ABSTRACT
BACKGROUND AND PURPOSE: The electrocardiogram (ECG) is frequently obtained in the work-up of COVID-19 patients. So far, no study has evaluated whether ECG-based machine learning models have added value to predict in-hospital mortality specifically in COVID-19 patients. METHODS: Using data from the CAPACITY-COVID registry, we studied 882 patients admitted with COVID-19 across seven hospitals in the Netherlands. Raw format 12-lead ECGs recorded within 72â¯h of admission were studied. With data from five hospitals (nâ¯= 634), three models were developed: (a) a logistic regression baseline model using age and sex, (b) a least absolute shrinkage and selection operator (LASSO) model using age, sex and human annotated ECG features, and (c) a pre-trained deep neural network (DNN) using age, sex and the raw ECG waveforms. Data from two hospitals (nâ¯= 248) was used for external validation. RESULTS: Performances for models a, b and c were comparable with an area under the receiver operating curve of 0.73 (95% confidence interval [CI] 0.65-0.79), 0.76 (95% CI 0.68-0.82) and 0.77 (95% CI 0.70-0.83) respectively. Predictors of mortality in the LASSO model were age, low QRS voltage, ST depression, premature atrial complexes, sex, increased ventricular rate, and right bundle branch block. CONCLUSION: This study shows that the ECG-based prediction models could be helpful for the initial risk stratification of patients diagnosed with COVID-19, and that several ECG abnormalities are associated with in-hospital all-cause mortality of COVID-19 patients. Moreover, this proof-of-principle study shows that the use of pre-trained DNNs for ECG analysis does not underperform compared with time-consuming manual annotation of ECG features.
ABSTRACT
For all patients with cardiovascular disease requiring an intervention, this is a major life event. The heart team concept is one of the most exciting and effective team modalities to ensure cost-effective application of invasive cardiovascular care. It optimises patient selection in a complex decision-making process and identifies risk/benefit ratios of different interventions. Informed consent and patient safety should be at the centre of these decisions. To deal with increased load of medical data in the future, artificial intelligence could enable objective and effective interpretation of medical imaging and decision support. This technical support is indispensable to meet current patient and societal demands for informed consent, shared decision-making, outcome improvement and safety. The heart team should be restructured with clear leadership, accountability, and process and outcome measurement of interventions. In this way, the heart team concept in the Netherlands will be ready for the future.
ABSTRACT
Quality-of-care registries have been shown to improve quality of healthcare and should be facilitated and encouraged. The data of these registries are also very valuable for medical data research. While fully acknowledging the importance of re-using already available data for research purposes, there are concerns about how the applicable privacy legislation is dealt with. These concerns are also articulated in the new European law on privacy, the 'General Data Protection Regulation' (GDPR) which has come into force on 25 May 2018. The aim of this review is to examine what the implications of the new European data protection rules are for quality-of-care registries in Europe while providing examples of three quality-of-care registries in the field of cardiology and cardiothoracic surgery in Europe. A general overview of the European and national legal framework (relevant data protection and privacy legislation) applying to quality-of-care registries is provided. One of the main rules is that non-anonymous patient data may, in principle, not be used for research without the patient's informed consent. When patient data are solely and strictly used for quality control and improvement, this rule does not apply. None of the described registries (NHR, SWEDEHEART, and NICOR) currently ask specific informed consent of patients before using their data in the registry, but they do carry out medical data research. Application of the GDPR implies that personal data may only be used for medical data research after informing patients and obtaining their explicit consent.
Subject(s)
Computer Security/legislation & jurisprudence , Health Records, Personal , Informed Consent/legislation & jurisprudence , Privacy/legislation & jurisprudence , Quality of Health Care/legislation & jurisprudence , Registries , Thoracic Surgery/legislation & jurisprudence , Europe , HumansABSTRACT
The assessment of quality of care is becoming increasingly important in healthcare, both globally and in the Netherlands. The Dutch Minister of Health declared 2015 to be the year of transparency, thereby aiming to improve quality of care by transparent reporting of outcome data. With the increasing importance of transparency, knowledge on quality measurement will be essential for a cardiologist in daily clinical care. To that end, this paper provides a comprehensive overview of the Dutch healthcare structure, quality indicators and the current and future assessment of quality of cardiac care in the Netherlands.
ABSTRACT
Klinefelter syndrome (KS) is a frequent genetic disorder due to one or more supernumerary X chromosomes. KS is associated with an increased risk for venous thromboembolic events like deep venous thrombosis and pulmonary embolism. This paper describes a 37-year-old male patient with KS referred to our tertiary center with chronic thromboembolic pulmonary hypertension, and who was successfully treated by pulmonary endarterectomy.