Heart age estimated using explainable advanced electrocardiography

Thomas Lindow, Israel Palencia-Lamela, Todd T Schlegel, Martin Ugander
doi: https://doi.org/10.1101/2021.12.08.21267378

Abstract

Background Electrocardiographic (ECG) Heart Age conveying cardiovascular risk has been estimated by both Bayesian and artificial intelligence approaches. We hypothesized that explainable measures from the 10-second 12-lead ECG could successfully predict Bayesian ECG Heart Age.

Methods Advanced analysis was performed on ECGs from healthy subjects and patients with cardiovascular risk or proven heart disease. Regression models were used to predict a Bayesian 5-minute ECG Heart Age from the standard resting 10-second 12-lead ECG. The difference between 10-second ECG Heart Age and chronological age was compared.

Results In total, 2,771 subjects were included (n=1682 healthy volunteers, n=305 with cardiovascular risk factors, n=784 with cardiovascular disease). Overall, 10-second Heart Age showed strong agreement with the 5-minute Heart Age (R2=0.94, p<0.001, mean±SD bias 0.0±5.1 years). The difference between 10-second ECG Heart Age and chronological age was 0.0±5.7 years in healthy individuals, 7.4±7.3 years in subjects with cardiovascular risk factors (p<0.001), and 14.3±9.2 years for patients with cardiovascular disease (p<0.001).

Conclusions ECG Heart Age can be accurately estimated from a 10-second 12-lead ECG in a transparent and explainable fashion based on known ECG measures, without artificial intelligence techniques. The difference between ECG Heart Age and chronological age increases markedly with cardiovascular risk and disease.

Background

Cardiovascular disease is a significant contributor to mortality, and pathological processes begin early and can progress silently for many years1-3. Many of the risk factors for developing cardiovascular disease are lifestyle-related4. Fortunately, several of these are modifiable, and risk can therefore be reduced for example by smoking cessation, dietary changes and increased physical activity5,6. To accomplish this, an individual must fully understand their risk and become motivated to act upon it. One way of communicating risks to the patient is to present the risk as a “Heart Age”, which can be contrasted to the patient’s chronological age. A Heart Age can either be determined by translating risk factor scores to what age that score would represent in an individual with no risk factors, or it can be based on electrocardiographic changes7-12. Describing the risk to the patient using Heart Age has been reported to reduce metabolic risk factors and may have the advantage of being easily understood by the patient11,13. A similar approach has been applied when conveying risk to smokers by describing how ‘old’ their lungs are, and such an approach increased the chance of smoking cessation14.

Moving beyond but not excluding basic ECG measurements such as heart rate and waveform amplitudes and durations, the diagnostic output from the ECG can be further substantially improved by using combinations of advanced ECG measures from 12-lead-ECG-derived vectorcardiography and waveform complexity15-17. An accurate ECG Heart Age using advanced, 5-minute, 12-lead ECG was developed in 2014, based on Bayesian statistics9. The Bayesian approach was transparent in so much as the advanced ECG measures contributing to the ECG Heart Age were well-described. After being trained on a set of healthy individuals, this approach yielded increased ECG Heart Ages for subjects at risk of cardiovascular disease and even higher ECG Heart Ages for those with established cardiovascular disease9. However, the reliance on 5-minute, high-fidelity 12-lead ECG recordings lessens the likelihood of widespread clinical use. If standard 10-second ECG recordings could be used instead, the clinical impact might be enhanced. Moreover, artificial intelligence has been used to estimate ECG Heart Age using the 10-second resting 12-lead ECG7,8,10. However, artificial intelligence techniques are limited by their “black box” approach, whereby the clinician does not have transparency as to the exact source(s) of the changes in the ECG that can affect an ECG Heart Age or other output18,19.Therefore, the aim of the study was to predict 5-minute ECG Heart Age from measures available by 10-second 12-lead ECG, and to compare the 10-second ECG Heart Age to chronological age in healthy subjects, subjects with cardiovascular risk factors, and patients with established cardiovascular disease. We hypothesized that 10-second 12-lead ECG recordings could accurately predict Bayesian ECG Heart Ages derived from 5-minute 12-lead ECG recordings.

Methods

A database of de-identified patients with both 5-minute and 10-second 12-lead ECG recordings was utilized for the study9. Within that database, healthy individuals, patients at cardiovascular risk, and patients with established cardiovascular disease were included. All healthy subjects were low-risk asymptomatic volunteers with absence of any cardiovascular or systemic disease, based on clinical history and physical examination. Exclusion criteria for the healthy subjects included increased blood pressure at physical examination (≥140/90 mm Hg), treatment for hypertension or diabetes, or active smoking. Patients with established cardiovascular disease were included based on the presence of either coronary heart disease (determined by coronary angiography with at least one obstructed vessel (≥50%) in at least one major native coronary vessel or coronary graft, or, if coronary angiography was either unavailable or clinically not indicated, one or more reversible perfusion defects on 99m-Tc-tetrofosmin single-photon emission computed tomography (SPECT)20-22), left ventricular hypertrophy (LVH) based on imaging evidence of at least moderate, concentric wall thickening according to guidelines of the American Society of Echocardiography23, left systolic dysfunction (left ventricular ejection fraction ≤50%) at echocardiography, cardiac magnetic resonance imaging (CMR) or SPECT, or findings suggestive of dilated/hypertrophic/ischemic cardiomyopathy at echocardiography or CMR16. Finally, subjects at cardiovascular risk were included based on the presence of cardiovascular risk factors such as hypertension or diabetes but no confirmed established cardiovascular disease9.

Based on the above, three groups of study participants were included: healthy subjects, subjects at cardiovascular risk, and patients with established cardiovascular disease. By methodological design, onlyhealthy subjectswere initially included when considering optimal measures of ECG available from 10-second 12-lead ECG for predicting the 5-minute ECG Heart Age. The 10-second ECG measures considered for the prediction model included: (1) From the conventional ECG: heart rate, R-to-R, P-wave, PR, QRS, QT, QTc, and TQ interval durations, as well as the conventional ECG amplitudes and axes; (2) From the transformation of the 12-lead ECG to the Frank X, Y and Z lead vectorcardiogram (VCG) via Kors’ transform24-28: the spatial means and peaks QRS-T angles, the spatial ventricular gradient and its individual QRS and T components, the spatial QRS- and T-wave axes (azimuths and elevations), waveform amplitudes and areas, including those in the three individual vectorcardiographic planes, and spatial QRS-and T-wave velocities; and (3) measures of QRS-and T-wave waveform complexity based on singular value decomposition after signal averaging29-31.

For all study participants, the difference between the 10-second ECG Heart Age and the chronological age was also calculated. Results were then compared between the three groups defined above. All participants gave written informed consent. All recordings were obtained under Institutional Review Board (IRB) approvals from NASA’s Johnson Space Center and partner hospitals that fall under IRB exemptions for previously collected and de-identified data. The study was performed in accordance with the Declaration of Helsinki.

Statistical analysis

Continuous variables were described using mean and standard deviation (SD). The chi-squared test was used to test for proportional differences between groups. Student’s t test was used to compare group means. The 10-second ECG Heart Age was first derived only in the healthy subject group through a process of feature selection, via optimized stepwise procedures, of measures available from 10-s, standard-fidelity ECG recordings, first using univariable linear regression, and finally as multivariable linear regression, to best predict the 5-minute ECG Heart Age9. The best model was defined as the model with highest R2 value that was also the most parsimonious, i.e., a model with statistically equal performance that incorporated a lesser number of measures was considered more parsimonious due to less proneness to over-fitting. The best model from the healthy volunteers was then applied forward to an expanded population that also included the other two groups (cardiovascular risk and cardiovascular disease, respectively), while not changing any of the included variables, and only allowing optimization of the included coefficients. The final sex-specific model was then applied across all three groups as the 10-secondECG Heart Age, and comparisons with 5-minute ECG Heart Age are presented as scatter plots and Bland-Altman plots. A two-sided p-value of 0.05 was used as to define statistical significance. Statistical analyses were performed using SAS JMP version 11.0, SAS Institute Inc, Cary, NC, USA, and R version 3.5.3R, R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/.

Results

In total, 2,771 patients were included (n=1682 healthy volunteers, n=305 subjects with cardiovascular risk factors, n=784 with cardiovascular disease). Baseline characteristics are presented in Table 1. The ECG measures included in the final prediction models are presented in Table 2 (males) and Table 3 (females).

View this table:
Table 1.

Baseline characteristics

View this table:
Table 2.

Measures included in the 10-second ECG Heart Age (male patients)

View this table:
Table 3.

Measures included in the 10-second ECG Heart Age (female patients)

The 10-second ECG Heart Age showed excellent agreement with the 5-minute Heart Age (R2=0.94, p<0.001, mean±SD bias 0.0±5.2 years), Figure 1. Agreement was strong for both males and females (R2=0.91, p<0.001, and R2=0.92, p<0.001 respectively). In healthy subjects, there was no difference in ECG Heart Age and chronological age (0.0±5.7 years). In subjects with cardiovascular risk factors, the difference was higher (7.4±7.3 years, p<0.001). Patients with cardiovascular disease showed the largest difference between ECG Heart Age and chronological age (14.3±9.2 years, p<0.001 when compared to subjects at cardiovascular risk), Figure 2.

Figure 1.
Figure 1.

Left panel: Scatter plot showing the relation between the 10-second ECG Heart Age and the 5-minute ECG Heart Age in all participants. The R2 value was 0.94 (p<0.001). Right panel: Bland-Altman plot showing the difference between the 10-second and 5-minute ECG Heart Age in relation to the mean of both ECG Heart Ages. The agreement between methods is strong, with minimal deviation from the identity line (dashed) or bias (0.0±5.2 years).

Figure 2.
Figure 2.

Panel A: The difference between the 10-second ECG Heart Age and chronological age in healthy subjects (left, dark green), subjects at cardiovascular (CV) risk (middle, light blue), and patients with CV disease (right, yellow).On average, there is no difference between ECG Heart Age and chronological age in healthy subjects. ECG Heart Age is higher in subjects at CV risk, and highest for those with overt CV disease.

Panel B: Scatter plots showing the relationship between the 10-second ECG Heart Age and chronological age in healthy subjects (left, dark green), subjects at CV risk (middle, light blue), and patients with CV disease (right, yellow). The dashed diagonal line is the identity line, i.e. indicating no difference between ECG Heart Age and chronological age.

Discussion

We found that ECG Heart Age based on measures of advanced ECG can be accurately predicted from standard resting 10-second 12-lead ECG recordings. This facilitates widespread use of ECG Heart Age in routine clinical settings, since neither specialized ECG machines nor unusually lengthy recordings are necessary. Also, if digital ECG raw data are available and the recording is of acceptable quality, ECG Heart Age can be retrospectively determined. Further, we found ECG Heart Age to be similar to chronological age in healthy individuals, while the ECG Heart Age was increasingly older with increasing cardiovascular disease status. This suggests that the ECG Heart Age is likely to provide accurate cardiovascular risk prediction, although validation in other datasets is necessary.

Explainability and transparency of variables that contribute to ECG Heart Age

For an estimation of Heart Age to be accurate in predicting an age that is similar to the chronological age when the heart is healthy, and increased when the heart is diseased, it is desirable that the included ECG measures change with age, and that the change is augmented with increasing cardiovascular risk or disease severity. Beyond age itself, the two ECG measures that had the strongest influence (highest t ratio) on the model were P-wave duration and spatial QT duration, and these measures fit this description well. P-wave duration increases with age32, and increased P-wave duration can be seen in advanced cardiovascular pathologies, e.g. heart failure and cardiac amyloidosis33. Similarly, QT duration increases with age34. Further, QT prolongation is associated with increased of cardiovascular risk, even beyond the rare long QT syndromes35, and with incrementally increased risk in advanced ages36. These general characteristics are also true for increased heart rate and for leftward shifting of the frontal plane QRS axis37-40. The other measures included in the score track changes in the vectorcardiographic QRS and T, and in T-wave complexity by singular value decomposition. Such changes are also known to occur in conditions associated with increased cardiovascular risk, such as hypertension and diabetes41, and in established cardiovascular disease, in which they often provide strong diagnostic and prognostic information24-27,29-31. Notably, these changes are not easily detectable by visual interpretation of a standard 12-lead ECG. How these ECG measures can affect the ECG Heart Age is exemplified in Figure 3. Taken together, the described ECG measures that contribute in a multivariable fashion to the ECG Heart Age all have physiologically reasonable associations with age and disease in a way that is transparent to the assessing clinician, thus providing important explainability to the model.

Figure 3.
Figure 3.

Example of the transparency and explainability of the ECG Heart Age from two subjects with equal chronological age but different ECG Heart Ages, illustrated by a commensurately aged female heart, but a disproportionately aged male heart. The ECG measures for each of the two patients are shown in the table in the middle of the figure, presented in the order of the relative strength (strongest first, based on t ratio [not shown]) of contribution to the ECG Heart Age. Notably, P-wave duration is markedly different between these two patients, and heart rate is higher for the male than the female, helping drive the ECG Heart Age higher in the male. The R-wave amplitude in lead Y is also much larger in the female, and the difference between the spatial ventricular gradient and the spatial mean QRS in the female is also larger, likely due to preserved T-wave amplitudes, contributing to her relatively younger ECG Heart Age. Furthermore, likely due to ischemic myocardial injuries, T-wave complexity is increased in the male, suggesting thatincreased myocardial repolarization heterogeneity also contributes to driving ECG Heart Age higher in the male, in spite of his shorter QT interval.

Differences compared to Bayesian 5-minute ECG Heart Age

The original, Bayesian 5-minute ECG Heart Age requires information from measures of beat-to-beat heart rate and QT variability,9 and of the root-mean square voltage or other aspects of high-frequency (high fidelity) components of the QRS complex42-44. However, 10-second-duration recordings of standard fidelity do not allow for such measures, and therefore they were not included in the 10-second ECG Heart Age. However, unlike the original Bayesian 5-minute ECG Heart Age, the 10-second ECG Heart Age should be derivable from any standard 12-lead ECG machine, as long as it is sufficiently equipped with software that can measure the included measures and calculate the 10-second ECG Heart Age. The presented 10-second ECG Heart Age might therefore be anticipated to contribute to more widespread clinical penetration and use.

Comparison with other Heart (or Vascular) Ages

Different means of expressing cardiovascular risk by translating it into a heart or vascular age have recently been published13. Attia, et al, showed that by using a deep neural network (DNN) artificial intelligence (AI) technique, a patient’s chronological age could be predicted, and that if the difference between the predicted and actual age was small, prognosis was good 8. When Heart Age by Attia et al’s technique was older than the chronological age, the risk of future death was increased7. This corresponds well to the findings in our study that ECG Heart Age increased with increasing burden of cardiovascular risk. However, the prognostic value of the ECG Heart Age presented in the current study requires additional validation. Furthermore, another AI method similar to that of Attia et al reported similarly encouraging results10. However, although the results of such AI studies are promising, DNN-based AI techniques are inherently problematic in several respects, especially in relation to their lack of transparency and explainability, i.e., the ‘black box’ of AI18,19.Without the ability to know the exact features of the 12-lead ECG that are most important in a given DNN model’s output, both interpretability and ethical accountability are compromised45. Moreover, it is effectively impossible for a clinician to identify, when critically evaluating the diagnostic output of a DNN-based AI model, the possible contribution to the result from methodological artifact or bias merely related to noise or to differing technical specifications between different ECG machines46. Alternatively, it is not possible to know if unanticipated results are possibly just related to excess dependency on the particular characteristics of a given DNN AI model’s training set47. In addition, a major flaw in both DNN-based AI models is that the age predictions were made using datasets including individuals with both cardiovascular risk factors and established disease8,10. For ECG Heart Age to be used as a marker of cardiovascular risk, it is imperative that ECG Heart Age agrees with chronological age in healthy populations, since it is the deviations from the line of identity in this relationship that form the basis of the assumed risk increase8.

Hence, we believe that the pursuit of an ECG Heart Age developed from heart-healthy subjects of varying ages, but without a black-box DNN or related AI methodology is valuable, and that the present results provide sufficient confirmation of accuracy to encourage further development. In addition, models in which the assessments were based on age predictions in healthy subjects will likely outperform models that were not. And finally, the use of more transparent regression models will also increase the ability of clinicians to better understand the origin of any unexpected result, and to thereafter relay it to the patient with a more convincing sense of trust and ethical accountability45.

Limitations

Using the same dataset for training and validation, our results can only be considered as proof-of-concept. Although the difference between ECG Heart Age and the subjects’ chronological age increased with increasing cardiovascular risk and/or disease, the prognostic value is unknown. Also, although ECG Heart Age was highly accurate, the precision cannot be reliably defined in this study. These aspects therefore need to be addressed in future studies.

Conclusion

We show that ECG Heart Age can be accurately, transparently, and explainably estimated from a standard 10-s, resting 12-lead ECG utilizing multiple, discrete conventional and advanced ECG measures. The difference between ECG Heart Age and subjects’ chronological age increases with increasing cardiovascular risk and disease. The prognostic value of our transparent and explainable 10-second ECG Heart Age requires prospective evaluation in future studies.

Data Availability

All data utilized in the present study are available upon reasonable request to the authors.

Footnotes

  • Fundings: TL is currently under the support of postdoctoral research grants from The Swedish Heart-Lung Foundation (grant no 20200553), the Swedish Cardiac Society, the Royal Swedish Academy of Sciences (grant no LM2019-0013), Women and Health Foundation, Region Kronoberg (grant no 8301), The Swedish Heart and Lung Association (grant no LKH1387), Swedish Association of Clinical Physiology, and the Scandinavian Society of Clinical Physiology; Nuclear Medicine. The study was funded in part by grants (PI Ugander) from New South Wales Health, Heart Research Australia, and the University of Sydney.

  • Competing interests: TL, IPL: None TTS is owner and founder of Nicollier-Schlegel SARL, which performs ECG interpretation consultancy using software that can quantify the advanced ECG measures used in the current study. TTS and MU are owners and founders of Advanced ECG Systems, a company that is developing commercial applications of advanced ECG technology used in the current study.

References

  1. ↵
    Bergström, G. et al. Prevalence of Subclinical Coronary Artery Atherosclerosis in the General Population. Circulation 144, 916–929, doi:10.1161/circulationaha.121.055340 (2021).
    OpenUrlCrossRefPubMed
  2. ↵
    Townsend, N. et al. Cardiovascular disease in Europe: epidemiological update 2016. Eur Heart J 37, 3232–3245, doi:10.1093/eurheartj/ehw334 (2016).
    OpenUrlCrossRefPubMed
  3. ↵
    Badimon, L. & Vilahur, G. Thrombosis formation on atherosclerotic lesions and plaque rupture. J Int Med 276, 618–632, doi:10.1111/joim.12296 (2014).
    OpenUrlCrossRefPubMed
  4. ↵
    Andersson, C. & Vasan, R. S. Epidemiology of cardiovascular disease in young individuals. Nat Rev Cardiol 15, 230–240, doi:10.1038/nrcardio.2017.154 (2018).
    OpenUrlCrossRefPubMed
  5. ↵
    Elliot, C. A. & Hamlin, M. J. Combined diet and physical activity is better than diet or physical activity alone at improving health outcomes for patients in New Zealand’s primary care intervention. BMC Publ Health 18, 230, doi:10.1186/s12889-018-5152-z (2018).
    OpenUrlCrossRef
  6. ↵
    Lim, S. S. et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380, 2224–2260, doi:10.1016/s0140-6736(12)61766-8 (2012).
    OpenUrlCrossRefPubMedWeb of Science
  7. ↵
    Ladejobi, A. O. et al. The 12-lead electrocardiogram as a biomarker of biological age. Eur Heart J - Dig Health 2, 379–389, doi:10.1093/ehjdh/ztab043 (2021).
    OpenUrlCrossRef
  8. ↵
    Attia, Z. I. et al. Age and Sex Estimation Using Artificial Intelligence From Standard 12-Lead ECGs. Circ Arrhythm Electrophysiol 12, e007284, doi:10.1161/circep.119.007284 (2019).
    OpenUrlCrossRef
  9. ↵
    Ball, R. L., Feiveson, A. H., Schlegel, T. T., Starc, V. & Dabney, A. R. Predicting “heart age” using electrocardiography. J Pers Med 4, 65–78, doi:10.3390/jpm4010065 (2014).
    OpenUrlCrossRefPubMed
  10. ↵
    Lima, E. M. et al. Deep neural network-estimated electrocardiographic age as a mortality predictor. Nature Commun 12, 5117, doi:10.1038/s41467-021-25351-7 (2021).
    OpenUrlCrossRef
  11. ↵
    Lopez-Gonzalez, A. A. et al. Effectiveness of the Heart Age tool for improving modifiable cardiovascular risk factors in a Southern European population: a randomized trial. Eur J Prev Cardiol 22, 389–396, doi:10.1177/2047487313518479 (2015).
    OpenUrlCrossRefPubMed
  12. ↵
    Soureti, A., Hurling, R., Murray, P., van Mechelen, W. & Cobain, M. Evaluation of a cardiovascular disease risk assessment tool for the promotion of healthier lifestyles. Eur J Cardiovasc Prev Rehab 17, 519–523, doi:10.1097/HJR.0b013e328337ccd3 (2010).
    OpenUrlCrossRefPubMed
  13. ↵
    Groenewegen, K. A. et al. Vascular age to determine cardiovascular disease risk: A systematic review of its concepts, definitions, and clinical applications. Eur J Prev Cardiol 23, 264–274, doi:10.1177/2047487314566999 (2016).
    OpenUrlCrossRefPubMed
  14. ↵
    Parkes, G., Greenhalgh, T., Griffin, M. & Dent, R. Effect on smoking quit rate of telling patients their lung age: the Step2quit randomised controlled trial. BMJ 336, 598–600, doi:10.1136/bmj.39503.582396.25 (2008).
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Maanja, M. et al. The electrical determinants of increased wall thickness and mass in left ventricular hypertrophy. J Electrocardiol 58, 80–86, doi:10.1016/j.jelectrocard.2019.09.024 (2020).
    OpenUrlCrossRef
  16. ↵
    Schlegel, T. T. et al. Accuracy of advanced versus strictly conventional 12-lead ECG for detection and screening of coronary artery disease, left ventricular hypertrophy and left ventricular systolic dysfunction. BMC Cardiovasc Disord 10, 28, doi:10.1186/1471-2261-10-28 (2010).
    OpenUrlCrossRefPubMed
  17. ↵
    Johnson, K. et al. Advanced Electrocardiography Identifies Left Ventricular Systolic Dysfunction in Non-Ischemic Cardiomyopathy and Tracks Serial Change over Time. J Cardiovasc Dev Dis 2, 93–107, doi:10.3390/jcdd2020093 (2015).
    OpenUrlCrossRef
  18. ↵
    Gladding, P. A., Hewitt, W. & Schlegel, T. T. Going deep with ecg and aortic stenosis: Touchdown or incomplete pass? J Am Heart Assoc 9, e016193–e016193, doi:10.1161/JAHA.120.016193 (2020).
    OpenUrlCrossRef
  19. ↵
    The Lancet Respiratory Medicine, L. R. M. Opening the black box of machine learning. Lance Respir Med 6, 801–801, doi:10.1016/S2213-2600(18)30425-9 (2018).
    OpenUrlCrossRef
  20. ↵
    Vrtovec, B., Sinkovec, M., Starc, V., Radovancevic, B. & Schlegel, T. T. Coronary artery disease alters ventricular repolarization dynamics in type 2 diabetes. Pacing Clin Electrophysiol 28 Suppl 1, S178–181, doi:10.1111/j.1540-8159.2005.00076.x (2005).
    OpenUrlCrossRef
  21. Persson, E., Carlsson, M., Palmer, J., Pahlm, O. & Arheden, H. Evaluation of left ventricular volumes and ejection fraction by automated gated myocardial SPECT versus cardiovascular magnetic resonance. Clin Physiol Funct Imaging25, 135–141, doi:10.1111/j.1475-097X.2005.00599.x (2005).
    OpenUrlCrossRefPubMed
  22. ↵
    Trägårdh, E. et al. High-frequency electrocardiogram analysis in the ability to predict reversible perfusion defects during adenosine myocardial perfusion imaging. J Electrocardiol40, 510–514, doi:10.1016/j.jelectrocard.2007.03.242 (2007).
    OpenUrlCrossRefPubMed
  23. ↵
    Lang, R. M. et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc maging16, 233–270, doi:10.1093/ehjci/jev014 (2015).
    OpenUrlCrossRefPubMed
  24. ↵
    Yamazaki, T., Froelicher, V. F., Myers, J., Chun, S. & Wang, P. Spatial QRS-T angle predicts cardiac death in a clinical population. Heart rhythm 2, 73–78, doi:10.1016/j.hrthm.2004.10.040 (2005).
    OpenUrlCrossRefPubMedWeb of Science
  25. Fayn, J., Rubel, P., Pahlm, O. & Wagner, G. S. Improvement of the detection of myocardial ischemia thanks to information technologies. Int J Cardiol 120, 172–180, doi:10.1016/j.ijcard.2006.09.025 (2006).
    OpenUrlCrossRef
  26. Kardys, I. et al. Spatial QRS-T angle predicts cardiac death in a general population. European heart journal24, 1357–1364, doi:10.1016/S0195-668X(03)00203-3 (2003).
    OpenUrlCrossRefPubMedWeb of Science
  27. ↵
    Horinaka, S. et al. Ventricular gradient variability. New ECG method for detection of ischemic heart disease. J Electrocardiol 28, 177–183, doi:10.1016/S0022-0736(05)80255-9 (1995).
    OpenUrlCrossRefPubMed
  28. ↵
    Kors, J. A., van Herpen, G., Sittig, A. C. & van Bemmel, J. H. Reconstruction of the Frank vectorcardiogram from standard electrocardiographic leads: diagnostic comparison of different methods. Eur Heart J 11, 1083–1092 (1990).
    OpenUrlPubMedWeb of Science
  29. ↵
    Zabel, M. et al. Analysis of T-wave morphology from the 12-lead electrocardiogram for prediction of long-term prognosis in male US veterans. Circulation 105, 1066–1070, doi:10.1161/hc0902.104598 (2002).
    OpenUrlAbstract/FREE Full Text
  30. Okin, P. M. et al. Repolarization Abnormality for Prediction of All-Cause and Cardiovascular Mortality in American Indians: The Strong Heart Study. J Cardiovasc Electrophysiol 16, 945–951, doi:10.1111/j.1540-8167.2005.40808.x (2005).
    OpenUrlCrossRefPubMedWeb of Science
  31. ↵
    Priori, S. G. et al. Evaluation of the spatial aspects of T-wave complexity in the long-QT syndrome. Circulation 96, 3006–3012, doi:10.1161/01.CIR.96.9.3006 (1997).
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Chhabra, L., Devadoss, R., Chaubey, V. K. & Spodick, D. H. Interatrial Block in the Modern Era. Curr Cardiol Rev10, 181–189, doi:10.2174/1573403X10666140514101748 (2014).
    OpenUrlCrossRefPubMed
  33. ↵
    Lindow, T. & Lindqvist, P. The Prevalence of Advanced Interatrial Block and Its Relationship to Left Atrial Function in Patients with Transthyretin Cardiac Amyloidosis. J Clin Med 10, doi:10.3390/jcm10132764 (2021).
    OpenUrlCrossRef
  34. ↵
    Rautaharju, P. M., Mason, J. W. & Akiyama, T. New age-and sex-specific criteria for QT prolongation based on rate correction formulas that minimize bias at the upper normal limits. Int J Cardiol 174, 535–540, doi:10.1016/j.ijcard.2014.04.133 (2014).
    OpenUrlCrossRefPubMed
  35. ↵
    Shah, S. R., Park, K. & Alweis, R. Long QT Syndrome: A Comprehensive Review of the Literature and Current Evidence. Curr Prob Cardiol 44, 92–106, doi:10.1016/j.cpcardiol.2018.04.002 (2019).
    OpenUrlCrossRef
  36. ↵
    Nielsen, J. B. et al. Risk prediction of cardiovascular death based on the QTc interval: evaluating age and gender differences in a large primary care population. Eur Heart J 35, 1335–1344, doi:10.1093/eurheartj/ehu081 (2014).
    OpenUrlCrossRefPubMed
  37. ↵
    Goorakani, Y. et al. Correlation of resting heart rate with anthropometric factors and serum biomarkers in a population-based study: Fasa PERSIAN cohort study. BMC Cardiovas Disord 20, 319–319, doi:10.1186/s12872-020-01594-y (2020).
    OpenUrlCrossRef
  38. Ogliari, G. et al. Resting heart rate, heart rate variability and functional decline in old age. Can Med Assoc J 187, E442–e449, doi:10.1503/cmaj.150462 (2015).
    OpenUrlAbstract/FREE Full Text
  39. Morcet, J. F., Safar, M., Thomas, F., Guize, L. & Benetos, A. Associations between heart rate and other risk factors in a large French population. J Hyperten s 17, 1671–1676, doi:10.1097/00004872-199917120-00003 (1999).
    OpenUrlCrossRefPubMedWeb of Science
  40. ↵
    Evans, J. G., Prior, I. A. & Tunbridge, W. M. Age-associated change in QRS axis: intrinsic or extrinsic ageing? Gerontology 28, 132–137, doi:10.1159/000212523 (1982).
    OpenUrlCrossRefPubMed
  41. ↵
    Bergfeldt, L. et al. Spatial peak and mean QRS-T angles: A comparison of similar but different emerging risk factors for cardiac death. J Electrocardiol61, 112–120, doi:10.1016/j.jelectrocard.2020.05.013 (2020).
    OpenUrlCrossRef
  42. ↵
    Ringborn, M. M. D. et al. Comparison of high-frequency QRS components and ST-segment elevation to detect and quantify acute myocardial ischemia. J Electrocardiol 43, 113–120, doi:10.1016/j.jelectrocard.2009.11.009 (2010).
    OpenUrlCrossRefPubMed
  43. Trägårdh, E., Pahlm, O., Wagner, G. S. & Pettersson, J. Reduced high-frequency QRS components in patients with ischemic heart disease compared to normal subjects. J Electrocardiol 37, 157–162, doi:10.1016/j.jelectrocard.2004.02.004 (2004).
    OpenUrlCrossRefPubMed
  44. ↵
    Schlegel, T. T. et al. Real-Time 12-Lead High-Frequency QRS Electrocardiography for Enhanced Detection of Myocardial Ischemia and Coronary Artery Disease. Mayo Clin Proc 79, 339–350, doi:10.4065/79.3.339 (2004).
    OpenUrlCrossRefPubMedWeb of Science
  45. ↵
    Yoon, C. H., Torrance, R. & Scheinerman, N. Machine learning in medicine: should the pursuit of enhanced interpretability be abandoned? J Med Ethics, doi:10.1136/medethics-2020-107102 (2021).
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Brisk, R. et al. The effect of confounding data features on a deep learning algorithm to predict complete coronary occlusion in a retrospective observational setting. Eur Heart J - Dig Health 2, 127–134, doi:10.1093/ehjdh/ztab002 (2021).
    OpenUrlCrossRef
  47. ↵
    Siontis, K. C. et al. Use of Artificial Intelligence Tools Across Different Clinical Settings: A Cautionary Tale. Circ Cardiovasc Qual Outcomes14, e008153, doi:10.1161/circoutcomes.121.008153 (2021).
    OpenUrlCrossRef
Back to top
Posted December 08, 2021.
Download PDF
Data/Code
Email
Heart age estimated using explainable advanced electrocardiography
Thomas Lindow, Israel Palencia-Lamela, Todd T Schlegel, Martin Ugander
medRxiv 2021.12.08.21267378; doi: https://doi.org/10.1101/2021.12.08.21267378
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area