Search

Machine learning (ML) methods for the analysis of electrocardiography (ECG) data are gaining importance, substantially supported by the release of large public datasets. However, these current datasets miss important derived descriptors such as ECG features that have been devised in the past hundred years and still form the basis of most automatic ECG analysis algorithms and are critical for cardiologists' decision processes. ECG features are available from sophisticated commercial software but are not accessible to the general public. To alleviate this issue, we add ECG features from two leading commercial algorithms and an open-source implementation supplemented by a set of automatic diagnostic statements from a commercial ECG analysis software in preprocessed format. This allows the comparison of ML models trained on clinically versus automatically generated label sets. We provide an extensive technical validation of features and diagnostic statements for ML applications. We believe this release crucially enhances the usability of the PTB-XL dataset as a reference dataset for ML methods in the context of ECG data., (© 2023. The Author(s).)

Recent global developments underscore the prominent role big data have in modern medical science. But privacy issues constitute a prevalent problem for collecting and sharing data between researchers. However, synthetic data generated to represent real data carrying similar information and distribution may alleviate the privacy issue. In this study, we present generative adversarial networks (GANs) capable of generating realistic synthetic DeepFake 10-s 12-lead electrocardiograms (ECGs). We have developed and compared two methods, named WaveGAN* and Pulse2Pulse. We trained the GANs with 7,233 real normal ECGs to produce 121,977 DeepFake normal ECGs. By verifying the ECGs using a commercial ECG interpretation program (MUSE 12SL, GE Healthcare), we demonstrate that the Pulse2Pulse GAN was superior to the WaveGAN* to produce realistic ECGs. ECG intervals and amplitudes were similar between the DeepFake and real ECGs. Although these synthetic ECGs mimic the dataset used for creation, the ECGs are not linked to any individuals and may thus be used freely. The synthetic dataset will be available as open access for researchers at OSF.io and the DeepFake generator available at the Python Package Index (PyPI) for generating synthetic ECGs. In conclusion, we were able to generate realistic synthetic ECGs using generative adversarial neural networks on normal ECGs from two population studies, thereby addressing the relevant privacy issues in medical datasets., (© 2021. The Author(s).)

Loss-of-function (LoF) mutations in KCNQ1, encoding the voltage-gated K + channel K v 7.1, lead to long QT syndrome 1 (LQT1). LQT1 patients also present with post-prandial hyperinsulinemia and hypoglycaemia. In contrast, KCNQ1 polymorphisms are associated with diabetes, and LQTS patients have a higher prevalence of diabetes. We developed a mouse model with a LoF Kcnq1 mutation using CRISPR-Cas9 and hypothesized that this mouse model would display QT prolongation, increased glucose-stimulated insulin secretion and allow for interrogation of K v 7.1 function in islets. Mice were characterized by electrocardiography and oral glucose tolerance tests. Ex vivo, islet glucose-induced insulin release was measured, and beta-cell area quantified by immunohistochemistry. Homozygous mice had QT prolongation. Ex vivo, glucose-stimulated insulin release was increased in islets from homozygous mice at 12-14 weeks, while beta-cell area was reduced. Non-fasting blood glucose levels were decreased at this age. In follow-up studies 8-10 weeks later, beta-cell area was similar in all groups, while glucose-stimulated insulin secretion was now reduced in islets from hetero- and homozygous mice. Non-fasting blood glucose levels had normalized. These data suggest that K v 7.1 dysfunction is involved in a transition from hyper- to hyposecretion of insulin, potentially explaining the association with both hypoglycemia and hyperglycemia in LQT1 patients.

Deep learning-based tools may annotate and interpret medical data more quickly, consistently, and accurately than medical doctors. However, as medical doctors are ultimately responsible for clinical decision-making, any deep learning-based prediction should be accompanied by an explanation that a human can understand. We present an approach called electrocardiogram gradient class activation map (ECGradCAM), which is used to generate attention maps and explain the reasoning behind deep learning-based decision-making in ECG analysis. Attention maps may be used in the clinic to aid diagnosis, discover new medical knowledge, and identify novel features and characteristics of medical tests. In this paper, we showcase how ECGradCAM attention maps can unmask how a novel deep learning model measures both amplitudes and intervals in 12-lead electrocardiograms, and we show an example of how attention maps may be used to develop novel ECG features.

We aimed to investigate the pathogenicity of cardiac ion channel variants previously associated with SIDS. We reviewed SIDS-associated variants previously reported in databases and the literature in three large population-based cohorts; The ExAC database, the Inter99 study, and the UK Biobank (UKBB). Variants were classified according to the American College of Medical Genetics and Genomics (ACMG) guidelines. Of the 92 SIDS-associated variants, 59 (64%) were present in ExAC, 18 (20%) in Inter99, and 24 (26%) in UKBB. Using the Inter99 cohort, we found no difference in J-point amplitude and QTc-interval between carriers and non-carriers for 14/18 variants. There was no difference in the risk of syncope (P = 0.32), malignant ventricular arrhythmia (P = 0.96), and all-cause mortality (P = 0.59) between carriers and non-carriers. The ACMG guidelines reclassified 75% of all variants as variant-of-uncertain significance, likely benign, and benign. We identified ~2/3 of variants previously associated with SIDS and found no significant associations with electrocardiographic traits, syncope, malignant ventricular arrhythmia, or all-cause mortality. These data indicate that many of these variants are not highly penetrant, monogenic causes of SIDS and underline the importance of frequent reappraisal of genetic variants to avoid future misdiagnosis.

Genome-wide association studies (GWAS) of quantitative electrocardiographic (ECG) traits in large consortia have identified more than 130 loci associated with QT interval, QRS duration, PR interval, and heart rate (RR interval). In the current study, we meta-analyzed genome-wide association results from 30,000 mostly Dutch samples on four ECG traits: PR interval, QRS duration, QT interval, and RR interval. SNP genotype data was imputed using the Genome of the Netherlands reference panel encompassing 19 million SNPs, including millions of rare SNPs (minor allele frequency < 5%). In addition to many known loci, we identified seven novel locus-trait associations: KCND3, NR3C1, and PLN for PR interval, KCNE1, SGIP1, and NFKB1 for QT interval, and ATP2A2 for QRS duration, of which six were successfully replicated. At these seven loci, we performed conditional analyses and annotated significant SNPs (in exons and regulatory regions), demonstrating involvement of cardiac-related pathways and regulation of nearby genes.

We describe a Swedish family with the proband and his brother suffering from severe Romano-Ward syndome (RWS) associated with compound heterozygosity for two mutations in the KVLQT1 (also known as KCNQ1 and KCNA9) gene (R518X and A525T). The mutations were found to segregate as heterozygotes in the maternal and the paternal lineage, respectively. None of the heterozygotes exhibited clinical long QT syndrome (LQTS). No hearing defects were found in the proband. The data strongly indicates that the compound heterozygosity for R518X and A525T is the cause of an autosomal recessive form of RWS in this family. Our findings support the implication of a higher frequency of gene carriers than previously expected. We suggest that relatives of 'sporadic RWS' patients should be considered potential carriers, at risk of dying suddenly from drug-induced LQTS.