An Electrocardiographic System With Anthropometrics via Machine Learning to Screen Left Ventricular Hypertrophy among Young Adults

Nocturnal sentry duty and cardiometabolic characteristics in armed forces personnel

BACKGROUND

Sleep deprivation can lead to increased body weight and blood pressure (BP), but the latent effects of partial sleep deprivation related to required night sentry duties within a short-term period on cardiometabolic characteristic changes in military personnel are unclear.

AIM

To investigate the association between night sentry duty frequency in the past 3 months and cardiometabolic characteristics in armed forces personnel.

METHODS

A total of 867 armed forces personnel who were aged 18-39 years and did not take any antihypertensive medications in Taiwan in 2020 were included. The frequency of night sentry duty was self-reported via a questionnaire (average number of night sentry shifts per month for the past 3 months). Hemodynamic status was assessed via the resting BP and pulse rate (PR). Cardiometabolic risk factors were defined according to the International Diabetes Federation criteria. Multivariable linear regression analyses of the associations between night sentry duties and PR, BP, and other metabolic syndrome (MetS) marker levels were performed, with adjustments for age, sex, substance use, body mass index and aerobic fitness. Multiple logistic regression analysis was carried out to determine the associations between night sentry duties and the prevalence of each MetS feature.

RESULTS

There was an association between night sentry duties and PR [standardized β (standard error) = 0.505 (0.223), P =0.02], whereas there was no association with systolic and diastolic BP. In addition, there was an inverse association between night sentry duties and high-density lipoprotein cholesterol (HDL-C) levels [standardized β = -0.490 (0.213), P = 0.02], whereas there was no association with the other metabolic marker levels. Compared with personnel without night sentry duties, those with ≥ 1 night sentry shift/month had a greater risk of impaired fasting glucose (≥ 100 mg/dL) [odds ratio: 1.415 (confidence interval: 1.016-1.969)], whereas no associations with other MetS features were found.

CONCLUSION

Among military personnel, the burden of night sentry duty was positively associated with the resting PR but inversely associated with HDL-C levels. In addition, personnel with partial sleep deprivation may have a greater risk of impaired fasting glucose than those without partial sleep deprivation.

ECG-Based Multiclass Arrhythmia Classification Using Beat-Level Fusion Network

Cardiovascular disease (CVD) is one of the most severe diseases threatening human life. Electrocardiogram (ECG) is an effective way to detect CVD. In recent years, many methods have been proposed to detect arrhythmia using 12-lead ECG. In particular, deep learning methods have been proven to be effective and have been widely used. The attention mechanism has attracted extensive attention in many fields in a series of deep learning methods. Off-the-shelf solutions based on deep learning and attention mechanism for ECG classification mostly give weights to time points. None of the existing methods were considered using the attention mechanism dealing with ECG signals at the level of heartbeats. In this paper, we propose a beat-level fusion net (BLF-Net) for multiclass arrhythmia classification by assigning weights at the heartbeat level, according to the contribution of the heartbeat to diagnostic results. This algorithm consists of three steps: (1) segmenting the long ECG signal into short beats; (2) using a neural network to extract features from heartbeats; and (3) assigning weights to features extracted from heartbeats using an attention mechanism. We test our algorithm on the PTB-XL database and have superiority over state-of-the-art performance on six classification tasks. Besides, the principle of this architecture is clarified by visualizing the weight of the attention mechanism. The proposed BLF-Net is shown to be useful and automatically provides an effective network structure for arrhythmia classification, which is capable of aiding cardiologists in arrhythmia diagnosis.

Asian Female Athlete's Heart: The CHIEF Heart Study

BACKGROUND

The cardiac characteristics of Asian female endurance athletes and strength athletes have rarely been investigated.

METHODS

This study included 177 Taiwanese young women undergoing military training. Cardiac features were assessed by electrocardiography (ECG) and echocardiography. Then, all participants completed a 3000-meter run to assess endurance capacity, and 89 participants completed a 2-minute push-up test to assess muscular strength. Athletes were those whose exercise performance fell one standard deviation above the mean, and the remaining participants were defined as controls. Multiple logistic regression analysis was used to determine the predictors of the cardiac characteristics of female athletes.

RESULTS

Compared to the female controls, female endurance athletes had a greater QRS duration (ms) (92.12 ± 10.35 vs. 87.26 ± 9.89, p = 0.01) and a higher prevalence of right axis deviation (RAD) (34.9% vs. 11.1%, p < 0.001). There were no differences in any echocardiographic parameters. Greater QRS duration and RAD and lower systolic blood pressure were independent predictors of female endurance athletes [odds ratios (OR) and 95% confidence intervals: 1.05 (1.01-1.09), 2.91 (1.12-7.59) and 0.93 (0.88-0.98), respectively]. Female strength athletes had a greater right ventricular outflow tract (RVOT) (mm) (28.06 ± 3.57 vs. 25.38 ± 3.61, p = 0.007) but revealed no differences in ECG variables. Greater RVOT was the only predictor of female strength athletes [OR: 1.26 (1.05-1.50)].

CONCLUSIONS

In Asian military women, a wider QRS duration and the presence of RAD in ECG rather than heart structure and function were found to characterize endurance athletes, whereas a wider RVOT but no ECG features were found to characterize strength athletes.

Mitral valve prolapse and physical performance in Asian military males: The CHIEF Heart study

The impact of mitral valve prolapse (MVP) and mitral regurgitation (MR) on physical performance has not been examined. Of 1,808 physically fit Asian military males, we compared the physical fitness between 62 subjects with MVP (MVP(+)) and 1,311 age- and anthropometrics-matched controls from the 1,746 participants without MVP (MVP(-)). MVP and MR grade were defined based on the American Society of Echocardiography criteria. Aerobic endurance capacity was evaluated by a 3000-m run and muscular endurance capacity was separately evaluated by 2-min sit-ups and 2-min push-ups. Analysis of covariance was used to determine the difference between groups. As compared to the MVP(-), the MVP(+) completed the 3000-m run test faster (839.2 ± 65.3 sec vs. 866.6 ± 86.8 sec, p = 0.019), but did fewer push-ups (41.3 ± 3.92 vs. 48.0 ± 10.1, p = 0.02) and similar sit-ups within 2 min. Of the MVP(+), those with any MR (trivial, mild or moderate) completed the 3000-m run test faster than those without MR (830.6 ± 61.7 sec vs. 877.2 ± 61.7 sec, p = 0.02). Our findings suggest that in physically active Asian military males, the MVP(+) may have greater aerobic endurance capacity but lower muscular endurance capacity than the MVP(-). The presence of MR may play a role for the MVP(+) to have greater aerobic endurance capacity.

Hepatic and Systemic Inflammation for Left Ventricular Mass in Physically Fit Adults: CHIEF Heart Study

BACKGROUND

Both low-grade systemic and hepatic inflammation could result in increased left ventricular mass (LVM) in the general population. However, the associations, which might be modified by exercise, have not been clarified in physically active young adults.

METHODS

The study included 2,004 military males aged 18-43 years in eastern Taiwan. Systemic and hepatic inflammation was defined by the upper tertiles of blood white blood cell (WBC) counts (7.51-11.00 x 103/μL) and serum alanine aminotransferase (ALT: 30-120 U/L), respectively. LVM indexed for the body height ≥49 g/m2.7 was defined as left ventricular hypertrophy (LVH) based on echocardiography. Multiple logistic regression analysis adjusting for age, smoking, alcohol intake, physical fitness, and metabolic syndrome was utilized to determine the associations.

RESULTS

As compared to the lower WBC/lower ALT group, there tended to have an increased risk of LVH with the higher WBC/lower ALT group, the lower WBC/higher ALT group, and the higher WBC/higher ALT group [odds ratios: 0.89 (95% confidence intervals (CI): 0.41-1.94), 1.90 (95% CI: 0.86-4.22) and 2.48 (95% CI: 1.04-5.92); p-value for trend = 0.01].

CONCLUSION

Our study suggested that in physically active males, those with hepatic inflammation rather than low-grade systemic inflammation had a higher risk of LVH. Hepatic injury might be relevant to LVH as an early sign of end-organ damage regardless of physical fitness in young adults.

Muscular Strength and Carotid Intima-Media Thickness in Physically Fit Young Adults: The CHIEF Atherosclerosis Study

Background: Greater muscular strength (MusS) has been found to have an inverse association with subclinical atherosclerosis in children, as well as with mortality from cardiovascular diseases (CVDs) in middle-aged and elderly individuals. However, the association of the degree of MusS with atherosclerosis may differ by sex and has not been clarified in young adults. Methods and Results: A total of 1021 Taiwanese military personnel, aged 18−40 years, participated in annual health examinations in 2018−2020. MusS was separately assessed by 2-min push-up and 2-min sit-up numbers. Subclinical atherosclerosis was measured by the left carotid bulb intima−media thickness (cIMT) using high-resolution ultrasonography. Multiple linear regression with adjustments for age, sex, alcohol intake, cigarette smoking, anthropometric indices, blood pressure, and lipid profiles was utilized to determine the correlation between MusS and cIMT. Both 2-min push-up and 2-min sit-up numbers were inversely correlated with cIMT (standardized β: −0.089 and −0.072, respectively; both p-values < 0.05). In men, both 2-min push-up and 2-min sit-up numbers were inversely correlated with cIMT (standardized β: −0.076 and −0.086, respectively; both p-values < 0.05), while in women, 2-min push-up numbers but not 2-min sit-up numbers were inversely correlated with cIMT (standardized β: −0.204 and −0.01; p = 0.03 and 0.99, respectively). Conclusions: Among young adults, there was an inverse association between MusS and cIMT, emphasizing the beneficial impact of MusS on the regression of atherosclerosis. The study also revealed a sex difference and suggested that training of the upper arm muscles may be an effective preventive measure for young women to reduce the risk of early cardiovascular diseases.

Deep learning assessment of left ventricular hypertrophy based on electrocardiogram

Background

Current electrocardiogram (ECG) criteria of left ventricular hypertrophy (LVH) have low sensitivity. Deep learning (DL) techniques have been widely used to detect cardiac diseases due to its ability of automatic feature extraction of ECG. However, DL was rarely applied in LVH diagnosis. Our study aimed to construct a DL model for rapid and effective detection of LVH using 12-lead ECG.

Methods

We built a DL model based on convolutional neural network-long short-term memory (CNN-LSTM) to detect LVH using 12-lead ECG. The echocardiogram and ECG of 1,863 patients obtained within 1 week after hospital admission were analyzed. Patients were evenly allocated into 3 sets at 3:1:1 ratio: the training set (n = 1,120), the validation set (n = 371) and the test set 1 (n = 372). In addition, we recruited 453 hospitalized patients into the internal test set 2. Different DL model of each subgroup was developed according to gender and relative wall thickness (RWT).

Results

The LVH was predicted by the CNN-LSTM model with an area under the curve (AUC) of 0.62 (sensitivity 68%, specificity 57%) in the test set 1, which outperformed Cornell voltage criteria (AUC: 0.57, sensitivity 48%, specificity 72%) and Sokolow-Lyon voltage (AUC: 0.51, sensitivity 14%, specificity 96%). In the internal test set 2, the CNN-LSTM model had a stable performance in predicting LVH with an AUC of 0.59 (sensitivity 65%, specificity 57%). In the subgroup analysis, the CNN-LSTM model predicted LVH by 12-lead ECG with an AUC of 0.66 (sensitivity 72%, specificity 60%) for male patients, which performed better than that for female patients (AUC: 0.59, sensitivity 50%, specificity 71%).

Conclusion

Our study established a CNN-LSTM model to diagnose LVH by 12-lead ECG with higher sensitivity than current ECG diagnostic criteria. This CNN-LSTM model may be a simple and effective screening tool of LVH.

Electrocardiographic and cardiometabolic risk markers of left ventricular diastolic dysfunction in physically active adults: CHIEF heart study

AIM

This study was aimed to investigate the association of cardiometabolic and ECG markers with left ventricular diastolic dysfunction (LVDD) in physically active Asian young adults, which has not been clarified in prior studies.

METHODS AND RESULTS

A total of 2,019 men aged 18-43 years were included from the military in Taiwan. All the subjects underwent anthropometric, hemodynamic, and blood metabolic marker measurements. Physical fitness was investigated by time for a 3,000-m run. LVDD was defined by presence of either one of the three echocardiographic criteria: (1) mitral inflow E/A ratio < 0.8 with a peak E velocity of > 50 cm/s, (2) tissue Doppler lateral mitral annulus e' <10 cm/s, and (3) E/e' ratio > 14. Multiple logistic regressions with adjustments for age, physical fitness, and pulse rate were conducted to determine the association of cardiometabolic and ECG markers with LVDD. The prevalence of LVDD was estimated to be 4.16% (N = 84). Of the cardiometabolic markers, central obesity, defined as waist circumference ≥ 90 cm, was the only independent marker of LVDD [odds ratio (OR) and 95% confidence interval: 2.97 (1.63-5.41)]. There were no association for hypertension, prediabetes, and dyslipidemia. Of the ECG markers, left atrial enlargement and incomplete right bundle branch block/intraventricular conduction delay were the independent ECG markers of LVDD [OR: 2.98 (1.28-6.94) and 1.94 (1.09-3.47), respectively]. There was borderline association for Cornell-based left ventricular hypertrophy and inferior T wave inversion [OR: 1.94 (0.97-3.63) and 2.44 (0.98-6.08), respectively].

CONCLUSION

In the physically active Asian young male adults, central obesity and some ECG markers for left heart abnormalities were useful to identify LVDD.

Sex-specific cardiometabolic risk markers of left ventricular mass in physically active young adults: the CHIEF heart study

Greater physical fitness may lead to greater left ventricular mass (LVM) and reduce the effect of cardiometabolic risk factors on LVM. However, the cardiometabolic biomarkers associations for LVM have not been clarified in physically active young adults. This study included 2019 men and 253 women, aged 18-43 years, from the military in Taiwan. All participants underwent anthropometric and blood metabolic markers measurements, and completed a 3000-m run test for assessing fitness. LVM was calculated on the basis of an echocardiography. Multiple linear regression was used to determine the sex-specific associations between cardiometabolic risk markers and LVM indexed for the body height (g/m2.7). In men, age, systolic blood pressure (SBP), 3000-m running time, serum triglycerides, serum uric acid and waist circumference (WC) were correlated with LVM index (β = 0.07, 0.10, - 0.01, 0.01, 0.24 and 0.24, respectively; all p-values < 0.05). The correlations were not significant for fasting plasma glucose, total cholesterol and high-density lipoprotein cholesterol (HDL-C). In women, SBP, HDL-C and WC were correlated with LVM index in the univariate analysis (β = 0.07, - 0.05 and 0.32, respectively; all p-values < 0.05), whereas the correlation was only significant for WC in the multiple linear regression analysis (β = 0.20; p-value < 0.001). In physically active adults, the associations of cardiometabolic risk markers with LVM might vary by sex. Better endurance exercise performance associated with greater LVM was noted only in men, while greater WC was the only metabolic risk marker for greater LVM in both men and women.

Cardiorespiratory Fitness and Carotid Intima-Media Thickness in Physically Active Young Adults: CHIEF Atherosclerosis Study

Background: The relationship of cardiorespiratory fitness (CRF) with subclinical atherosclerosis affected by the body adiposity has been observed in children, whereas this relationship remains unclear in young adults. Methods and Results: A total of 1520 military recruits, aged 18−40 years, were included in Taiwan in 2018−2020. All subjects underwent detailed physical and blood laboratory examinations. CRF was evaluated by time for a 3000 m run, and subclinical atherosclerosis was evaluated by intima−media thickness of the bulb of the left common carotid artery (cIMT) utilizing high-resolution ultrasonography. Multivariable linear regression analysis with adjustments for age, sex, cigarette smoking, alcohol intake, systolic and diastolic blood pressure, high- and low-density lipoprotein cholesterols, fasting glucose, waist circumference, serum uric acid and serum triglycerides were utilized to determine the correlation between CRF and cIMT. CRF was independently correlated with cIMT (standardized β: 0.11, p < 0.001). Of the cardiometabolic risk markers, serum triglycerides were the only independent risk marker of cIMT (standardized β: 0.063, p = 0.03). In addition, the association of CRF with cIMT did not differ between those with a body mass index (BMI) ≥ 25 kg/m2 and those with BMI < 25 kg/m2 (standardized β: 0.103 and 0.117; p = 0.01 and 0.005, respectively). Conclusions: In physically active young men and women, there was an inverse association of cIMT with CRF, which was observed in both overweight/mild obesity and normal-weight individuals, highlighting the importance of endurance capacity on reducing risk of early atherosclerosis and implying that the moderation effect of body adiposity might not be present in this population.

Machine Learning for Electrocardiographic Features to Identify Left Atrial Enlargement in Young Adults: CHIEF Heart Study

BACKGROUND

Left atrial enlargement (LAE) is associated with cardiovascular events. Machine learning for ECG parameters to predict LAE has been performed in middle- and old-aged individuals but has not been performed in young adults.

METHODS

In a sample of 2,206 male adults aged 17-43 years, three machine learning classifiers, multilayer perceptron (MLP), logistic regression (LR), and support vector machine (SVM) for 26 ECG features with or without 6 biological features (age, body height, body weight, waist circumference, and systolic and diastolic blood pressure) were compared with the P wave duration of lead II, the traditional ECG criterion for LAE. The definition of LAE is based on an echocardiographic left atrial dimension > 4 cm in the parasternal long axis window.

RESULTS

The greatest area under the receiver operating characteristic curve is present in machine learning of the SVM for ECG only (77.87%) and of the MLP for all biological and ECG features (81.01%), both of which are superior to the P wave duration (62.19%). If the sensitivity is fixed to 70-75%, the specificity of the SVM for ECG only is up to 72.4%, and that of the MLP for all biological and ECG features is increased to 81.1%, both of which are higher than 48.8% by the P wave duration.

CONCLUSIONS

This study suggests that machine learning is a reliable method for ECG and biological features to predict LAE in young adults. The proposed MLP, LR, and SVM methods provide early detection of LAE in young adults and are helpful to take preventive action on cardiovascular diseases.

Athlete's Heart Assessed by Sit-Up Strength Exercises in Military Men and Women: The CHIEF Heart Study

BACKGROUND

Greater changes in cardiac structure and function in response to physical training have been observed more often in male athletes than in female athletes compared with their sedentary controls. However, studies for the sex-specific cardiac remodeling related to strength exercises in Asian athletes are rare.

METHODS

This study included 580 men and 79 women, with an average age of 25 years, for a 6-month military training program in Taiwan. Both men and women attended a 2-min sit-up test to assess muscular strength after the training. The test performance falling one standard deviation above the mean (16%) was to define the superior eliteness of athletes. Cardiac structure and function were investigated by electrocardiography and echocardiography for men and women. Multiple logistic regression was used to determine the predictors of elite athlete status.

RESULTS

In men, greater QTc interval, left ventricular mass adjusted to body surface area (LVMI), lateral mitral E'/A' ratio and right ventricular systolic pressure, and lower diastolic blood pressure were independent predictors of elite strength athletes in the sit-up test [odds ratio (OR) and 95% confidence intervals: 1.01 (1.00, 1.02), 1.02 (1.00, 1.04), 1.45 (1.06, 1.98), 1.13 (1.06, 1.23), and 0.96 (0.93, 0.99), respectively. In contrast, in women, the greater right ventricular outflow tract dimension was the only independent predictor of elite strength athletes in the sit-up test [OR: 1.26 (1.04, 1.53)].

CONCLUSIONS

In the 2-min sit-up test, cardiac characteristics differ between elite male and female athletes. While greater QTc interval, LVMI, and diastolic function of left ventricle predict the eliteness of male strength athletes, greater right ventricular chamber size characterizes elite female strength athletes.

Athlete's Heart in Asian Military Males: The CHIEF Heart Study

Background: Elite athlete's heart is characterized by a greater left ventricular mass indexed by body surface area (LVMI) and diastolic function; however previous studies are mainly conducted in non-Asian athletes compared to sedentary controls. Methods: This study included 1,388 male adults, aged 18-34 years, enrolled in the same unified 6-month physical training program in Taiwan. During the midterm exams of 2020, all trainees completed a 3-km run (endurance) test, and 577 were randomly selected to attend a 2-min push-up (muscular strength) test. Elite athletes were defined as the performance of each exercise falling one standard deviation above the mean (16%). Cardiac structure and function were measured by echocardiography and compared between elite and non-elite athletes. Multiple logistic regression analysis was used to determine the independent predictors of elite athlete status at each exercise modality. Results: As compared to non-elite controls, elite endurance athletes had greater LVMI (84.4 ± 13.6 vs. 80.5 ± 12.9 g/m2, p < 0.001) and lateral mitral E'/A' ratio (2.37 ± 0.73 vs. 2.22 ± 0.76, p < 0.01) with lower late diastolic A' (7.77 ± 2.16 vs. 8.30 ± 3.69 cm/s, p = 0.03). Elite strength athletes had greater LVMI (81.8 ± 11.4 vs. 77.5 ± 12.1, p = 0.004) and lateral mitral E'/A' ratio (2.36 ± 0.70 vs. 2.11 ± 0.71, p < 0.01) with a greater early diastolic E' (19.30 ± 4.06 vs. 18.18 ± 4.05 cm/s, p = 0.02). Greater LVMI and lower heart rate were independent predictors of elite endurance athletes [odds ratio (OR) and 95% confidence intervals: 1.03 (1.02, 1.04) and 0.96 (0.95, 0.98), respectively]. Greater LVMI, lateral mitral E'/A' ratio and right ventricular systolic pressure were independent predictors of elite strength athletes [OR: 1.03 (1.01, 1.05), 1.50 (1.06, 2.12), and 1.12 (1.05, 1.19), respectively]. Conclusions: Cardiac structural and functional characteristics differ between endurance and strength elite athletes. While greater LVMI predicts elite status in both groups of Asian athletes, consistent with findings from Western elite athletes, greater diastolic function, and right ventricular systolic pressure characterize strength elite athletes, while lower heart rate at rest predicts endurance elite athletic status.

Metabolically healthy obesity and physical fitness in military males in the CHIEF study

The metabolically healthy obese (MHO) characterized by the absence of metabolic syndrome have shown superior cardiorespiratory fitness (CRF) and similar muscular strength as compared with the metabolically unhealthy obese (MUO). However, this finding might be biased by the baseline sedentary behavior in the general population. This study utilized 3669 physically active military males aged 18-50 years in Taiwan. Obesity and metabolically unhealthy were respectively defined as body mass index ≥ 27.5 kg/m2 and presence of at least two major components of the metabolic syndrome, according to the International Diabetes Federation criteria for Asian male adults. Four groups were accordingly classified as the metabolically healthy lean (MHL, n = 2510), metabolically unhealthy lean (MUL, n = 331), MHO (n = 181) and MUO (n = 647). CRF was evaluated by time for a 3-km run, and muscular strengths were separately assessed by numbers of push-up and sit-up within 2 min. Analysis of covariance was utilized to compare the difference in each exercise performance between groups adjusting for age, service specialty, smoking, alcohol intake and physical activity. The metabolic syndrome prevalence in MUL and MUO was 49.8% and 47.6%, respectively. The performance of CRF did not differ between MHO and MUO (892.3 ± 5.4 s and 892.6 ± 3.0 s, p = 0.97) which were both inferior to MUL and MHL (875.2 ± 4.0 s and 848.6 ± 1.3 s, all p values < 0.05). The performance of muscular strengths evaluated by 2-min push-ups did not differ between MUL and MUO (45.3 ± 0.6 and 45.2 ± 0.4, p = 0.78) which were both less than MHO and MHL (48.4 ± 0.8 and 50.6 ± 0.2, all p values < 0.05). However, the performance of 2-min sit-ups were only superior in MHL (48.1 ± 0.1) as compared with MUL, MHO and MUO (45.9 ± 0.4, 46.7 ± 0.5 and 46.1 ± 0.3, respectively, all p values < 0.05). Our findings suggested that in a physically active male cohort, the MHO might have greater muscle strengths, but have similar CRF level compared with the MUO.

MLBF-Net: A Multi-Lead-Branch Fusion Network for Multi-Class Arrhythmia Classification Using 12-Lead ECG

Automatic arrhythmia detection using 12-lead electrocardiogram (ECG) signal plays a critical role in early prevention and diagnosis of cardiovascular diseases. In the previous studies on automatic arrhythmia detection, most methods concatenated 12 leads of ECG into a matrix, and then input the matrix to a variety of feature extractors or deep neural networks for extracting useful information. Under such frameworks, these methods had the ability to extract comprehensive features (known as integrity) of 12-lead ECG since the information of each lead interacts with each other during training. However, the diverse lead-specific features (known as diversity) among 12 leads were neglected, causing inadequate information learning for 12-lead ECG. To maximize the information learning of multi-lead ECG, the information fusion of comprehensive features with integrity and lead-specific features with diversity should be taken into account. In this paper, we propose a novel Multi-Lead-Branch Fusion Network (MLBF-Net) architecture for arrhythmia classification by integrating multi-loss optimization to jointly learning diversity and integrity of multi-lead ECG. MLBF-Net is composed of three components: 1) multiple lead-specific branches for learning the diversity of multi-lead ECG; 2) cross-lead features fusion by concatenating the output feature maps of all branches for learning the integrity of multi-lead ECG; 3) multi-loss co-optimization for all the individual branches and the concatenated network. We demonstrate our MLBF-Net on China Physiological Signal Challenge 2018 which is an open 12-lead ECG dataset. The experimental results show that MLBF-Net obtains an average [Formula: see text] score of 0.855, reaching the highest arrhythmia classification performance. The proposed method provides a promising solution for multi-lead ECG analysis from an information fusion perspective.

Prevalence and characteristics of mitral valve prolapse in military young adults in Taiwan of the CHIEF Heart Study

The prevalence of mitral valve prolapse (MVP) among middle- and older-aged individuals is estimated to be 2-4% in Western countries. However, few studies have been conducted among Asian individuals and young adults. This study included a sample of 2442 consecutive military adults aged 18-39 years in Hualien, Taiwan. MVP was defined as displacement of the anterior or posterior leaflet of the mitral valve to the mid portion of the annular hinge point > 2 mm in the parasternal long-axis view of echocardiography. Cardiac chamber size and wall thickness were measured based on the latest criteria of the American Society of Echocardiography. The clinical features of participants with MVP and those without MVP were compared using a two-sample t test, and the cardiac structures were compared using analysis of covariance with adjustment for body surface area (BSA). Eighty-two participants were diagnosed with MVP, and the prevalence was 3.36% in the overall population. Compared with those without MVP, participants with MVP had a lower body mass index (kg/m2) (24.89 ± 3.70 vs. 23.91 ± 3.45, p = 0.02) and higher prevalence of somatic symptoms related to exercise (11.0% vs. 4.9%, p = 0.02) and systolic click in auscultation (18.3% vs. 0.6%, p < 0.01). In addition, participants with MVP had greater left ventricular mass (gm) and smaller right ventricular wall thickness (mm) and dimensions (mm) indexed by BSA than those without MVP (149.12 ± 35.76 vs. 155.38 ± 36.26; 4.66 ± 0.63 vs. 4.40 ± 0.68; 26.57 ± 3.99 vs. 25.41 ± 4.35, respectively, all p-values < 0.01). In conclusion, the prevalence and clinical features of MVP in military young adults in Taiwan were in line with those in Western countries. Whether the novel MVP phenotype found in this study has any pathological meaning needs further investigation.

Tobacco Smoking and Association between Betel Nut Chewing and Metabolic Abnormalities Among Military Males: The CHIEF Study

AIM

To investigate the effect of smoking and alcohol intake on the association between betel nut chewing and each metabolic abnormality.

BACKGROUND

Betel nut chewing has been associated with metabolic syndrome.

OBJECTIVE

Whether the association is affected by tobacco or alcohol use is not clarified so far.

METHODS

The authors conducted a cross-sectional study using 6,657 military males, aged 18-50 years in eastern Taiwan in 2013-2014. Metabolic syndrome was defined according to the International Diabetes Federation's ethnic criteria for Asians. The population was classified as non-betel nut chewers (N =5,749), current chewers with both tobacco and alcohol use (N =615), and current chewers without tobacco and/or alcohol use (N =293). Multiple logistic regression analyses were stepwise adjusted for the confounders including alcohol and tobacco use to determine the association of betel chewing with the metabolic abnormalities.

RESULTS

As compared to the non-current chewers, the current chewers with both tobacco/alcohol use and those without had a higher risk of metabolic syndrome (odds ratios (OR) and 95% confidence intervals: 2.46 (2.00-3.02), and 2.04 (1.53-2.73), respectively) after controlling for age, service specialty, total cholesterol levels ≥200 mg/dL and exercise frequency (model 1). The association did not change much in the two chewing groups after additionally adjusting for alcohol consumption (model 2) (OR: 2.49 (1.99-3.12), and 2.04 (1.52-2.73), respectively), whereas the relationship reduced significantly in the chewers with both tobacco/alcohol use rather than those without after further adjusting for smoking (model 3) (OR: 2.18 (1.71-2.78) and 2.02 (1.51-2.71), respectively). This was in parallel with the pattern for the association of betel nut chewing with serum triglycerides >150 mg/dL in the chewers with both tobacco/alcohol use and those without in model 1 (OR: 2.90 (2.40-3.51) and 1.90 (1.45-2.49), respectively, p =0.011), in model 2 (OR: 2.82 (2.30-3.46) and 1.89 (1.44-2.49), respectively, p =0.040), and in model 3 (2.26 (1.81-2.81) and 1.87 (1.42-2.45), respectively, p =0.76).

CONCLUSION

Our findings suggest that tobacco smoking but not alcohol intake could increase the relationship of betel nut chewing with metabolic syndrome, which is likely mediated by a synergic effect on increasing serum triglycerides levels.

Psychological stress and long-term blood pressure variability of military young males: The cardiorespiratory fitness and hospitalization events in armed forces study

BACKGROUND

Acute stress might increase short-term heart rate variability and blood pressure variability (BPV); however, chronic stress would not alter short-term BPV in animal models.

AIM

To examine the association of psychological stress with long-term BPV in young male humans.

METHODS

We prospectively examined the association of chronic psychological stress with long-term BPV in 1112 healthy military males, averaged 32.2 years from the cardiorespiratory fitness and hospitalization events in armed forces study in Taiwan. Psychological stress was quantitatively evaluated with the Brief Symptom Rating Scale (BSRS-5), from the least symptom of 0 to the most severe of 20, and the five components of anxiety, insomnia, depression, interpersonal sensitivity, and hostility (the severity score in each component from 0 to 4). Long-term BPV was assessed by standard deviation (SD) for systolic and diastolic blood pressure (SBP and DBP), and average real variability (ARV), defined as the average absolute difference between successive measurements of SBP or DBP, across four visits in the study period from 2012 to 2018 (2012-14, 2014-15, 2015-16, and 2016-18).

RESULTS

The results of multivariable linear regressions showed that there were no correlations of the BSRS-5 score with SDSBP, SDDBP, ARVSBP, and ARVDBP after adjusting for all the covariates [β(SE): -0.022 (0.024), -0.023 (0.026), -0.001 (0.018), and 0.001 (0.020), respectively; P > 0.05 for all]. In addition, there were also no correlations between each component of the BSRS score and the long-term BPV indexes.

CONCLUSION

Our findings suggest that chronic psychological stress might not be associated with long-term BPV in military young male humans.

Association of Liver Transaminase Levels and Long-Term Blood Pressure Variability in Military Young Males: The CHIEF Study

Background: An inverse relationship of serum liver transaminases and mortality might be due to better blood pressure control in hypertensive patients. Whether it holds true regarding such an association for long-term blood pressure variability (BPV) in those without antihypertensive therapy is unclear. Methods: A population of 1112 military males without antihypertensive medications, aged 32 years, was collected from a retrospective longitudinal study in Taiwan. Serum liver aspartate and alanine transaminase (AST and ALT) levels were obtained from a 12 h-fast blood sample of each participant. BPV was assessed by standard deviation (SD) and average real variability (ARV) of systolic and diastolic blood pressure (SBP and DBP), respectively across 4 visits during the study period (2012-2014, 2014-2015, 2015-2016, and 2016-2018). Multivariable linear regression analysis was utilized to determine the association adjusting for demographics, anthropometric indexes, SBP, DBP, and lipid profiles. Results: In the unadjusted model, ALT was significantly and positively correlated with SDDBP and ARVDBP (β (standard errors) = 0.36 (0.16) and 0.24 (0.12), respectively), and so was AST (β = 0.19 (0.08) and 0.14 (0.06), respectively). All the associations were insignificant with adjustments. However, ALT was significantly and negatively correlated with SDSBP and ARVSBP (β = -0.35 (0.14) and -0.25 (0.11), respectively) and so was AST (β = -0.14 (0.07) and -0.12 (0.06), respectively) with adjustments. Conclusion: Our findings suggested that serum liver transaminases were negatively correlated with long-term systolic BPV in young male adults without antihypertensive therapy, and the clinical relevance needs further investigations.

Association of Single Measurement of dipstick proteinuria with physical performance of military males: the CHIEF study

BACKGROUND

Proteinuria, a marker of kidney injury, may be related to skeletal muscle loss. Whether the severity of proteinuria is associated with physical performance is unclear.

METHODS

We examined the association of proteinuria severity with physical performance cross-sectionally in 3357 military young males, free of chronic kidney disease, from the cardiorespiratory fitness and hospitalization events in armed Forces (CHIEF) study in Taiwan. The grades of proteinuria were classified according to one dipstick urinalysis which were collected at morning after an 8-h fast as unremarkable (0, +/-, and 1+), moderate (2+) and severe (3+ and 4+). Aerobic physical performance was evaluated by time for a 3000-m run and anaerobic physical performance was evaluated by numbers of 2-min sit-ups and 2-min push-ups, separately. Multiple linear regressions were used to determine the relationship.

RESULTS

As compared with unremarkable proteinuria, moderate and severe proteinuria were dose-dependently correlated with 3000-m running time (β: 4.74 (95% confidence intervals (CI): - 0.55, 10.02) and 7.63 (95% CI: 3.21, 12.05), respectively), and inversely with numbers of 2-min push-ups (β = - 1.13 (- 1.97, - 0.29), and - 1.00 (- 1.71, - 0.28), respectively) with adjustments for age, service specialty, body mass index, blood pressure, alcohol intake, smoking, fasting plasma glucose, blood urea nitrogen, serum creatinine and physical activity. However, there was no association between proteinuria severity and 2-min sit-ups.

CONCLUSIONS

Our findings show a relationship of dipstick proteinuria with aerobic physical performance and parts of anaerobic physical performance in military healthy males. This mechanism is not fully understood and requires further investigations.

A 12-Lead ECG-Based System With Physiological Parameters and Machine Learning to Identify Right Ventricular Hypertrophy in Young Adults

OBJECTIVE

The presence of right ventricular hypertrophy (RVH) accounts for approximately 5-10% in young adults. The sensitivity estimated by commonly used 12-lead electrocardiographic (ECG) criteria for identifying the presence of RVH is under 20% in the general population. The aim of this study is to develop a 12-lead ECG system with the related information of age, body height and body weight via machine learning to increase the sensitivity and the precision for detecting RVH.

METHOD

In a sample of 1,701 males, aged 17-45 years, support vector machine is used for the training of 31 parameters including age, body height and body weight in addition to 28 ECG data such as axes, intervals and wave voltages as the inputs to link the output RVH. The RVH is defined on the echocardiographic finding for young males as right ventricular anterior wall thickness > 5.5 mm.

RESULTS

On the system goal for increasing sensitivity, the specificity is controlled around 70-75% and all data tested in the proposed method show competent sensitivity up to 70.3%. The values of area under curve of receiver operating characteristic curve and precision-recall curve using the proposed method are 0.780 and 0.285, respectively, which are better than 0.518 and 0.112 using the Sokolow-Lyon voltage criterion, respectively, for detecting unspecific RVH.

CONCLUSION

We present a method using simple physiological parameters with ECG data to effectively identify more than 70% of the RVH among young adults. Clinical Impact: This system provides a fast, precise and feasible diagnosis tool to screen RVH.