An Integrated Heart Sound–Electrocardiogram Synchronization System for Early Detecting Left Ventricular Systolic Dysfunction in Pediatric Populations

ABSTRACT

Background

Early detection of pediatric left ventricular systolic dysfunction (LVSD) remains challenging in resource‐limited settings. This study evaluates a novel synchronized heart sound‐electrocardiogram (HS‐ECG) system for LVSD detection in underserved pediatric populations.

Hypothesis

Cardiac acoustic biomarkers (CABs) enable age‐specific early detection of pediatric LVSD.

Methods

This prospective cohort study enrolled 212 children aged 3–16 years, classified into case (LVEF ≤ 55%, n = 80) and control (LVEF > 55%, n = 132) groups with age stratification (3–6 and 7–16 years). All underwent synchronized HS–ECG recording using a high‐fidelity wearable device. Signals were processed via wavelet analysis to derive CABs including electromechanical activation time (EMAT), pre ‐ ejection period (PEP), left ventricular ejection time (LVET), left ventricular systolic time (LVST), their rate‐corrected values (EMATc, PEPc, LVETc, LVSTc), and the PEP/LVET ratio. Echocardiographic LVEF served as the reference standard.

Results

In children aged 3–6 years, EMATc showed strong inverse correlation with LVEF (r = −0.529, p < 0.001) and AUC = 0.821 (cutoff: 11.80%; sensitivity 82.5%, specificity 84.7%). In those aged 7–16 years, EMAT demonstrated superior performance (r = −0.609, p < 0.001; AUC = 0.896; cutoff: 82.50 ms; sensitivity 82.5%, specificity 93.2%).

Conclusion

The synchronized HS‐ECG system provides clinically valuable biomarkers for pediatric LVSD diagnosis, demonstrating distinct age‐dependent diagnostic patterns: EMATc shows optimal performance in younger children (3–6 years), while EMAT exhibits superior efficacy in older children (7–16 years).

This study validates a novel synchronized heart sound‐electrocardiogram (HS‐ECG) system for early detection of pediatric left ventricular systolic dysfunction (LVSD). The optimal diagnostic biomarker is age‐dependent: EMATc for children aged 3–6 years and EMAT for those aged 7–16 years.

1. Introduction

Left ventricular systolic dysfunction (LVSD) represents a primary manifestation of heart failure in pediatric populations, characterized by structural and functional abnormalities that compromise ventricular contractility and reduce cardiac output relative to physiological demands. LVSD further disrupts neuroendocrine regulation, leading to a complex clinical syndrome with multisystem involvement (Jayaprasad 2016; Krittanawong et al. ²⁰²³; Subspecialty Group of Cardiology, the Society of Pediatrics, Chinese Medical Association; Pediatric Cardiovascular Disease Committee, College of Cardiovascular Physicians, Chinese Medical Doctor Association; Editorial Board, Chinese Journal of Pediatrics ²⁰²¹). Affected children often require long‐term therapeutic intervention and face elevated risks of life‐threatening complications, including respiratory failure, renal impairment, and sepsis (Nandi and Rossano 2015). Epidemiological data from a 2009 US clinical survey indicate a mortality rate of 6.7% in pediatric heart failure, underscoring the imperative for early detection and intervention in LVSD (Rossano et al. 2012).

Notably, although echocardiography remains the gold standard for assessing left ventricular function, its implementation in resource‐limited settings is significantly constrained by high costs, limited equipment availability, and dependency on specialized technical expertise. These barriers are particularly pronounced in remote rural regions and socioeconomically disadvantaged urban communities, where they substantially impede early diagnosis of pediatric LVSD and complicate long‐term remote monitoring of chronic cases. Consequently, the development of accessible, cost‐effective, and scalable alternative diagnostic tools is critically needed.

A novel noninvasive synchronized heart sound–electrocardiogram (HS–ECG) monitoring system enables quantitative analysis of derived parameters—including electromechanical activation time (EMAT) and left ventricular ejection time (LVET)—which have demonstrated significant correlations with LVSD in adult populations (Romer et al. 2018; Fudim et al. ²⁰²²; Butler et al. ²⁰²⁵; Erath et al. ²⁰²⁰). However, its applicability in pediatric cohorts remains inadequately explored, primarily due to developmental variations in cardiac physiology and the absence of established pediatric reference values. This study therefore aims to rigorously evaluate the feasibility and diagnostic potential of this synchronized monitoring approach for early detection of LVSD in children.

2. Research Methods and Resources

2.1. Ethical Approval

The study protocol received approval from the Ethics Committee of the Children's Hospital of Soochow University (Approval No. 2022CS181). Written informed consent was obtained from all participants after detailed explanation of the study objectives and procedures.

2.2. Study Population

We conducted a prospective study that recruited 212 pediatric patients (aged 3–16 years) who underwent echocardiographic evaluation at our institution between October 2022 and December 2023. Participants were stratified into a control group (n = 132; left ventricular ejection fraction [LVEF] ≥ 55%) and a case group (n = 80; LVEF < 55%) based on LVEF measurements (Subspecialty Group of Cardiology, the Society of Pediatrics, Chinese Medical Association; Pediatric Cardiovascular Disease Committee, College of Cardiovascular Physicians, Chinese Medical Doctor Association; Editorial Board, Chinese Journal of Pediatrics ²⁰²¹). To account for potential heterogeneity in cardiac acoustic biomarkers (CABs) across development, each group was subsequently subdivided into younger (3–6 years; n = 99; cases = 40, controls = 59) and older (7–16 years; n = 113; cases = 40, controls = 73) subgroups (Subspecialty Group of Cardiology, the Society of Pediatrics, Chinese Medical Association; Pediatric Cardiovascular Disease Committee, College of Cardiovascular Physicians, Chinese Medical Doctor Association; Editorial Board, Chinese Journal of Pediatrics ²⁰²¹; Zhu et al. ²⁰⁰⁶).

2.3. Inclusion Criteria

Case group participants met all of the following criteria: (1) age ≥ 3 years; (2) echocardiographically confirmed LVSD (LVEF < 55%) (Subspecialty Group of Cardiology, the Society of Pediatrics, Chinese Medical Association; Pediatric Cardiovascular Disease Committee, College of Cardiovascular Physicians, Chinese Medical Doctor Association; Editorial Board, Chinese Journal of Pediatrics ²⁰²¹); (3) diagnosis of LVSD, cardiac insufficiency, or heart failure according to the 2020 Chinese Medical Association guidelines (Subspecialty Group of Cardiology, the Society of Pediatrics, Chinese Medical Association; Pediatric Cardiovascular Disease Committee, College of Cardiovascular Physicians, Chinese Medical Doctor Association; Editorial Board, Chinese Journal of Pediatrics ²⁰²¹); and (4) ability to complete the synchronized HS‐ECG monitoring protocol.

Control group participants were required to: (1) be ≥ 3 years of age; (2) exhibit preserved left ventricular systolic function (LVEF ≥ 55%); (3) have no history of LVSD, heart failure, or cardiac insufficiency; and (4) demonstrate ability to complete the full monitoring protocol (Subspecialty Group of Cardiology, the Society of Pediatrics, Chinese Medical Association; Pediatric Cardiovascular Disease Committee, College of Cardiovascular Physicians, Chinese Medical Doctor Association; Editorial Board, Chinese Journal of Pediatrics ²⁰²¹).

2.4. Exclusion Criteria

Exclusion criteria for both groups included: (1) age < 3 years (due to reduced compliance and increased signal interference); (2) cardiac conduction abnormalities affecting QRS duration (e.g., ventricular preexcitation, sustained ventricular arrhythmias, bundle branch blocks, or paced rhythms); or (3) inability to tolerate the monitoring procedure.

2.5. Data Collection

2.5.1. Clinical Data Collection

Demographic and anthropometric parameters—including medical record number, name, sex, age, height, weight, blood pressure, and diagnosis—were retrieved from electronic medical records. Body surface area (BSA) was calculated using the formula: BSA (m²) = √[weight (kg) × height (cm)/3600] (Redlarski et al. 2016).

2.5.2. Echocardiographic Data Acquisition

All pediatric participants underwent comprehensive transthoracic echocardiography in the awake state using either the GE Vivid E90/E95 (GE Vingmed Ultrasound, Horten, Norway) or Philips iE33 (Philips Healthcare, Andover, Massachusetts, USA) ultrasound systems. LVEF was quantified via the biplane Simpson's method. All studies were performed and analyzed by board‐certified echocardiographers in strict accordance with standardized imaging protocols to ensure methodological accuracy and reproducibility.

2.5.3. HS–ECG Synchronized Data Acquisition

All data were acquired by investigators using the HR‐0536 portable HS‐ECG recorder (Wenxin Technology, Fuzhou, China). The device integrated a reusable core module with a high‐fidelity piezoelectric acoustic sensor and disposable adhesive patches containing medical‐grade electrodes for single‐lead ECG acquisition. During recording, the patch was applied to standard auscultation sites (mitral area) to simultaneously capture phonocardiographic and electrocardiographic waveforms. Data were transmitted via Bluetooth to a cloud‐based platform for automated analysis and report generation (Figure 1).

FIGURE 1.

Synchronized monitoring device for HS–ECG and automatic analysis results.

Participants remained supine with quiet respiration during examination. Synchronized 30‐s recordings were obtained immediately after echocardiographic assessment. Data processing utilized a proprietary analytical platform employing wavelet‐based signal processing to derive the following CABs (Figure 2) (Fudim et al. 2022; Butler et al. ²⁰²⁵):

1. Electromechanical Activation Time (EMAT): The time interval from the onset of the QRS complex on the electrocardiogram to mitral valve closure.

2. EMATc: EMAT expressed as a percentage of the electrocardiogram R‐R interval.

3. Pre‐ejection Period (PEP): The time interval from the onset of the QRS complex on the electrocardiogram to aortic valve opening.

4. PEPc: PEP expressed as a percentage of the electrocardiogram R‐R interval.

5. Left Ventricular Ejection Time (LVET): The duration from aortic valve opening to aortic valve closure.

6. LVETc: LVET expressed as a percentage of the electrocardiogram R‐R interval.

7. PEP/LVET: The ratio of PEP to LVET.

8. Left Ventricular Systolic Time (LVST): The time interval from mitral valve closure to aortic valve closure.

9. LVSTc: LVST expressed as a percentage of the electrocardiogram R‐R interval.

FIGURE 2.

Schematic representation of cardiac acoustic biomarkers (Pressure waveforms are included for physiological reference). Ao, aorta; ECG, electrocardiogram; EMAT, electromechanical activation time; LA, left atrium; LV, left ventricle; LVET, left ventricular ejection time; LVST, left ventricular systolic time; PEP, pre‐ejection period; S1, first heart sound; S2, second heart sound.

3. Statistical Analysis

All statistical analyzes were performed using SPSS (version 27.0; IBM, Armonk, New York, USA) and GraphPad Prism (version 10.0; GraphPad Software, San Diego, California, USA). Categorical variables were expressed as frequencies (percentages) and compared using Pearson's chi‐square tests. The normality of continuous variables was assessed using the Shapiro–Wilk test. Normally distributed variables were presented as mean ± standard deviation (SD) and analyzed using independent samples t‐tests. Nonnormally distributed variables were reported as median (interquartile range) and compared using Mann–Whitney U tests. Bivariate correlations between LVEF and CABs were quantified using Spearman's rank correlation coefficients. Binary logistic regression analysis was employed to identify independent factors associated with LVSD diagnosis. The diagnostic performance of significant parameters was evaluated using receiver operating characteristic (ROC) curve analysis, with area under the curve (AUC) calculations. A two‐tailed p < 0.05 was considered statistically significant for all analyzes.

4. Results

4.1. Study Findings in the 3‐ To 6‐Year‐Old Children Cohort

4.1.1. Demographic Characteristics

The study cohort comprised 40 pediatric patients diagnosed with LVSD (male: female ratio, 24:16; median age, 4.0 years) and 59 age‐ and sex‐matched controls (male: female ratio, 33:26; median age, 4.0 years). Compared with controls, LVSD patients showed significantly impaired cardiovascular function, characterized by reduced systolic blood pressure (85.0 [78.0–95.0] vs. 95.0 [86.0–102.0] mmHg; p = 0.004) and lower LVEF (45.0% [35.5–50.8] vs. 62.0% [58.0–65.0]; p < 0.001). Compensatory tachycardia was observed in LVSD patients (109.7 ± 25.6 vs. 99.4 ± 20.6 bpm; p = 0.030). No significant intergroup differences were found in age, sex distribution, height, weight, or body surface area (Table 1).

TABLE 1.

Comparison of demographic characteristics and cardiac acoustic biomarkers between case and control groups in the 3‐ to 6‐year‐old cohort.

Variable	Cases (n = 40)	Controls (n = 59)	p
Demographic characteristics
Age, years	4.00 (3.00–5.00)	4.00 (3.00–5.00)	0.604
Male, n (%)	24 (60.0)	33 (55.9)	0.688
Height, cm	105.00 (96.75–110.00)	107.00 (102.00–112.00)	0.162
Weight, kg	16.91 ± 3.78	18.25 ± 3.18	0.060
BSA, m2	0.70 ± 0.11	0.74 ± 0.09	0.052
SBP, mmHg	85.00 (78.00–95.00)	95.00 (86.00–102.00)	0.004
HR, bpm	109.68 ± 25.59	99.40 ± 20.64	0.030
LVEF, %	45.00 (35.50–50.75)	62.00 (58.00–65.00)	< 0.001
Cardiac acoustic biomarkers
EMAT, ms	70.00 (61.25–81.00)	60.00 (53.00–70.00)	< 0.001
EMATc, %	12.99 ± 2.68	10.08 ± 2.10	< 0.001
PEP, ms	85.50 (74.25–109.50)	78.00 (70.00–89.00)	0.017
PEPc, %	16.13 (13.40–19.08)	12.41 (10.45–15.79)	< 0.001
LVET, ms	236.68 ± 53.43	274.41 ± 42.69	< 0.001
LVETc, %	42.89 (34.53–46.59)	44.80 (40.91–47.59)	0.055
LVST, ms	255.00 (230.75–283.75)	293.00 (270.00–316.00)	< 0.001
LVSTc, %	45.49 (39.55–49.63)	47.59 (43.38–51.32)	0.113
PEP/LVET, %	40.58 ± 13.52	29.70 ± 7.57	< 0.001

Note: Data are presented as median (interquartile range), mean ± standard deviation, or number (percentage). Significant p‐values (< 0.05) are highlighted in bold.

Abbreviations: BSA, body surface area; EMAT, electromechanical activation time; EMATc, heart rate‐corrected EMAT; HR, heart rate; LVEF, left ventricular ejection fraction; LVET, left ventricular ejection time; LVST, left ventricular systolic time; PEP, pre‐ejection period; SBP, systolic blood pressure.

4.1.2. Cardiac Acoustic Biomarkers

4.1.2.1. Intergroup Differences

In the cohort of children aged 3–6 years, comparative assessment revealed significant alterations in CABs between LVSD patients and matched controls. The case group exhibited markedly elevated values for EMAT (70.00 [61.25–81.00] vs. 60.00 [53.00–70.00] ms; p < 0.001), EMATc (12.99% ± 2.68% vs. 10.08% ± 2.10%; p < 0.001), PEP (85.50 [74.25–109.50] vs. 78.00 [70.00–89.00] ms; p = 0.017), PEPc (16.13% [13.40–19.08] vs. 12.41% [10.45–15.79]; p < 0.001), and the PEP/LVET ratio (40.58% ± 13.52% vs. 29.70% ± 7.57%; p < 0.001). Conversely, LVET (236.68 ± 53.43 vs. 274.41 ± 42.69 ms; p < 0.001) and LVST (255.00 [230.75–283.75] vs. 293.00 [270.00–316.00] ms; p < 0.001) were significantly reduced in the case group. No statistically significant intergroup differences were observed for LVETc or LVSTc (p > 0.05), as detailed in Table 1.

4.1.2.2. Correlation Analysis

Bivariate correlation analysis revealed significant associations between CABs and LVEF. LVEF demonstrated a strong inverse correlation with EMATc (r = −0.529, p < 0.001). Weaker negative correlations were observed with EMAT (r = −0.273, p = 0.006), PEPc (r = −0.390, p < 0.001), and the PEP/LVET ratio (r = −0.372, p < 0.001). Positive correlations were identified with LVET (r = 0.396, p < 0.001) and LVST (r = 0.415, p < 0.001). No significant correlation was found between LVEF and PEP (p > 0.05).

4.1.2.3. Logistic Regression Analysis

Binary logistic regression analysis identified EMATc as an independent relevant factor for LVSD diagnosis (odds ratio [OR] = 1.666, 95% confidence interval [CI]: 1.334–2.081; p < 0.001) following forward stepwise selection from clinically relevant parameters showing significant intergroup differences.

4.1.2.4. Diagnostic Performance of EMATc by ROC Curve Analysis

ROC curve analysis for EMATc confirmed its diagnostic utility, demonstrating an AUC of 0.821 (95% CI: 0.731–0.910; p < 0.001). The optimal diagnostic cutoff for LVSD was established at 11.80%, corresponding to a Youden's index of 0.672, with sensitivity and specificity of 82.5% and 84.7%, respectively (Figure 3).

FIGURE 3.

Receiver operating characteristic (ROC) curve of heart rate–corrected electromechanical activation time (EMATc) in left ventricular systolic dysfunction (LVSD) diagnosis.

4.2. Study Findings in the 7‐ To 16‐Year‐Old Children Cohort

4.2.1. Demographic Characteristics

This analysis included 40 children aged 7–16 years with LVSD and 73 control subjects. The case group contained a higher percentage of male participants (85.0% vs. 61.6%; p = 0.010). Nonetheless, the two groups were well‐matched for other baseline characteristics, including age, height, weight, and BSA (all p > 0.05). Cardiovascular function was markedly impaired in the case group, as evidenced by a significantly reduced LVEF (38.5% [34.0–49.8] vs. 61.0% [57.0–64.0]; p < 0.001). No statistically significant differences were observed in systolic blood pressure or heart rate between the two groups (Table 2).

TABLE 2.

Comparison of demographic characteristics and cardiac acoustic biomarkers between case and control groups in the 7‐ to 16‐year‐old cohort.

Variable	Cases (n = 40)	Controls (n = 73)	p
Demographic characteristics
Age, years	12.00 (8.00–14.00)	11.00 (8.00–14.00)	0.541
Male, n (%)	34 (85.0)	45 (61.6)	0.010
Height, cm	159.00 (125.25–168.00)	155.00 (133.50–168.50)	0.812
Weight, kg	44.25 (24.00–59.50)	45.00 (28.00–56.50)	0.542
BSA, m2	1.43 (0.91–1.68)	1.38 (1.03–1.64)	0.627
SBP, mmHg	103.28 ± 18.43	108.60 ± 11.03	0.100
HR, bpm	81.00 (70.00–98.00)	79.00 (69.50–87.00)	0.249
LVEF, %	38.50 (34.00–49.75)	61.00 (57.00–64.00)	< 0.001
Cardiac acoustic biomarkers
EMAT, ms	91.80 ± 17.60	64.95 ± 10.67	< 0.001
EMATc, %	12.04 (10.03–16.16)	8.41 (7.05–9.53)	< 0.001
PEP, ms	108.73 ± 22.28	83.56 ± 16.76	< 0.001
PEPc, %	15.05 (11.25–18.76)	10.85 (8.85–12.65)	< 0.001
LVET, ms	287.90 ± 40.55	302.71 ± 37.17	0.052
LVETc, %	38.42 (33.59–45.54)	39.37 (35.44–43.41)	0.430
LVST, ms	293.50 (278.25–327.50)	318.00 (298.50–347.00)	0.006
LVSTc, %	41.39 (36.12–48.61)	42.92 (38.18–45.41)	0.501
PEP/LVET, %	38.81 (28.35–46.14)	26.71 (22.92–32.14)	< 0.001

Note: Data are presented as median (interquartile range), mean ± standard deviation, or number (percentage). Significant p‐values (< 0.05) are highlighted in bold.

Abbreviations: BSA, body surface area; EMAT, electromechanical activation time; EMATc, heart rate‐corrected EMAT; HR, heart rate; LVEF, left ventricular ejection fraction; LVET, left ventricular ejection time; LVST, left ventricular systolic time; PEP, pre‐ejection period; SBP, systolic blood pressure.

4.2.2. Cardiac Acoustic Biomarkers

4.2.2.1. Intergroup Differences

Evaluation of children aged 7–16 years demonstrated substantial differences in CABs between cases and controls. The case group showed significantly elevated values for EMAT (91.80 ± 17.60 vs. 64.95 ± 10.67 ms; p < 0.001), EMATc (12.04% [10.03–16.16] vs. 8.41% [7.05–9.53]; p < 0.001), PEP (108.73 ± 22.28 vs. 83.56 ± 16.76 ms; p < 0.001), PEPc (15.05% [11.25–18.76] vs. 10.85% [8.85–12.65]; p < 0.001), and the PEP/LVET ratio (38.81% [28.35–46.14] vs. 26.71% [22.92–32.14]; p < 0.001). LVST was significantly reduced in cases (293.50 [278.25–327.50] vs. 318.00 [298.50–347.00] ms; p = 0.006). No significant intergroup differences were observed for LVET, LVETc, or LVSTc (p > 0.05), as detailed in Table 2.

4.2.2.2. Correlation Analysis

Bivariate correlation analysis revealed strong associations between these CABs and LVEF. LVEF showed a pronounced inverse correlation with EMAT (r = −0.609, p < 0.001) and a moderate inverse correlation with EMATc (r = −0.558, p < 0.001). Negative correlations of lesser magnitude were observed with PEP (r = −0.483, p < 0.001), PEPc (r = −0.478, p < 0.001), and the PEP/LVET ratio (r = −0.472, p < 0.001). A positive correlation was identified between LVEF and LVST (r = 0.308, p < 0.001).

4.2.2.3. Logistic Regression Analysis

Binary logistic regression analysis incorporating all significant parameters identified EMAT as an independent relevant factor for LVSD (odds ratio [OR] = 1.138, 95% CI: 1.087–1.191; p < 0.001) through forward stepwise selection.

4.2.2.4. Diagnostic Performance of EMATc by ROC Curve Analysis

ROC curve analysis demonstrated excellent diagnostic performance for EMAT, with an AUC of 0.896 (95% CI: 0.825–0.966; p < 0.001). The optimal diagnostic threshold for LVSD was determined to be 82.50 ms, corresponding to a Youden's index of 0.757, which provides a sensitivity of 82.5% and a specificity of 93.2% (Figure 4).

FIGURE 4.

Receiver operating characteristic (ROC) curve of electromechanical activation Time (EMAT) in left ventricular systolic dysfunction (LVSD) diagnosis.

5. Discussion

To our knowledge, this study represents the first application of a novel integrated HS–ECG synchronization monitoring system for early detection of LVSD in pediatric populations. Our methodology incorporated ECG RR interval calibration to mitigate heart rate confounding and implemented rigorous age‐stratified analysis in a cohort of 212 children aged 3–16 years. Comparative assessment of CABs between LVSD patients and controls with normal cardiac function revealed significant alterations strongly correlated with impaired LVEF. A key finding was the age‐dependent diagnostic utility of these biomarkers: the heart rate–corrected EMATc emerged as the optimal diagnostic biomarker for LVSD in younger children (aged 3–6 years), whereas the uncorrected EMAT demonstrated superior diagnostic efficacy in older children (aged 7–16 years).

The cardiac cycle requires precise electromechanical coupling, initiated by an electrical stimulus that triggers myocardial activation and subsequent mechanical contraction. The EMAT, defined as the interval from QRS onset to the generation of sufficient ventricular pressure for mitral valve closure, serves as a quantitative measure of this coupling efficiency (Burkhoff et al. 2020). EMAT is influenced by preload, afterload, and the rate of left ventricular pressure rise. In LVSD, this interval is necessarily prolonged to achieve adequate mitral valve closure pressure (Trabelsi et al. 2020). Despite its established role in adult cardiology, comprehensive data on EMAT in pediatric populations remain globally scarce.

Our analysis revealed significant prolongation of EMAT in pediatric LVSD patients relative to matched controls, accompanied by a strong inverse correlation with LVEF, corroborating adult findings (Moyers et al. 2007). Notably, absolute EMAT values in children were shorter than in adults. In the 7–16 years cohort, mean EMAT was 91.80 ms in cases versus 159.82 ms in adults; controls similarly exhibited lower values (64.95 vs. 91.58 ms) (Li et al. 2020). A progressive, age‐dependent elongation of EMAT was observed, with median values of 70 ms (cases) and 60 ms (controls) in the 3–6 years subgroup, potentially reflecting age‐related cardiac dimensional scaling (Kampmann et al. 2000; Li et al. ²⁰¹⁰).

Within the 7–16 years cohort, EMAT demonstrated the strongest correlation with LVEF among all CABs and was identified as an independent relevant factor of LVSD through multivariate analysis. ROC curve analysis confirmed its high diagnostic accuracy, with an optimal cutoff value of 82.50 ms (AUC = 0.896; sensitivity: 82.5%; specificity: 93.2%). These results indicate that EMAT may serve as a clinically useful biomarker to facilitate the early diagnosis of LVSD in this specific age group.

To address heart rate dependence, we evaluated the rate‐corrected index, EMATc (calculated as EMAT/RR interval) (Shitara et al. 2021; Wang et al. ²⁰¹³; Sung et al. ²⁰²⁰; Chang et al. ²⁰²⁰; Zhang et al. ²⁰²⁰; Zhang and Liu ²⁰²²). A prior multicenter study demonstrated that the heart rate–corrected EMATc was significantly elevated in patients with heart failure with preserved ejection fraction (HFpEF) compared to hypertensive controls and was further increased in those with heart failure with reduced ejection fraction (HFrEF) (Wang et al. 2013). These collective findings indicate that EMATc not only assists in the diagnosis of HFrEF but also holds substantial diagnostic value in HFpEF, highlighting its broad applicability across heart failure phenotypes. Moreover, an international investigation revealed that EMATc predicts postdischarge adverse events in acute heart failure. Implementing a treatment target of EMATc < 15% resulted in significantly improved 1‐year clinical outcomes compared to a symptom‐guided approach (Sung et al. 2020).

Consistent with these observations, our results confirmed a significant correlation between EMATc and LVEF, which was most robust in the 3–6‐year cohort. In this age group, EMATc emerged as an independent diagnostically relevant factor for LVSD, with an optimal diagnostic threshold of 11.80% (AUC = 0.821; sensitivity: 82.5%; specificity: 84.7%). The enhanced diagnostic performance of EMATc in young children is likely attributable to their characteristically elevated heart rates and greater intrinsic heart rate variability, underscoring the particular clinical importance of rate correction in this pediatric demographic (Zhu et al. 2006).

The PEP refers to the interval from QRS onset to aortic valve opening, encompassing both EMAT and the isovolumetric contraction period. Prolongation of PEP indicates a reduced rate of left ventricular pressure rise and is observed in individuals with LVSD (Efstratiadis and Michaels 2008). In our study, both PEP and the heart rate–corrected index (PEPc) were significantly elevated in the case group compared to controls, consistent with previous findings (Reant et al. 2010). Although these parameters showed significant intergroup differences, subsequent comparative analysis revealed that their diagnostic performance was not superior to that of EMAT and its rate–corrected form (EMATc).

LVET reflects the duration of ventricular ejection and is closely associated with stroke volume. It is modulated by preload, afterload, left ventricular geometry, and contractility (Triposkiadis et al. 2018). Patients with HFrEF consistently exhibit shortened LVET (Alhakak et al. 2021). The underlying mechanism involves competing physiological influences: although prolongation of the PEP in LVSD delays ejection onset—potentially shortening LVET—a concomitant reduction in the velocity and extent of myocardial fiber shortening also occurs. The latter effect dominates, ultimately resulting in net shortening of LVET (Alhakak et al. 2021).

In our pediatric cohort, children aged 3–6 years with LVSD demonstrated significantly shorter LVET than controls, and LVET correlated significantly with LVEF. However, the rate‐corrected form (LVETc) showed no significant intergroup difference. No significant differences in LVET or LVETc were observed in the 7–16 years subgroup—a finding inconsistent with certain previous studies (Biering‐Sørensen et al. ²⁰¹⁸; Biering‐Sorensen ²⁰¹⁶). These discrepancies may stem from variations in ethnicity, age distribution, or underlying etiology differences. Consequently, the diagnostic performance of the PEP/LVET ratio was limited in our cohort.

LVST includes both the isovolumetric contraction period and LVET. In LVSD, isovolumetric contraction prolongs while LVET shortens; these opposing directional changes limit the clinical utility of LVST. Studies in adults have reported conflicting results: Kosmicki et al. found significantly shorter LVST in 215 patients with LVSD compared to 218 controls (Kosmicki et al. 2010), whereas a Chinese study (76 cases vs. 113 controls) reported no significant difference in LVST or its rate–corrected form (LVSTc) (Li et al. 2020). Similarly, a study involving pacemaker‐implanted patients showed no significant difference in LVST between those with and without LVSD, regardless of pacing site (Huang et al. 2022). In our study, LVST was significantly shorter in the case group, but its rate‐corrected form (LVSTc) showed no statistical difference. These inconsistencies may arise from population differences in race, age, sample size, clinical characteristics, or disease composition.

6. Study Limitations

This study has several important limitations. First, the sample size was constrained by the low overall incidence of LVSD in pediatric populations. Second, although we applied a heart rate correction method widely adopted in adults (dividing parameters by the ECG RR interval), the suitability of this approach for pediatric patients remains uncertain. Alternative correction strategies may better account for developmental variations in heart rate dynamics that are unique to children. Finally, because our analysis controlled for the influence of QRS duration on EMAT and PEP, we did not assess CABs in children with prolonged QRS duration—including those with intraventricular conduction delay, ventricular preexcitation, ventricular rhythms, or paced rhythms. We plan to focus specifically on this subgroup in subsequent investigations.

To more rigorously evaluate the efficacy and generalizability of this technology, we intend to initiate a multicenter study to expand both the sample size and age representation, implement comprehensive long‐term follow‐up, and systematically examine its feasibility and practical utility in remote patient management.

7. Conclusion

This first‐of‐its‐kind study establishes the clinical validity of a novel noninvasive synchronized HS–ECG monitoring system for assessing LVSD in pediatric populations. We identified EMAT and its heart rate–corrected derivative (EMATc) as robust, age‐stratified acoustic biomarkers for LVSD diagnosis, with EMATc demonstrating superior performance in younger children (aged 3–6 years) and EMAT showing greater efficacy in older children (aged 7–16 years). These results provide a practical, efficient, and scalable framework for early detection and risk stratification of cardiac dysfunction in children. The technology provides distinct advantages over conventional diagnostic methods, including full noninvasiveness, objective measurement, rapid assessment capability, and operational simplicity—making it suitable for both clinical and community‐based screening applications.

Author Contributions

Chun‐Mei Gao contributed to conceptualization, data curation, methodology, formal analysis, investigation, and writing. Fan Xie, Han Gao, Ye Chen, Hui Wang, Qiu‐Qin Xu, Xin‐Lu Hu, Ling Sun, Yun‐Jia Tang, Shu‐Hui Wang contributed to investigation and project administration. Hai‐Tao Lv contributed to conceptualization, investigation, and supervision. All authors read and approved the final manuscript.

Funding

This work was fully supported by the National Natural Science Foundation of China (Nos. 82171797, 82371806, 82270529, and 82470523).

Ethics Statement

The institutional review board of Children's Hospital of Soochow University approved this study (No: 2022CS181).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.