and non-cardiac conditions detectable from the ECG and to understand which ECG features contribute to the disease classification.

Methods and results: Using large datasets of ECGs and connected electronic health records from two separate medical centres, we independently trained PheWASNet, a multi-task deep learning model, to detect 1243 different disease phenotypes from the raw ECG waveform. We confirmed that the ECG can be used to detect chronic kidney disease (AUC = 0.80), cirrhosis (AUC = 0.80), and sepsis (AUC = 0.84), as well as a range of cardiac diseases, and also found new detectable conditions, including respiratory failure (AUC = 0.86), neutropenia (AUC = 0.83), and menstrual disorders (AUC = 0.84). We found that of the 37 non-cardiac strongly detectable conditions, 35 were detectable by the model output for just four diseases, suggesting that they have similar effects on the ECG. We found that high performance in some conditions including neutropenia, respiratory failure, and sepsis can be explained by linear models based on conventional measurements taken from the ECG.

Conclusion: Our study uncovers a range of diseases detectable in the ECG, including many previously unknown phenotypes, and makes progress towards understanding ECG features that allow this detection.

Aims: Whole blood RNA expression is modulated in response to signals from tissues, including the vessel wall. The primary objective of this study was to explore the ability of whole blood transcriptomes, analysed using artificial intelligence (AI), to predict coronary artery calcifications (CAC).

Methods and results: A total of 196 subjects [men aged 40-70 years and women aged 50-70 years without known cardiovascular disease (CVD)] were non-consecutively enrolled for CAC assessment via chest computed tomography. Whole blood RNA was isolated and sequenced. Different AI models were trained using clinical and transcriptomic variables as distinctive features to identify the presence of CAC (Agatston score >0). Finally, we compared the predictive performance of these models. The prevalence of CAC was 43.9%. The combined AI model, incorporating transcriptome data along with age, sex, body mass index, smoking status, diabetes, and hypercholesterolaemia, achieved an area under the curve (AUC) of 0.92 (95% CI, 0.88-0.95) for predicting the presence of CAC, with a sensitivity of 92%, specificity of 80%, positive predictive value of 81%, negative predictive value of 91%, and an overall accuracy of 86%. The combined AI model demonstrated significantly improved discrimination compared with the transcriptomic model (AUC 0.79; P = 0.009), the clinical variables model (AUC 0.72; P < 0.001), and the CVD risk model (AUC 0.68; P < 0.001).

Conclusion: In this pilot study, an AI model integrating whole blood transcriptome data with clinical risk factors demonstrated the ability to predict CAC, providing incremental value over clinical models. Further studies are needed to achieve more robust validation.

Aims: Treatment capacities on intensive care units (ICUs) are a limited resource reserved for high-risk patients. To facilitate risk stratification of ICU patients, several scoring systems have been developed over time. Among them, the Simplified Acute Physiology Score 3 (SAPS3) is the gold standard, but lacks specificity for cardiac ICU patients. Here, we propose a novel, fully automated, electrocardiogram-based cardiac autonomic risk stratification score (CAF_ICU) that substantially adds to current risk stratification strategies.

Methods and results: CAF_ICU is based on periodic repolarization dynamics, a marker of sympathetic overactivity and deceleration capacity of heart rate, a parameter of vagal imbalance. We developed CAF_ICU in a retrospective cohort of 355 ICU patients and subsequently validated the score in a cohort of 702 ICU patients, enrolled between February-November 2018 and December 2018-April 2020 at a large cardiac ICU in a tertiary hospital. The primary endpoint of the study was 30-day intrahospital mortality. Thirty (8.5%) and 100 (14.2%) patients reached the primary endpoint in the training and validation cohorts, respectively. CAF_ICU was significantly higher in non-survivors than survivors (2.58 ± 1.34 vs. 1.76 ± 0.97 units; P = 0.003 in the training cohort and 2.20 ± 1.05 vs. 1.70 ± 0.83 units; P < 0.001 in the validation cohort) and was a strong predictor of mortality in both the training [hazard ratio (HR) 25.67; 95% confidence interval (CI) 3.50-188.40; P = 0.001] and validation cohorts (HR 4.70; 95% CI 2.79-7.92; P < 0.001). In the pooled cohort, CAF_ICU significantly improved risk stratification based on SAPS3 (IDI-increase 0.033; 95% CI 0.010-0.061; P < 0.001).

Conclusion: ECG-based automatic autonomic risk stratification by means of PRD and DC is highly predictive of short-term mortality in the ICU and can be combined with the SAPS3-Score to identify patients with increased risk for intrahospital mortality. This method can be integrated in conventional monitors and may improve risk stratification strategies in cardiac ICUs.

Aims: Cardiac rehabilitation (CR) programmes are effective, but they are underutilized. Digitally enabled CR programmes (DeCR) offer alternative means of healthcare delivery. We aimed to assess the effects of a DeCR programme on cardiovascular risk factors and healthcare utilization.

Methods and results: In this observational cohort study that used propensity score matching, privately insured Australian patients, recruited nationally following a cardiac hospitalization, were given a digital app and received weekly telehealth consultations. Risk factors were assessed before and after the intervention. Propensity scoring methods were used to compare differences in 30-day, 90-day, and 12-month rehospitalizations, hospital-days, and mortality rates in the DeCR group with patients who undertook: (i) usual care (n = 266) or (ii) face-to-face CR (F2F-CR, n = 115). Overall, 172 intervention patients (70% men, age 68 ± 10 years, 36% living in regional/remote areas) were enrolled (59% agreed to participate and 91% completed final follow-up). The DeCR group had significant improvements in most risk factors. Rehospitalization and mortality rates were similar between the DeCR group and both comparison groups at all time points (all P > 0.05). Patients in the DeCR group spent significantly fewer days in hospital compared with usual care within 30-days (P = 0.026), 90-days (P = 0.003), and 12-months (P = 0.04) post-discharge. Cardiac-related rehospitalization bed days were reduced at 30-days (P = 0.005) and 90-days (P = 0.017) but not 12-months (P = 0.20). There were no group differences between DeCR and F2F-CR across any outcomes (all P > 0.05).

Conclusion: DeCR was associated with lower healthcare utilization than usual care, yet comparable compared with F2F-CR. DeCR represents a suitable option for cardiac patients post-discharge.

Lay summary: We investigated whether a digitally enabled cardiac rehabilitation (DeCR) programme, delivered to patients following a heart disease hospitalization, improved patients' cardiovascular disease risk factors and whether they had a reduction in rehospitalizations, spent fewer days in hospital and improved survival compared with matched controls who undertook either face-to-face cardiac rehabilitation (F2F-CR) or usual care.• DeCR was associated with similar healthcare utilization outcomes compared with F2F-CR. This suggests that the potential benefits of DeCR may be comparable. Additionally, DeCR programmes create an opportunity for patients to choose the style of CR to undertake and have an advantage of broader access.• The DeCR group spent significantly fewer readmission days in hospital compared with the usual care group, which may reflect differences in the nature of rehospitalizations when they occur.

Aims: Hypertension is a leading cause of death worldwide, yet many hypertensive cases remain undiagnosed. Wearable, cuffless blood pressure (BP) monitors could be deployed at scale, but their accuracy remains undetermined.

Methods and results: This study validated a popular consumer-grade wearable BP monitor (W-BPM, Aktiia), using a medical-grade ambulatory device (A-BPM, Mobil-O-Graph), for reference. Thirty-one participants (aged 19-62 years, 17 (55%) females, in office BP 121 ± 15 over 77 ± 12 mmHg) simultaneously wore both devices for 24 h. Systolic BP (SBP), diastolic BP (DBP), and heart rate (HR) were measured in pre-set intervals by the A-BPM and at rest by the W-BPM. Agreement was assessed using standard methods. Accuracy in identifying high BP (mean 24 h SBP/DBP > 130/80 mmHg) was assessed. Compared to A-BMP, mean SBP and DBP tended to be slightly lower during the day and not significantly different at night. Nocturnal BP dipping and BP variability were significantly underestimated by the W-BPM. Agreement between the two devices was poor to moderate (limits of agreement of about -30/+30 mmHg for SBP and -20/+15 mmHg for DBP, correlation coefficients between 0.20 and 0.42). Sensitivity and specificity for high BP detection were around 50% and 80%, respectively. Limiting the analysis to measures taken in similar conditions (within 10 min and with HR within ±10 b.p.m.) did not improve agreement.

Conclusion: Low agreement suggests that the cuffless device may not be a suitable replacement for standard 24 h cuff-based ambulatory monitoring. Further data are required to assess the clinical role of cuffless BP monitors.

Aims: Despite advancements in disease-modifying therapies, the rate of hospitalizations in patients with heart failure (HF) remains high, with an increased risk of future adverse events and healthcare costs. In this context, the AZIMUTH study aims to evaluate the large-scale applicability of a smartphone app-based model of care to improve the quality of care and clinical outcomes of HF patients.

Methods and results: The AZIMUTH trial is a multicentre, prospective, pragmatic, interventional, single-cohort study enrolling HF patients. Three hundred patients will be recruited from four different sites. For comparative analyses, historical data from participating hospitals for the 6 months before enrolment and propensity-matching score analyses from GENERATOR HF DataMart, will be used. The estimated duration of the study is 6 months. During the whole observational period, the patients are asked to provide information regarding their clinical status, transmit remote clinical parameters, and periodically answer validated questionnaires, the Kansas City Cardiomyopathy Questionnaire Health and Morisky Medication Adherence Scale 8-item, on a mobile application, through which healthcare providers implement therapeutic adjustments and remote clinical assessments. The primary objective of this study is to evaluate the feasibility, usability, and perceived benefits for key stakeholders (patients and clinical staff) of the AZIMUTH digital platform in the enrolled patients when compared to standard of care. Secondary endpoints will be the description of the rate of hospital readmissions, ambulatory visits and prescribed therapy in the 6 months following enrolment in the experimental group compared to both the historical and propensity-matched cohorts.

Conclusion: The AZIMUTH aims to enhance HF management by leveraging digital technologies to support the care process and enhance monitoring, engagement, and personalized treatment for HF patients.