Modular ECG Analysis System Research Articles

Feasibility and findings of electrocardiogram recording in older adults with intellectual disabilities: results of the Healthy Ageing and Intellectual Disabilities study.

Older adults with intellectual disabilities (ID) have a high risk of cardiovascular diseases (CVD). At the same time, challenging diagnostic work-up increases the likelihood of underdiagnosis of CVD in this population. To limit this underdiagnosis, it would be beneficial to use objective measures such as the electrocardiogram (ECG). However, little is known about the feasibility of ECG recording and the prevalence of ECG abnormalities in this population. Therefore, the aims of this study were to investigate the feasibility of resting ECG recording, to study the prevalence of ECG abnormalities, and to compare the frequency of ECG abnormalities with medical records in older adults with ID. A cross-sectional study was performed within a cohort of older adults (≥60years) with ID as part of the Healthy Ageing and Intellectual Disabilities (HA-ID) study. A resting 12-lead ECG was attempted, and the ECG recording was considered feasible if the recording could be made and if the ECG could be interpreted by a cardiologist and the Modular ECG Analysis System (MEANS). ECGs were assessed for the presence of ECG abnormalities and medical record review was performed. If the cardiologist or MEANS concluded that there was evidence of myocardial infarction, atrial fibrillation or QTc prolongation on the ECG in the absence of this ECG diagnosis in the participant's medical record, this was classified as a previously undiagnosed ECG diagnosis. ECG recording was feasible in 134 of the 200 participants (67.0%). Of these 134 participants (70.6±5.8years; 52.2% female), 103 (76.9%) had one or more ECG abnormality, with the most prevalent being prolonged P-wave duration (27.6%), QTc prolongation (18.7%), minor T-wave abnormalities (17.9%), first degree atrioventricular block (12.7%) and myocardial infarction (6.7%). Eight out of 9 (88.9%) myocardial infarctions and all cases of (significant) QTc prolongation (100%) were previously undiagnosed. This study showed that ECG recording is feasible in the majority of older adults with ID and revealed a substantial underdiagnosis of ECG abnormalities. These results stress the importance of ECG recording and warrant further research into the yield of opportunistic ECG screening in older adults with ID.

Impact of QRS misclassifications on heart-rate-variability parameters (results from the CARLA cohort study).

Heart rate variability (HRV), an important marker of autonomic nervous system activity, is usually determined from electrocardiogram (ECG) recordings corrected for extrasystoles and artifacts. Especially in large population-based studies, computer-based algorithms are used to determine RR intervals. The Modular ECG Analysis System MEANS is a widely used tool, especially in large studies. The aim of this study was therefore to evaluate MEANS for its ability to detect non-sinus ECG beats and artifacts and to compare HRV parameters in relation to ECG processing. Additionally, we analyzed how ECG processing affects the statistical association of HRV with cardiovascular disease (CVD) risk factors. 20-min ECGs from 1,674 subjects of the population-based CARLA study were available for HRV analysis. All ECGs were processed with the ECG computer program MEANS. A reference standard was established by experienced clinicians who visually inspected the MEANS-processed ECGs and reclassified beats if necessary. HRV parameters were calculated for 5-minute segments selected from the original 20-minute ECG. The effects of misclassified typified normal beats on i) HRV calculation and ii) the associations of CVD risk factors (sex, age, diabetes, myocardial infarction) with HRV were modeled using linear regression. Compared to the reference standard, MEANS correctly classified 99% of all beats. The averaged sensitivity of MEANS across all ECGs to detect non-sinus beats was 76% [95% CI: 74.1;78.5], but for supraventricular extrasystoles detection sensitivity dropped to 38% [95% CI: 36.8;38.5]. Time-domain parameters were less affected by false sinus beats than frequency parameters. Compared to the reference standard, MEANS resulted in a higher SDNN on average (mean absolute difference 1.4ms [95% CI: 1.0;1.7], relative 4.9%). Other HRV parameters were also overestimated as well (between 6.5 and 29%). The effect estimates for the association of CVD risk factors with HRV did not differ between the editing methods. We have shown that the use of the automated MEANS algorithm may lead to an overestimation of HRV due to the misclassification of non-sinus beats, especially in frequency domain parameters. However, in population-based studies, this has no effect on the observed associations of HRV with risk factors, and therefore an automated ECG analyzing algorithm as MEANS can be recommended here for the determination of HRV parameters.

Heart rate: an accessible risk indicator in adult congenital heart disease

BackgroundHigher resting heart rate has been described as a risk factor for adverse outcome in healthy individuals and cardiovascular patients. The aim of this study was to evaluate resting heart...

Comparison of Machine Learning Algorithms Using Manual/Automated Features on 12-Lead Signal Electrocardiogram Classification: A Large Cohort Study on Students Aged Between 6 to 18 Years Old.

An electrocardiogram (ECG) has been extensively used to detect rhythm disturbances. We sought to determine the accuracy of different machine learning in distinguishing abnormal ECGs from normal ones in children who were examined using a resting 12-Lead ECG machine, and we also compared the manual and automated measurement using the modular ECG Analysis System (MEANS) algorithm of ECG features. Altogether, 10745 ECGs were recorded for students aged 6 to 18. Manual and automatic ECG features were extracted for each participant. Features were normalized using Z-score normalization and went through the student's t-test and chi-squared test to measure their relevance. We applied the Boruta algorithm for feature selection and then implemented eight classifier algorithms. The dataset was split into training (80%) and test (20%) partitions. The performance of the classifiers was evaluated on the test data (unseen data) by 1000 bootstrap, and sensitivity (SEN), specificity (SPE), AUC, and accuracy (ACC) were reported. In univariate analysis, the highest performance was heart rate and RR interval in the manual dataset and heart rate in an automated dataset with AUC of 0.72 and 0.71, respectively. The best classifiers in the manual dataset were random forest (RF) and quadratic-discriminant-analysis (QDA) with AUC, ACC, SEN, and SPE equal to 0.93, 0.98, 0.69, 0.99, and 0.90, 0.95, 0.75, 0.96, respectively. In the automated dataset, QDA (AUC: 0.89, ACC:0.92, SEN:0.71, SPE:0.93) and stack learning (SL) (AUC:0.89, ACC:0.96, SEN:0.61, SPE:0.99) reached best performances. This study demonstrated that the manual measurement of 12-Lead ECG features had better performance than the automated measurement (MEANS algorithm), but some classifiers had promising results in discriminating between normal and abnormal cases. Further studies can help us evaluate the applicability and efficacy of machine-learning approaches for distinguishing abnormal ECGs in community-based investigations in both adults and children.

The prevalence and predictors of Stroke among atrial fibrillation patients attending at Gondar University Referral Hospital, Northwest Ethiopia

Introduction:Recently, stroke is becoming the major public health problem in developing countries including Ethiopia. Atrial fibrillation patients are the most vulnerable group for the occurrence of stroke. Knowing the predictors and being aware for it is important for preventing severe complications and death. Therefore, the aim of this study is to assess the prevalence of stroke and identifying significant predictors.Methods:A hospital-based cross-sectional study was performed from 1 December 2018 to 30 September 2019 at University of Gondar Referral Hospital. A total of 242 patients with atrial fibrillation were included in the study. Atrial fibrillation patients were diagnosed by using 12 lead electrocardiographic tracing, in addition to thorough medical history and physical examination and analyzed by the Modular ECG Analysis System (MEANS). We used Epi info 7 and SPSS version 22 software for data entry and analysis purpose, respectively. Both bivariable and multivariable binary logistic regression model were computed to show the relationship of dependent and independent variables.Result:The prevalence of stroke among atrial fibrillation patients was 19.4% (95% confidence interval (CI): 14.9–25.2). Patients with heart failure (adjusted odds ratio (AOR): 5.70, 95% CI: 2.50–13.24) and thyroid disorder (AOR: 4.98, 95% CI: 1.47–16.85) are at risk of developing stroke.Conclusion:The prevalence of stroke was higher compared with others studies. Patients with heart failure and thyroid disorders were the risk factor for the development of stroke; therefore, physicians and cardiologists may better to consider all these two disorders when they diagnose stroke in patients with atrial fibrillation.

The potential yield of ECG screening of hypertensive patients: the Utrecht Health Project

Several guidelines for hypertension and cardiovascular risk management recommend an ECG in hypertensive patients to improve risk prediction. We estimated the prevalence of clinically relevant ECG abnormalities and the number needed to screen (NNS) with a routine ECG to prevent the occurrence of one death in the next 10 years conditional on adequate treatment and follow-up. The study population consisted of 866 hypertensive participants recruited from the Utrecht Health Project (UHP), a dynamic population study in Utrecht. Baseline measurements included an ECG and the risk factors that enable a Systematic COronary Risk Evaluation (SCORE) risk estimation for each participant. ECGs were interpreted using Modular ECG Analysis System for computerized recognition of ECG abnormalities. NNS to prevent one death was computed by the reciprocal of the prevalence of the ECG abnormalities multiplied by number needed to treat to prevent one death when the ECG abnormality is managed according to the prevailing clinical guidelines. The population consisted of 54.2% men with a mean age of 53.2 years (SD 11.5). The prevalence of ECG abnormalities was 17.6 [n = 95% confidence interval (CI) 15.0-20.1]. Prevalence of atrial fibrillation or prior myocardial infarction was 2.1% (95%CI 1.1-3.0) and of other ECG abnormalities related to increased cardiovascular disease risk 15.4% (95%CI 13.1-17.9). NNS to prevent one death from cardiovascular disease within 10 years was estimated at 260 (95%CI 220-308). Our findings support the existing recommendations to routinely record an ECG in unselected hypertensive patients as the prevalence of relevant abnormalities is considerable and NNS to prevent one death is lower than that in other widely accepted tests.

Methodology of QT‐Interval Measurement in the Modular ECG Analysis System (MEANS)

QT prolongation as can be induced by drugs, signals the risk of life-threatening arrhythmias. The methodology of QT measurement in the modular ECG analysis system (MEANS) is described. In the simultaneously recorded leads of the standard 12-lead electrocardiogram (ECG), the QRS complexes are detected by a spatial velocity function. They are typed as dominant or nondominant, and a representative complex per lead is obtained by averaging over the dominant complexes. QRS onset and T end are determined by a template technique, and QT is measured. MEANS performance was evaluated on the 125 ECGs of the common standards for quantitative electrocardiography (CSE) multilead database, of which the waveform boundaries have been released. MEANS detected correctly all 1445 complexes of the CSE library, with one false-positive detection due to a sudden baseline jump. All dominant complexes were correctly typed. The average of the differences between MEANS and reference was less than 2 ms (=1 sample) for both QRS onset and T end, and 2.1 ms for QT duration. The standard deviation of the differences was 3.8, 8.4, and 10.4 ms, respectively. A standard deviation of 10.4 ms for QT measurement seems large when related to the regulatory requirement that a prolongation as small as 5 ms should be detected. However, QT variabilities as encountered in different individuals will be larger than when measured in one individual during pharmacological intervention. Finally, if the U wave is part of the total repolarization, then T and U form a continuum and the end of T becomes questionable.

Incidence of recognized and unrecognized myocardial infarction in men and women aged 55 and older: the Rotterdam Study

Contemporary data on the incidence of unrecognized myocardial infarction (MI) among subjects aged 55 and older are limited. We studied the incidence of recognized and unrecognized MI in the Rotterdam Study, a population-based cohort of men and women aged 55 and older. The baseline examination was performed during 1990-93, with follow-up examinations during 1994-95, and 1997-2000. Baseline and follow-up 12-lead ECGs were analysed by the Modular ECG Analysis System. The 5148 participants who had no evidence of prevalent infarction were the subjects for analysis. Incident recognized infarction was defined as the occurrence of a fatal or non-fatal event coded as I21 according to the International Classification of Diseases, 10th edition. A repeat ECG was available in 4187 subjects. An unrecognized infarction was considered to have occurred if there was electrocardiographic evidence in the absence of a clinically recognized event. During a median follow-up of 6.4 years, 141 incident recognized infarctions occurred and the incidence rate of this event was 5.0 per 1000 person years. The incidence was higher in men (8.4) than in women (3.1). The incidence rate of unrecognized infarction was 3.8 per 1000 person years. Men (4.2) and women (3.6) had approximately similar incidence. Hence, the proportion of unrecognized infarction was lower in men (33%) than in women (54%). This difference in proportion of unrecognized infarctions was independent of age. A high proportion of incident MIs remains clinically unrecognized. As a history of MI is associated with an increased risk of repeat cardiovascular complications, our data suggest a need for periodical electrocardiographic screening to recognize (prevalent) infarctions and to install effective preventive treatment in those aged 55 and older.

The Electrical T-Axis and the Spatial QRS-T Angle Are Independent Predictors of Long-Term Mortality in Patients Admitted with Acute Ischemic Chest Pain

Objective: To investigate whether the orientation of the electrical T-axis and the spatial QRS-T angle provide independent diagnostic and prognostic information in patients presenting with acute chest pain. Methods: Patients with symptoms suggestive of acute cardiac pathology, who were seen by a general practitioner and for whom a prehospital electrocardiogram (ECG) was recorded by the ambulance service between 1992 and 1994 were investigated. The ECGs (n = 2,261) for our study population were stored for off-line analysis by the Modular ECG Analysis System. QRS- and T-axes were computed from the reconstructed vectorcardiographic X, Y and Z leads. During the year 2000, a follow-up of the entire cohort was performed, and the vital status of the patients was determined via the civil registrar’s office. Cox multivariable regression analyses were performed to evaluate the relation between the orientation of the T-axis, the spatial QRS-T angle and long-term mortality. Results: An abnormal orientation of the T-axis and the spatial QRS-T angle were associated with an increased likelihood of cardiac diseases and an increased risk of all-cause mortality during short- and long-term follow-up. Conclusions: We conclude that the frontal T-axis and the spatial QRS-T angle are important determinants of diagnosis and prognosis in patients presenting with acute chest pain. The reintroduction of vectorcardiography in routine clinical practice might therefore be reconsidered.

Comparison of computer-assigned Minnesota Codes with the visual standard method for new coronary heart disease events.

The Minnesota Code is the most widely used electrocardiogram (ECG) classification system for epidemiologic studies and has been incorporated into several Computer algorithms. The authors compared the Modular ECG Analysis System (MC-MEANS) and NOVACODE computer ECG findings with the Visual coding standard for agreement and prognostic associations with coronary heart disease (CHD) events occurring during follow-up from 1987 to 1995 in 2,116 individuals participating in the Atherosclerosis Risk in Communities (ARIC) Study. The exact agreement between Visual and computer findings was greater than 90% for all Minnesota Code categories except Q-code, which was 77% for MC-MEANS and 81% for NOVACODE. Approximately 60% of all Q-codes were assigned by computer methods only. Among the 2,116 participants, there were 246 (11.6%) new coronary events. Unadjusted relative risks for codes assigned by the three methods were similar. When computer methods disagreed on code severity, the CHD occurrence rates for MC-MEANS-detected severer code versus NOVACODE-detected severer code were 21% and 7%, respectively. This study provides clear evidence that computers assign more and severer Minnesota Codes with similar prognostic importance as does the Visual method; it also alerts researchers to potential problems in pooling Minnesota Code data read by different methods.

Minimal lead sets for reconstruction of 12-lead electrocardiograms

It may not always be possible to record all precordial leads of the standard 12-lead electrocardiogram (ECG). Especially in monitoring situations, a minimal lead set from which the 12-lead ECG can be reconstructed, would be valuable. This article assesses how well missing precordial leads could be synthesized from the remaining leads of the 12-lead ECG. A total of 2,372 diagnostic 12-lead ECG recordings were obtained from subjects with chest pain suggestive for acute myocardial infarction. Representative average beats were computed from the digital 12-lead ECG recordings with our Modular ECG Analysis System. The recordings were divided into a learning set and a test set. We considered all lead sets with one or more precordial leads removed, but always including limb leads I and II. By using the learning set, general reconstruction coefficients were computed to synthesize the missing precordial leads to each lead set. Performance of the synthesis was assessed by cross correlation between the original and the reconstructed leads. Also, patient-specific reconstruction coefficients were derived for each ECG in the test set and correlations were determined. High correlation coefficients were found with both reconstruction techniques. For different sizes of lead sets, the best patient-specific reconstructions had higher correlation values than the general reconstructions. For example, when 2 precordial leads were excluded, the best patient-specific median correlation was 0.994 compared to 0.963 for the best general reconstruction correlation. General reconstruction allows synthesis of 2 or 3 excluded precordial leads in good approximation. When patient-specific reconstruction can be applied, a minimal lead set including the limb leads and only 2 precordial leads suffices.

Both decreased and increased heart rate variability on the standard 10-second electrocardiogram predict cardiac mortality in the elderly: the Rotterdam Study.

Decreased heart rate variability has been associated with an adverse prognosis in patients after myocardial infarction. Studies carried out in the population at large show contradictory results. The authors examined the association between heart rate variability on a standard 10-second electrocardiogram and cardiac and all-cause mortality in the Rotterdam Study, a population-based cohort study of men and women aged > or =55 years, using data collected between 1990 and 1996 (mean follow-up = 4 years). Heart rate variability, taken as the standard deviation of normal R-R intervals (SDNN), was computed by means of the Modular ECG Analysis System. After exclusion of subjects with arrhythmia and those with fewer than six normal R-R intervals, the study population consisted of 2,088 men and 3,184 women. Cox's proportional hazards model was used to examine the age- and sex-adjusted risk for cardiac, noncardiac, and total mortality in relation to quartiles of SDNN, using the third quartile of SDNN as the reference category. Subjects in the lowest quartile of SDNN relative to those in the third quartile had an 80 percent age- and sex-adjusted increased risk for cardiac mortality (hazard ratio = 1.8; 95% confidence interval: 1.0, 3.2). Interestingly, for subjects in the highest quartile of SDNN, an even more pronounced risk for cardiac mortality was present (hazard ratio = 2.3; 95% confidence interval: 1.3, 4.0). Additional adjustment for possible confounders did not materially change the risk estimates. The authors conclude that heart rate variability measured on the standard 10-second electrocardiogram can be used to identify older men and women with an increased risk for cardiac mortality. In the elderly, increased heart rate variability is an even stronger indicator of cardiac mortality than decreased heart rate variability. Further studies are needed to confirm these findings and to elucidate their physiologic meaning.

Prolonged QT interval predicts cardiac and all-cause mortality in the elderly. The Rotterdam Study.

To examine the association between heart-rate corrected QT prolongation and cardiac and all-cause mortality in the population-based Rotterdam Study among men and women aged 55 years or older and to compare the prognostic value of the QT interval, using different formulas to correct for heart rate. After exclusion of participants with arrhythmias or bundle branch block on the ECG, the study population consisted of 2083 men and 3158 women. The QT interval was computed by the Modular ECG Analysis System (MEANS). Data were analysed using Cox' proportional hazards model. Participants in the highest quartile of the heart-rate corrected QT interval had about a 70% age- and sex-adjusted increased risk for both all-cause mortality (hazard ratio (HR) 1.8; 95% CI:1.3-2.4) and cardiac mortality (HR 1.7; 95% CI:1.0-2.7) compared to those in the lowest quartile. In women, the increased risk associated with prolonged QT for cardiac death was more pronounced than in men. These risk estimates did not change after adjustment for potential confounders, including history of myocardial infarction, hypertension and diabetes mellitus. A prolonged heart-rate corrected QT interval is an independent predictor for cardiac and all-cause mortality in older men and women. The risk associated with prolonged QT is hardly affected by the heart-rate correction formula used.

Prolonged QT Interval: A Tricky Diagnosis?

Prolonged heart-rate adjusted QT intervals on the electrocardiogram (ECG) are associated with an increased risk for coronary heart disease and sudden death. However, the diagnosis of the prolonged QT interval is hampered by lack of standards. We studied variations in the prevalence of prolonged QT, based on different common definitions, in a large nonhospitalized population, and compared our results with other studies applying the same definitions. The study population consisted of 2,200 men and 3,366 women participants of the Rotterdam Study, ≥55 years old. The QT interval was computed by our Modular ECG Analysis System (MEANS). Three different formulas to adjust QT for heart rate were used: Bazett’s formula (QTc), a linear regression equation (QTlr), and the QT index (QTI). Prolonged QT occurred frequently in both men and women, and its prevalence increased with age. Women had longer heart-rate adjusted QT intervals than men (mean QTc 433 ms vs 422 ms), and mean values for QTlr were lower than for QTc (mean QTlr 422 ms in women and 412 ms in men). Prevalence was highest for prolonged QTlr (31% in men and 26% in women) and lowest for prolonged QTI (6% in men and 9% in women). Comparison with other studies applying the same correction formulas showed large discrepancies in prevalence estimates of prolonged QTc and QTlr, and to a lesser degree of prolonged QTI, possibly due to differences in measurement techniques. Future research is needed to relate QT interval to prognosis, to obtain measurement technique specific reference values of heart-rate adjusted QT measurements, and to obtain age- and sex-specific threshold values for prolonged QT. Such data are needed to use the QT interval with confidence.

Improvement off automated electrocardiographic diagnosis by combination of computer interpretations of the electrocardiogram and vectorcardiogram

In the international project “Common Standards for Quantitative Electrocardiography” (CSE), diagnostic results of different computer programs for the interpretation of the electrocardiogram (ECG) and of the vectorcardiogram (VCG) were combined, and it was shown that the “combined program” performs better than each program separately. Because the program MEANS (Modular ECG Analysis System) comprises 2 different classification programs — one for the ECG, the other for the VCG — this allowed investigation of whether the combined interpretations would yield a better diagnostic result than either one separately. This approach requires that a VCG always be recorded in addition to the ECG. To circumvent this complication, the VCG was reconstructed from the simultaneously recorded ECG leads. This reconstructed VCG was then interpreted by the VCG classification program, whereupon the diagnostic interpretations of the ECG and the reconstructed VCG were combined. For the validation, the CSE database of documented ECGs and VCGs (n = 1,220) was used. The combination of the ECG and VCG interpretations yielded a better diagnostic result than each interpretation program separately (total accuracy 74.2% (ECG + VCG) vs 69.8% (ECG) and 70.2% (VCG), p < 0.001 in both cases). The results for the reconstructed VCG (total accuracy 70.5%) are comparable to those for the ECG and the VCG (p > 0.10 in both cases). The performance of the combined interpretations of ECG and reconstructed VCG (total accuracy 73.6%) is approximately the same as that of the combined ECG and VCG (p > 0.10). Thus, the performance of an ECG computer program can be improved by incorporating both ECG and VCG classificatory knowledge, using only the ECG itself.

Methodology of the Modular ECG Analysis System MEANS

The methodology, used in the Modular ECG Analysis System (MEANS) is described. MEANS consists of modules for signal analysis and diagnostic classification. The basic structure of the modular interpretation system remained intact over a period of 20 years, while all modules underwent many changes as a function of experience and insight, and the continuously changing information technology. The article describes the advantages of a modular approach to decision-support systems, the most important ones being easier maintenance of the software package and separate optimization and testing of each module. The overall evaluation of MEANS was done in the CSE study. Evaluation results for modules and for the entire system are presented.

Multilead ECG analysis

This paper describes the results of our recent research in computer-assisted ECG VCG interpretation. It comprises new developments which were initiated by the advent of relatively inexpensive microcomputers. Our previous systems performed an off-line analysis of ECGs. Currently, there is a trend to move computer power near to the patient and to provide on-line analysis of ECGs. Besides the advantage of the direct availability of the ECG interpretation, quality control will reduce the number of uninterpretable ECGs and hence the number of repeated recordings. This paper describes the requirements that were established for a system for on-line ECG analysis. The system is based on our modular approach just like our off-line system, Modular ECG A Nalysis System (MEANS). Changes in the Methods and software had to be made mainly because of the simultaneity of all ECG leads and the concurrency of the processing tasks. Other modifications and extensions of the algorithms necessary to meet the requirements of on-line ECG interpretation especially those related to processing speed, are discussed, and evaluation results are presented

The advantage of modular software design in computerized ECG analysis.

In the early 1970s, modular software design was advocated as a means to obtain better structured and more reliable software systems. Such a modular approach was also introduced for computerized ECG processing in 1972. The TNO Modular ECG/VCG interpretation system was developed along these lines. During the last decade, the algorithms of this system were thoroughly improved and evaluated. This resulted in 1983 in a new version of this system named MEANS (Modular ECG ANalysis System). This paper outlines the ideas of the modular approach as they were originally defined in 1972 and developed over the years. The usefulness and advantages of these ideas will be discussed based on 12 years of experience. It appeared that modularity is the ideal approach to the development and maintenance of complex systems such as for ECG/VCG interpretation.