ライフサイエンスコーパス: ECG

1 ECG biomarkers and vulnerability window for reentry were

2 ECG criteria for LBBB incompletely predicted CCB, and in

3 ECG imaging (ECGI) has been used to guide treatment of v

4 ECG measurements derived from the convolutional neural n

5 ECG measures of LV mass are estimated as heritable in tw

6 ECG recordings documented the onset of AF.

7 ECG-based detection of HCM by an artificial intelligence

8 ECG-based traits such as LV mass and Sokolow-Lyon durati

9 ECG-gated computed tomography scans were acquired at end

10 rmative standards were developed for all 102 ECG variables including heart rate; P, R, and T axis; R-

11 Data from 9536, 1301, and 1768 ECGs of adult patients who underwent coronary angiograph

12 We identified 36 186 ECGs from the University of California, San Francisco da

13 526 patients in the training dataset, 64 340 ECGs from 18 116 patients in the internal validation dat

14 A total of 412,461 ECGs were used to develop a variational autoencoder (VAE

15 Of these, 752,517 ECG samples were selected (1256 +/- 244 per subject) to

16 rmal sinus rhythm ECGs for analysis: 454 789 ECGs recorded from 126 526 patients in the training data

17 the internal validation dataset, and 130 802 ECGs from 36 280 patients in the testing dataset.

18 ally, PTT is determined by (1) measuring (a) ECG and ear, finger, or toe PPG waveforms or (b) two of

19 Male sex and an abnormal ECG are associated with a higher risk of developing HCM.

20 ry triggers for MET activation were abnormal ECG alerts, specifically asystole (n = 5), and pulseless

21 Electronically acquired ECG values based on the largest pediatric and young adul

22 icted groups (dead versus alive 1 year after ECG) over a 25-year follow-up period.

23 prognostic value and interaction between AI-ECG AF model output and CHARGE-AF score.

24 When included in the same model, both AI-ECG AF model output (hazard ratio, 1.76 per SD after log

25 and 0.72 (95% CI, 0.69-0.75) for combined AI-ECG and CHARGE-AF score.

26 del output, CHARGE-AF score, and combined AI-ECG and CHARGE-AF score.

27 value of AI-enabled electrocardiography (AI-ECG) as a predictor of future AF and assess its performa

28 tistics were 0.69 (95% CI, 0.66-0.72) for AI-ECG AF model output, 0.69 (95% CI, 0.66-0.71) for CHARGE

29 C statistics were calculated for AI-ECG AF model output, CHARGE-AF score, and combined AI-EC

30 In the present study, both the AI-ECG AF model output and CHARGE-AF score independently pr

31 The AI-ECG may offer a means to assess risk with a single test

32 We calculated the probability of AF using AI-ECG, among participants in the population-based Mayo Cli

33 Participants with AI-ECG AF model output of >0.5 at the baseline visit had cu

34 onitor-detected AF using a 14-day ambulatory ECG monitor was similar in the 4 race/ethnic groups: 7.1

35 d in an ancillary study involving ambulatory ECG monitoring and had follow-up for clinically detected

36 olving both a resting and 12-lead ambulatory ECG, an exercise stress test, and genetic screening.

37 participants who wore a leadless, ambulatory ECG monitor (Zio XT Patch) for up to 2 weeks were aged 7

38 Long-term ambulatory ECG monitors provide the opportunity for unbiased AF det

39 The T-wave morphology restitution (TMR), an ECG marker quantifying ventricular repolarization dynami

40 highly dependent on the duration of analysed ECG fragments and the heart rate.

41 orks have been used to automatically analyze ECG tracings and outperform physicians in detecting cert

42 We analyzed ECG recordings from 55 222 individuals from a UK middle-

43 performed using easily obtained clinical and ECG features.

44 ritish families (427 individuals) by CMR and ECG, and undertook heritability analyses using variance-

45 were screened from nonselective imaging and ECG databases.

46 ge, are associated with early structural and ECG changes.

47 hemia determine arrhythmia vulnerability and ECG alterations.

48 ized data of heart rate, activity level, and ECGs from 7500 AliveCor users.

49 e bottom y axis was incorrectly labelled as 'ECG (muV)'; the correct label is 'ECG (mV)'.

50 bone fracture underwent clinical assessment, ECG, and contrast-enhanced cardiovascular magnetic reson

51 -disease correlations and improved automated ECG readings in high-volume cardiac screening efforts in

52 ve the accuracy and scalability of automated ECG analysis.

53 a development of algorithms for an automatic ECG selection.

54 approach can dramatically benefit automatic ECG data analysis by providing cardiologist level accura

55 Breathing, ECG and microvascular blood flow were simultaneously mon

56 y controls (2 markedly and 3 marginally) but ECGs were otherwise normal.

57 tion (AF), which is challenging to detect by ECG analysis when in paroxysmal form.

58 atients with left ventricular hypertrophy by ECG criteria and 0.95 (95% CI: 0.90 to 1.00) among patie

59 ithout changes in heart rate, as measured by ECG and ex vivo optical mapping.

60 measurement, being greater when measured by ECG than CMR.

61 2) has been associated with a characteristic ECG pattern of short-long RR intervals.

62 ongenetic causes with certain characteristic ECG features preceding polymorphic ventricular tachyarrh

63 ecular causes result in these characteristic ECG patterns and how these patterns are mechanistically

64 ed in the supine position at the Mayo Clinic ECG laboratory between Dec 31, 1993, and July 21, 2017,

65 ng the different genotype-dependent clinical ECG features.

66 We evaluated measures from prior clinical ECG and echocardiograms, manually over-read to evaluate

67 d left precordial leads were the most common ECG abnormalities.

68 reduce the rate of misdiagnosed computerized ECG interpretations and improve the efficiency of expert

69 d vasovagal syncope documented by continuous ECG and video EEG monitoring.

70 (Atrial Fibrillation Detected by Continuous ECG Monitoring Using Implantable Loop Recorder to Preven

71 (Atrial Fibrillation Detected by Continuous ECG Monitoring Using Implantable Loop Recorder to Preven

72 ng implantable loop recorders for continuous ECG monitoring post-AF ablation show that VLR occurs in

73 stic yield of 4 versus 2 weeks of continuous ECG monitoring.

74 complex, and QT interval from the continuous ECG waveform using features extraction logic, then the P

75 Widely available digital ECG data and the algorithmic paradigm of deep learning(2

76 According to CineECG algorithm, each ECG was classified as Normal, Brugada, RBBB, or Undeterm

77 t coronary angiography within 24 h from each ECG were used for development, internal and external val

78 Such an ultra-early, ECG-based clinical decision support tool, when combined

79 function were evaluated by echocardiography, ECG, and in Langendorff-perfused hearts.

80 Electrocardiogram (ECG) acquisition is increasingly widespread in medical a

81 .82 to 4.65) and abnormal electrocardiogram (ECG) (HR: 4.02; 95% CI: 2.51 to 6.44); TNNI3 variants ha

82 ing 500 time points of an electrocardiogram (ECG) trace in a genome-wide association study (GWAS), Ve

83 Here, we used an electrocardiogram (ECG)-depth recorder tag to measure blue whale heart rate

84 art-Phone (Android) based electrocardiogram (ECG) acquisition and monitoring system (cvrPhone), and a

85 subjects failed to deploy electrocardiogram (ECG)-based hypoglycemic detection systems reliably.

86 determine if they exhibit electrocardiogram (ECG) abnormalities involving cardiac conduction that are

87 thod, obtained by inverse electrocardiogram (ECG) from standard 12-lead ECG, to localize the electric

88 The 12-lead electrocardiogram (ECG) is readily available during initial patient evaluat

89 ay be detected by 12-lead electrocardiogram (ECG), single-lead monitors (such as ambulatory blood pre

90 Resting electrocardiogram (ECG) of 5-min was collected prior to surgical treatments

91 We used standard 10-s electrocardiogram (ECG) recordings of 60 subjects from the Honolulu Asia Ag

92 On standard electrocardiogram (ECG) PQ interval is known to be moderately heart rate de

93 The electrocardiogram (ECG) is a widely used medical test, consisting of voltag

94 intelligence (AI) to the electrocardiogram (ECG), a routine method of measuring the heart's electric

95 terpretation accuracy of electrocardiograms (ECGs).

96 ntelligence (AI)-enabled electrocardiograph (ECG) using a convolutional neural network to detect the

97 The analysis of equine electrocardiographic (ECG) recordings is complicated by the absence of agreed

98 Electrocardiography (ECG) provides gold standard HRV measurements but is inco

99 it was initiated and an electrocardiography (ECG) patch was mailed to the participant, to be worn for

100 tion using clinical and electrocardiography (ECG) variables as a first step in the detection of LV di

101 of HCM based on 12-lead electrocardiography (ECG).

102 l infarction (MI) using electrocardiography (ECG) is the cornerstone of effective treatment and preve

103 a validated artificial intelligence-enabled ECG algorithm for the identification of LVSD (defined as

104 The artificial intelligence-enabled ECG algorithm identified LVSD with an area under the rec

105 curacy of an artificial intelligence-enabled ECG to identify patients presenting with dyspnea who hav

106 he state-of-art approaches, where the entire ECG beats have been considered.

107 Forty-five ECGs were available for corrected QT interval (QTc) meas

108 of an end-to-end deep learning approach for ECG analysis across a wide variety of diagnostic classes

109 construct smoothed adversarial examples for ECG tracings that are invisible to human expert evaluati

110 ve not used an implantable loop recorder for ECG monitoring.

111 ical event, 1-year all-cause mortality, from ECG voltage-time traces.

112 Median time from ECG to echocardiogram was 1 day (Q1: 1, Q3: 2).

113 ds were implemented to extract features from ECGs including simple heart rate variability (HRV) metri

114 -3-gallate (GCG), and epicatechin-3-gallate (ECG)) and not the catechol-type catechins (catechin and

115 the order of minutes), epicatechin gallate (ECG) and epigallocatechin gallate (ECGC) exhibited very

116 and gallated catechins the most potent: CG > ECG > EGCG >= GCG when compared to the non-gallated cate

117 Historical ECG interpretative concepts were reassessed and new conc

118 6, 9, and 12 months using continuous 24-hour ECG monitors.

119 However, ECG LV hypertrophy traits are likely to be influenced by

120 s and improve the efficiency of expert human ECG interpretation by accurately triaging or prioritizin

121 e coumadin ridge and 2 vascular landmarks in ECG-gated computed tomography provides a viable method o

122 nresponse rates, highlighting limitations in ECG selection criteria: QRS duration (QRSd) >=150 ms and

123 Titchener T-complexity on the heart rate in ECG strips obtained at low heart rates (25-60 bpm) and p

124 Data included ECG recordings from all four studies, controlled exposur

125 To understand how clinicians interpret ECGs, and potentially other medical images, visual trans

126 belled as 'ECG (muV)'; the correct label is 'ECG (mV)'.

127 etected not only with a conventional 12 lead ECG but also with a life type 6-lead ECG device that emp

128 al sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the Mayo Clinic E

129 ed if they had at least one standard 12-lead ECG acquired on the date of the ED visit and an echocard

130 ith higher accuracy than a validated 12-lead ECG algorithm (83.3% versus 38.9%; P=0.015).

131 will summarize all of the published 12-lead ECG algorithms used to guide localization of OT ventricu

132 s) at 12 months of follow-up using a 12-lead ECG and Holter monitoring at 3, 6, and 12 months.

133 e history, physical examination, and 12-lead ECG are each critical to the diagnosis and evaluation of

134 trained and validated using digital 12-lead ECG from 2,448 patients with a verified HCM diagnosis an

135 females) in whom drug-free day-time 12-lead ECG Holters were available.

136 The surface 12-lead ECG is routinely used to localize the anatomic site of o

137 Continuous 12-lead ECG of 5 episodes of long-lasting VF occurring in 3 pati

138 DLA successfully detected MI using a 12-lead ECG or a 6-lead ECG.

139 age and self-reported sex using only 12-lead ECG signals.

140 eening pathways using the ubiquitous 12-lead ECG to detect asymptomatic paroxysmal AF in at-risk popu

141 hine learning methods can extend the 12-lead ECG to quantitative applications well beyond its current

142 tricular ejection fraction using the 12-lead ECG varies by race/ethnicity and to (2) determine whethe

143 ar activation in real time using the 12-lead ECG was developed.

144 12-lead ECG was performed at admission, after 7 and 14 days; QTc

145 Compared with standard 12-lead ECG, CineECG at baseline had a 100% positive predictive

146 asive CineECG, based on the standard 12-lead ECG, opens new prospective for diagnosing patients with

147 lectrocardiogram (ECG) from standard 12-lead ECG, to localize the electrical activity pathway in pati

148 554 temporal-spatial features of the 12-lead ECG, we train and test multiple classifiers on two indep

149 lar ejection fraction <=35% from the 12-lead ECG.

150 ciations with maximum PWD across the 12-lead ECG.

151 mporospatial isochrone from standard 12-lead ECG.

152 uration confirmed by rhythm strip or 12-lead ECG.

153 MI and is difficulty to apply to limb 6-lead ECG based life type or wearable devices.

154 12 lead ECG but also with a life type 6-lead ECG device that employs our DLA.

155 r (VAE) that reconstructed precordial 6-lead ECG using limb 6-lead ECG.

156 ic curves of the DLA with VAE using a 6-lead ECG were 0.880 and 0.854, respectively, and the performa

157 lgorithm (DLA) for detecting MI using 6-lead ECG.

158 cted precordial 6-lead ECG using limb 6-lead ECG.

159 detected MI using a 12-lead ECG or a 6-lead ECG.

160 palpate, record, and interpret a single-lead ECG (SLECG).

161 nvolving a total 59 482 one-min, single-lead ECG segments recorded from 156 subjects.

162 el for arrhythmia detection from single-lead ECG(6) is vulnerable to this type of attack.

163 Ten methods to compute single-lead ECG-derived respiration (EDR) were compared under multip

164 /phenotyping) and data fidelity (single-lead ECG/22-channel EEG).

165 pertly labeled dataset consisting of 12-lead ECGs from 40,258 patients with four arrhythmia classes:

166 Additionally, 12-lead ECGs were collected at diagnosis, before initiation of m

167 nus rhythm using standard 10-second, 12-lead ECGs.

168 nge of distinct arrhythmias from single-lead ECGs with high diagnostic performance similar to that of

169 The susceptibility of deep learning ECG algorithms to adversarial misclassification implies

170 Models trained using patient-level ECG profiles enabled surprising quantitative estimates o

171 wireless physiological data using Lifetouch (ECG-derived heart and respiratory rate) and WristOx2 (pu

172 For each measured ECG sample, 5-minute history of preceding cardiac cycles

173 nostic tasks involving different modalities (ECG/EEG/EHR), required level of characterization (abnorm

174 Multiple ECG algorithms have been developed to assist preprocedur

175 100 randomly selected patients with multiple ECGs over the course of decades were identified to asses

176 raged a unique opportunity in which neonatal ECGs and hydroxychloroquine blood levels were available

177 Of them, 111 had a normal ECG, different cardiac diseases, or were lost to follow-

178 : 0.90 to 1.00) among patients with a normal ECG.

179 the discriminating features of these normal ECGs were not apparent to expert reviewers.

180 Normative ECG values for children are based on relatively few subj

181 ve was to create the foundation of normative ECG standards in the young utilizing Z-scores.

182 titative foundation of traditional and novel ECG variables.

183 T-complexity and Lempel-Ziv '78 analysis of ECG recordings of healthy Thoroughbred horses are highly

184 ore quantitative, reproducible assessment of ECG variables.

185 r defibrillator therapies or on the basis of ECG-documented arrhythmia.

186 iagnosed as affected based on combination of ECG abnormalities with positive genotyping (QTc, 482+/-3

187 An effective method for compression of ECG signals, which falls within the transform lossy comp

188 simplicity is crucial for the development of ECG-based ambulatory systems.

189 can lead to more confident establishment of ECG-disease correlations and improved automated ECG read

190 emia alarming system using short excerpts of ECG signal.

191 Expression of ECG variables by Z-scores lends an objective and reprodu

192 ind approach where the localized features of ECG are being taken into considerations unlike the state

193 ask Force reviewed the benefits and harms of ECG screening for AFib in adults aged 65 years or older

194 Unsupervised machine learning of ECG waveforms identified CRT subgroups with relevance be

195 We explored unsupervised machine learning of ECG waveforms to identify CRT subgroups that may differe

196 G, the simultaneous noninvasive recording of ECG and skin sympathetic nerve activity (SKNA), directly

197 AF was detected by 4 weeks versus 2 weeks of ECG monitoring.

198 hy in patients with ST-segment elevations on ECG after OHCA, while the role of coronary angiography i

199 uld be taken when evaluating these models on ECGs that may have been altered, particularly when incen

200 oes affect accuracy of interpretation in one ECG.

201 is that included 97 829 patients with paired ECGs and echocardiograms.

202 For further evaluation, arterial phase ECG-synchronized CT angiography from the skull base to t

203 For further evaluation, arterial phase ECG-synchronized CT angiography from the skull base to t

204 n clinical COVID-19 studies, early postnatal ECGs should be considered.

205 segment elevation on their postresuscitation ECG, the prevalence of coronary artery disease has been

206 segment elevation on their postresuscitation ECG, the prevalence of coronary artery disease has been

207 land Clinic, 155 underwent both preoperative ECG-gated contrast-enhanced CT and TEE.

208 Additionally, we sought to classify prior ECG characteristics in male and female subjects with SCD

209 record, nor significant differences in prior ECG or echocardiogram findings compared with matched con

210 th SCD were more likely to have normal prior ECG tracings (22.2% versus 15.3% in men, P<0.001).

211 models were developed using signal-processed ECG, traditional ECG, and clinical features and were tes

212 nd-tidal CO(2) (capnography) and heart rate (ECG) were performed for 5 min at rest (normocapnia) and

213 l hypoglycemia using a few heartbeats of raw ECG signal recorded with non-invasive, wearable devices,

214 Previously recorded ECGs were available and analyzed in 1101 subjects (18.8%

215 ealthy participants, together with reference ECG and arterial finger PPG signals for validation.

216 rrence of AF was based on physician-reported ECG-verified events.

217 tor analysis identified three representative ECG parameters: standard deviation of NN-intervals (SDNN

218 trained a DNN with 1,169,662 12-lead resting ECGs obtained from 253,397 patients, in which 99,371 eve

219 ion to the interpretation of 12-lead resting ECGs, even in cases that are interpreted as normal by ph

220 Among the 450 participants who returned ECG patches containing data that could be analyzed - whi

221 fibrillation before the normal sinus rhythm ECG tested by the model.

222 22 patients with 649 931 normal sinus rhythm ECGs for analysis: 454 789 ECGs recorded from 126 526 pa

223 free 60-second strips of normal sinus-rhythm ECGs were converted to binary strings using threshold cr

224 ication of horses with PAF from sinus-rhythm ECGs with high accuracy.

225 y analysis of apparently normal sinus-rhythm ECGs.

226 rate variability, derived from the patient's ECG.

227 In these data, a single 10-second ECG yielded a sensitivity (and negative predictive value

228 sing to 8.3% (67%) for twice-daily 30-second ECGs during 14 days and to 11% (68%), 13% (68%), 15% (69

229 inically significant changes in vital signs, ECGs, or clinical chemistry laboratory values, including

230 Specific ECG characteristics including early (consistent TZ) and

231 In this retrospective study ECG recordings were obtained during routine clinical wor

232 e, 34% had atrial fibrillation on subsequent ECG patch readings and 84% of notifications were concord

233 ogical function by echocardiography, surface ECG and conscious telemetry, intracardiac electrograms a

234 The f-waves in surface ECG leads V(1), aVL, and II, which reflect well electric

235 he accuracy of ECGI versus a 12-lead surface ECG algorithm was compared.

236 We sought to study surface ECG waveforms and effect of ablation in long-lasting VF

237 e value 100%, sensitivity 100%) than surface ECG criteria for correction with His bundle pacing.

238 s in atrial fibrillation (AF) in the surface ECG by quantifying the amplitude, irregularity, and domi

239 haracteristics of the f-waves on the surface ECG could discriminate LPeAF from other types of AF.

240 heters and simultaneous 57-lead body surface ECGs.

241 assification scheme based on 12-lead surface ECGs that attains the accuracy performance level of prof

242 We obtained surface ECGs and analyzed arrhythmia susceptibility; we observed

243 We analysed, by telemetric ECG recording, whether pharmacological inhibition of the

244 -19 who were receiving continuous telemetric ECG monitoring and had a definitive disposition of hospi

245 or older and found inadequate evidence that ECG identifies AFib more effectively than usual care.

246 The ECG is an inexpensive, ubiquitous, painless test which c

247 The ECG remains the most widely used diagnostic test for cha

248 x and age have long been known to affect the ECG.

249 tions at two granularity levels: between the ECG leads, and between smaller grid-cells, whose size wa

250 were detected as the time delays between the ECG R-wave and ear PPG foot, R-wave and finger PPG foot

251 The proposed methodology first extracts the ECG localized features including PR interval, QRS comple

252 espiratory information were derived from the ECG and compared to those extracted from a reference res

253 e to derive respiratory information from the ECG signal.

254 and the clinical inferences derived from the ECG while at the same time preserving interpretability f

255 left ventricular ejection fraction from the ECG.

256 However, the ECG LV mass was positively correlated with the lateral d

257 ion population or by racial variation in the ECG.

258 nd practically complete normalization of the ECG abnormalities (their QTc shortened from 492+/-37 to

259 in a higher frequency range than that of the ECG and myopotential, it is possible to use high-pass or

260 ng clinicians to visualise which part of the ECG signal (e.g., T-wave, ST-interval) is significantly

261 abled the temporospatial localization of the ECG waveforms, deriving the mean temporospatial isochron

262 by exploiting the localized features of the ECG.

263 f marked repolarization abnormalities on the ECG performed during the mandatory medical visit necessa

264 92) for observing atrial fibrillation on the ECG simultaneously with a subsequent irregular pulse not

265 74) for observing atrial fibrillation on the ECG simultaneously with a subsequent irregular tachogram

266 te to sex and age-related differences on the ECG.

267 ECG-a ubiquitous, low-cost test-permits the ECG to serve as a powerful screening tool in asymptomati

268 We segmented the ECG into standard component waveforms and intervals usin

269 Applying artificial intelligence to the ECG allows prediction of patient sex and estimation of a

270 Application of AI to the ECG-a ubiquitous, low-cost test-permits the ECG to serve

271 are genotype-negative, >40% normalize their ECG after detraining, but the abnormalities tend to recu

272 estimations of HRV are highly correlated to ECG HRV for both time and frequency domain parameters an

273 versarial examples do not extend directly to ECG signals, as such methods introduce square-wave artef

274 In this research, we compare SPG and PPG to ECG for estimation of HRV during an orthostatic challeng

275 Machine learning applied to ECGs may help identify subjects at high risk of having p

276 oped using signal-processed ECG, traditional ECG, and clinical features and were tested using the tes

277 en EVM-based heart tracking and ground truth ECG, from both color (RGB) and infrared (IR) videos, in

278 One hundred two ECG variables were collected from a retrospective cohort

279 sed as affected by LQTS based on unequivocal ECG abnormalities (QTc, 472+/-33 ms).

280 levation myocardial infarction network using ECG assessment of reperfusion coupled with clinical outc

281 By using ECGs collected over a 34-year period in a large regional

282 ex anatomy of the OT limits predictive value ECG criteria alone for localization for these arrhythmia

283 curacy and robust compatibility with various ECG data sources.

284 A direct correlation exists between PM VAs ECG inconsistencies and type-specific PMCs.

285 Five-second VF ECG segments were collected with and without chest compr

286 easures of the ventricular fibrillation (VF) ECG waveform can assess myocardial physiology and predic

287 AF was ascertained using study visit ECGs and hospital records.

288 In vivo ECG recording and whole heart optical mapping of intact

289 a viable alternative for HRV estimation when ECG measurements are impractical.

290 Our study demonstrates that while ECG characteristics vary by race, this did not impact th

291 Both are associated with ECG and mechanical changes and clinical worsening over 1

292 nd are enriched for variants associated with ECG measures and atrial fibrillation.

293 EGM transitions correlated with ECG transitions during threshold testing and can help ac

294 ic PMCs were more prevalent in patients with ECG inconsistencies.

295 Similar results with ECG and EGC, but not in line with FQ kinetics, highlight

296 In a systematic review, screening with ECG identified more new cases of AFib than no screening.

297 onnections of the PM and correlate them with ECG inconsistencies and ablation results.

298 iuretic peptide) <100 pg/mL), and those with ECG-LVH and abnormal levels of either biomarker (maligna

299 3 groups: those without ECG-LVH, those with ECG-LVH and normal biomarkers (hs-cTnT (high sensitivity

300 e large subset of patients (n = 45,285) with ECGs interpreted as 'normal' by a physician, the perform

301 were classified into 3 groups: those without ECG-LVH, those with ECG-LVH and normal biomarkers (hs-cT