A Deep Learning-Enabled Electrocardiogram Model for the Identification of a Rare Inherited Arrhythmia: Brugada Syndrome.

BACKGROUND: Brugada syndrome is a major cause of sudden cardiac death in young people and has distinctive electrocardiographic (ECG) features. We aimed to develop a deep learning-enabled ECG model for automatic screening for Brugada syndrome to identify these patients at an early point in time, thus allowing for life-saving therapy. METHODS: A total of 276 ECGs with a type 1 Brugada ECG pattern (276 type 1 Brugada ECGs and another randomly retrieved 276 non-Brugada type ECGs for 1:1 allocation) were extracted from the hospital-based ECG database for a 2-stage analysis with a deep learning model. After trained network for identifying right bundle branch block pattern, we transferred the first-stage learning to the second task to diagnose the type 1 Brugada ECG pattern. The diagnostic performance of the deep learning model was compared with that of board-certified practicing cardiologists. The model was further validated in an independent ECG data set collected from hospitals in Taiwan and Japan. RESULTS: The diagnoses by the deep learning model (area under the receiver operating characteristic curve [AUC] 0.96, sensitivity 88.4%, specificity 89.1%) were highly consistent with the standard diagnoses (kappa coefficient 0.78). However, the diagnoses by the cardiologists were significantly different from the standard diagnoses, with only moderate consistency (kappa coefficient 0.63). In the independent ECG cohort, the deep learning model still reached a satisfactory diagnostic performance (AUC 0.89, sensitivity 86.0%, specificity 90.0%). CONCLUSIONS: We present the first deep learning-enabled ECG model for diagnosing Brugada syndrome, which appears to be a robust screening tool with a diagnostic potential rivalling trained physicians.

Genome-wide association study identifies novel susceptibility loci for migraine in Han Chinese resided in Taiwan.

Background Susceptibility genes for migraine, despite it being a highly prevalent and disabling neurological disorder, have not been analyzed in Asians by genome-wide association study (GWAS). Methods We conducted a two-stage case-control GWAS to identify susceptibility genes for migraine without aura in Han Chinese residing in Taiwan. In the discovery stage, we genotyped 1005 clinic-based Taiwanese migraine patients and 1053 population-based sex-matched controls using Axiom Genome-Wide CHB Array. In the replication stage, we genotyped 27 single-nucleotide polymorphisms with p < 10-4 in 1120 clinic-based migraine patients and 604 sex-matched normal controls by using Sequenom. Variants at LRP1, TRPM8, and PRDM, which have been replicated in Caucasians, were also genotyped. Results We identified a novel susceptibility locus (rs655484 in DLG2) that reached GWAS significance level for migraine risk in Han Chinese ( p = 1.45 × 10-12, odds ratio [OR] = 2.42), and also another locus (rs3781545in GFRA1) with suggestive significance ( p = 1.27 × 10-7, OR = 1.38). In addition, we observed positive association signals with a similar trend to the associations identified in Caucasian GWASs for rs10166942 in TRPM8 (OR = 1.33, 95% confidence interval [CI] = 1.14-1.54, Ppermutation = 9.99 × 10-5; risk allele: T) and rs1172113 in LRP1 (OR = 1.23, 95% CI = 1.04-1.45, Ppermutation = 2.9 × 10-2; risk allele: T). Conclusion The present study is the first migraine GWAS conducted in Han-Chinese and Asians. The newly identified susceptibility genes have potential implications in migraine pathogenesis. DLG2 is involved in glutamatergic neurotransmission, and GFRA1 encodes GDNF receptors that are abundant in CGRP-containing trigeminal neurons. Furthermore, positive association signals for TRPM8 and LRP1 suggest the possibility for common genetic contributions across ethnicities.

Evaluation of RGS4 as a candidate gene for schizophrenia.

Several studies have suggested that the regulator of G-protein signaling 4 (RGS4) may be a positional and functional candidate gene for schizophrenia. Three single nucleotide polymorphisms (SNP) located at the promoter region (SNP4 and SNP7) and the intron 1 (SNP18) of RGS4 have been verified in different ethnic groups. Positive results have been reported in these SNPs with different numbers of SNP combinatory haplotypes. In this study, these three SNP markers were genotyped in 218 schizophrenia pedigrees of Taiwan (864 individuals) for association analysis. Among these three SNPs, neither SNP4, SNP7, SNP18 has shown significant association with schizophrenia in single locus association analysis, nor any compositions of the three SNP haplotypes has shown significantly associations with the DSM-IV diagnosed schizophrenia. Our results fail to support the RGS4 as a candidate gene for schizophrenia when evaluated from these three SNP markers.