Detection of abnormal left ventricular geometry in patients without cardiovascular disease through machine learning: An ECG-based approach

A review of evaluation approaches for explainable AI with applications in cardiology

Explainable artificial intelligence (XAI) elucidates the decision-making process of complex AI models and is important in building trust in model predictions. XAI explanations themselves require evaluation as to accuracy and reasonableness and in the context of use of the underlying AI model. This review details the evaluation of XAI in cardiac AI applications and has found that, of the studies examined, 37% evaluated XAI quality using literature results, 11% used clinicians as domain-experts, 11% used proxies or statistical analysis, with the remaining 43% not assessing the XAI used at all. We aim to inspire additional studies within healthcare, urging researchers not only to apply XAI methods but to systematically assess the resulting explanations, as a step towards developing trustworthy and safe models.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10462-024-10852-w.

Searching for the Best Machine Learning Algorithm for the Detection of Left Ventricular Hypertrophy from the ECG: A Review

Background: Left ventricular hypertrophy (LVH) is a powerful predictor of future cardiovascular events. Objectives: The objectives of this study were to conduct a systematic review of machine learning (ML) algorithms for the identification of LVH and compare them with respect to the classical features of test sensitivity, specificity, accuracy, ROC and the traditional ECG criteria for LVH. Methods: A search string was constructed with the operators "left ventricular hypertrophy, electrocardiogram" AND machine learning; then, Medline and PubMed were systematically searched. Results: There were 14 studies that examined the detection of LVH utilizing the ECG and utilized at least one ML approach. ML approaches encompassed support vector machines, logistic regression, Random Forest, GLMNet, Gradient Boosting Machine, XGBoost, AdaBoost, ensemble neural networks, convolutional neural networks, deep neural networks and a back-propagation neural network. Sensitivity ranged from 0.29 to 0.966 and specificity ranged from 0.53 to 0.99. A comparison with the classical ECG criteria for LVH was performed in nine studies. ML algorithms were universally more sensitive than the Cornell voltage, Cornell product, Sokolow-Lyons or Romhilt-Estes criteria. However, none of the ML algorithms had meaningfully better specificity, and four were worse. Many of the ML algorithms included a large number of clinical (age, sex, height, weight), laboratory and detailed ECG waveform data (P, QRS and T wave), making them difficult to utilize in a clinical screening situation. Conclusions: There are over a dozen different ML algorithms for the detection of LVH on a 12-lead ECG that use various ECG signal analyses and/or the inclusion of clinical and laboratory variables. Most improved in terms of sensitivity, but most also failed to outperform specificity compared to the classic ECG criteria. ML algorithms should be compared or tested on the same (standard) database.

Prognostic implications of heart failure stages among Chinese community populations: insight from a nationwide population-based study

Background

In light of high burden of heart failure (HF) in China, studies of prognostic implication of HF stages are important. We aimed to evaluate the relationship between HF stages and mortality risk in Chinese community populations.

Methods

Nationwide representative populations aged ≥35 years (n = 23,284, mean age 56.9 years, women 53.2%) were enrolled from 2012 to 2016. According to the international HF guidelines, participants were divided into stage A, B and C, and those who did not qualify these stages were categorized as apparently-healthy group. Association between HF stages and all-cause, cardiovascular [CV] and non-CV death was evaluated using multivariable-adjusted Cox proportional regression analysis.

Findings

During a median follow-up of 4.7 years (109,902.8 person-years), 1314 deaths occurred. Age-adjusted incidence rate of all-cause death was 5.3 in apparently-healthy, 7.8 in stage A, 8.6 in stage B and 24.6 in stage C groups per 1000 person-years. In reference to apparently-healthy group, adjusted hazard ratio for all-cause death was 1.90 (95% CI: 1.47-2.45), 2.43 (95% CI: 1.89-3.13) and 6.40 (95% CI: 4.56-8.99) for stage A, B and C. Advancing HF stages were associated with increasing risks for all-cause, CV and non-CV death (P-trend <0.05). For all-cause death, population attributable fraction due to stage A, B and C were 21.2%, 33.4% and 4.9%, accounting for 1,933,385, 3,045,993 and 446,867 deaths in China in 2018.

Interpretation

Advancing HF stages were associated with increasing risk mortality. Development and implementation of early screening and targeted interventions are urgently needed to reduce HF burdens in China.

Funding

This work was supported by the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (grant 2017-I2M-1-004), the Projects in the Chinese National Science & Technology Pillar Program during the Twelfth Five-year Plan Period (No.: 2011BAI11B01), and the Project Entrusted by the National Health Commission of the People's Republic of China (NHC2020-609).

T-wave inversion through inhomogeneous voltage diffusion within the FK3V cardiac model

The heart beats are due to the synchronized contraction of cardiomyocytes triggered by a periodic sequence of electrical signals called action potentials, which originate in the sinoatrial node and spread through the heart's electrical system. A large body of work is devoted to modeling the propagation of the action potential and to reproducing reliably its shape and duration. Connection of computational modeling of cells to macroscopic phenomenological curves such as the electrocardiogram has been also intense, due to its clinical importance in analyzing cardiovascular diseases. In this work, we simulate the dynamics of action potential propagation using the three-variable Fenton-Karma model that can account for both normal and damaged cells through a the spatially inhomogeneous voltage diffusion coefficient. We monitor the action potential propagation in the cardiac tissue and calculate the pseudo-electrocardiogram that reproduces the R and T waves. The R-wave amplitude varies according to a double exponential law as a function of the (spatially homogeneous, for an isotropic tissue) diffusion coefficient. The addition of spatial inhomogeneity in the diffusion coefficient by means of a defected region representing damaged cardiac cells may result in T-wave inversion in the calculated pseudo-electrocardiogram. The transition from positive to negative polarity of the T-wave is analyzed as a function of the length and the depth of the defected region.

ECG-based data-driven solutions for diagnosis and prognosis of cardiovascular diseases: A systematic review

Cardiovascular diseases (CVD) are a leading cause of death globally, and result in significant morbidity and reduced quality of life. The electrocardiogram (ECG) plays a crucial role in CVD diagnosis, prognosis, and prevention; however, different challenges still remain, such as an increasing unmet demand for skilled cardiologists capable of accurately interpreting ECG. This leads to higher workload and potential diagnostic inaccuracies. Data-driven approaches, such as machine learning (ML) and deep learning (DL) have emerged to improve existing computer-assisted solutions and enhance physicians' ECG interpretation of the complex mechanisms underlying CVD. However, many ML and DL models used to detect ECG-based CVD suffer from a lack of explainability, bias, as well as ethical, legal, and societal implications (ELSI). Despite the critical importance of these Trustworthy Artificial Intelligence (AI) aspects, there is a lack of comprehensive literature reviews that examine the current trends in ECG-based solutions for CVD diagnosis or prognosis that use ML and DL models and address the Trustworthy AI requirements. This review aims to bridge this knowledge gap by providing a systematic review to undertake a holistic analysis across multiple dimensions of these data-driven models such as type of CVD addressed, dataset characteristics, data input modalities, ML and DL algorithms (with a focus on DL), and aspects of Trustworthy AI like explainability, bias and ethical considerations. Additionally, within the analyzed dimensions, various challenges are identified. To these, we provide concrete recommendations, equipping other researchers with valuable insights to understand the current state of the field comprehensively.

Echocardiographic artificial intelligence for pulmonary hypertension classification

OBJECTIVE

The classification of pulmonary hypertension (PH) is crucial for determining the appropriate therapeutic strategy. We investigated whether machine learning (ML) algorithms may assist in echocardiographic PH prediction, where current guidelines recommend integrating several different parameters.

METHODS

We obtained physical and echocardiographic data from 885 patients who underwent right heart catheterisation (RHC). Patients were classified into three groups: non-PH, precapillary PH and postcapillary PH, based on values obtained from RHC. Using 24 parameters, we created predictive models employing four different classifiers and selected the one with the highest area under the curve. We then calculated the macro-average classification accuracy for PH on the derivation cohort (n=720) and prospective validation data set (n=165), comparing the results with guideline-based echocardiographic assessment obtained from each cohort.

RESULTS

Logistic regression with elastic net regularisation had the highest classification accuracy, with areas under the curves of 0.789, 0.766 and 0.742 for normal, precapillary PH and postcapillary PH, respectively. The ML model demonstrated significantly better predictive accuracy than the guideline-based echocardiographic assessment in the derivation cohort (59.4% vs 51.6%, p<0.01). In the independent validation data set, the ML model's accuracy was comparable to the guideline-based PH classification (59.4% vs 57.8%, p=0.638).

CONCLUSIONS

This preliminary study suggests promising potential for our ML model in predicting echocardiographic PH. Further research and validation are needed to fully assess its clinical utility in PH diagnosis and treatment decision-making.

Electrocardiogram Interpretation Using Artificial Intelligence: Diagnosis of Cardiac and Extracardiac Pathologic Conditions. How Far Has Machine Learning Reached?

Artificial intelligence (AI) is already widely used in different fields of medicine, making possible the integration of the paraclinical exams with the clinical findings in patients, for a more accurate and rapid diagnosis and treatment decision. The electrocardiogram remains one of the most important, fastest, cheapest, and noninvasive methods of diagnosis in cardiology, despite the rapid development and progression of the technology. Even if studied a long time ago, it still has a lot of less understood features that, with a better understanding, can give more clues to a correct and prompt diagnosis in a short time. The use of AI in the interpretation of the ECG improved the accuracy and the time to diagnosis in different cardiovascular diseases, and more than this, explaining the decision to make AI diagnosis improved the human understanding of the different features of the ECG that might be considered for a more accurate diagnosis. The purpose of this article is to provide an overview of the most recently published articles about the use of AI in ECG interpretation.

Predicting left ventricular hypertrophy from the 12-lead electrocardiogram in the UK Biobank imaging study using machine learning

Aims

Left ventricular hypertrophy (LVH) is an established, independent predictor of cardiovascular disease. Indices derived from the electrocardiogram (ECG) have been used to infer the presence of LVH with limited sensitivity. This study aimed to classify LVH defined by cardiovascular magnetic resonance (CMR) imaging using the 12-lead ECG for cost-effective patient stratification.

Methods and results

We extracted ECG biomarkers with a known physiological association with LVH from the 12-lead ECG of 37 534 participants in the UK Biobank imaging study. Classification models integrating ECG biomarkers and clinical variables were built using logistic regression, support vector machine (SVM) and random forest (RF). The dataset was split into 80% training and 20% test sets for performance evaluation. Ten-fold cross validation was applied with further validation testing performed by separating data based on UK Biobank imaging centres. QRS amplitude and blood pressure (P < 0.001) were the features most strongly associated with LVH. Classification with logistic regression had an accuracy of 81% [sensitivity 70%, specificity 81%, Area under the receiver operator curve (AUC) 0.86], SVM 81% accuracy (sensitivity 72%, specificity 81%, AUC 0.85) and RF 72% accuracy (sensitivity 74%, specificity 72%, AUC 0.83). ECG biomarkers enhanced model performance of all classifiers, compared to using clinical variables alone. Validation testing by UK Biobank imaging centres demonstrated robustness of our models.

Conclusion

A combination of ECG biomarkers and clinical variables were able to predict LVH defined by CMR. Our findings provide support for the ECG as an inexpensive screening tool to risk stratify patients with LVH as a prelude to advanced imaging.

Are Electrocardiographic Criteria Reliable for Left Ventricular Hypertrophy Detection in Indian Adults?

BACKGROUND

Left ventricular hypertrophy (LVH) detection is vital to the risk stratification of adults at risk of adverse cardiovascular events such as coronary heart disease, cerebrovascular disease, and aortic aneurysms. Electrocardiogram (ECG), a non-invasive, cost-effective instrument has been widely used as a screening tool for LVH. The objective of this study was to determine the diagnostic accuracy of seven frequently used ECG criteria in high-risk Indian adults in comparison with echocardiography.

METHODS

ECG and transthoracic echocardiography were performed in adults older than 18 years with at least one cardiac risk factor (chronic hypertension, obesity, ischemic heart disease, and type 2 diabetes mellitus). Precision and accuracy were calculated for the various ECG criteria against LVH based on left ventricular mass index (LVMI) and cardiac remodeling by echocardiography.

RESULTS

A total of 220 participants were enrolled. Of these, 96 had LVH by echocardiography. There was marked variability in LVH detection by the different ECG criteria: 28 by Sokolow-Lyon criteria, 26 by Cornell criteria, 24 by Lewis criteria, 46 by Scott criteria, eight by Romhilt-Estes criteria, six by Modified Cornell criteria, and only two by Roberts criteria. Agreement statistics between ECG criteria and LVMI showed that none of them had a good agreement for LVH detection.

CONCLUSION

None of the ECG criteria were sensitive enough to rule out ventricular hypertrophy. In the context of cardiac remodeling, the ECG criteria had high sensitivity but low specificity and, hence, limited clinical relevance.

Survey and Evaluation of Hypertension Machine Learning Research

Background Machine learning (ML) is pervasive in all fields of research, from automating tasks to complex decision-making. However, applications in different specialities are variable and generally limited. Like other conditions, the number of studies employing ML in hypertension research is growing rapidly. In this study, we aimed to survey hypertension research using ML, evaluate the reporting quality, and identify barriers to ML's potential to transform hypertension care. Methods and Results The Harmonious Understanding of Machine Learning Analytics Network survey questionnaire was applied to 63 hypertension-related ML research articles published between January 2019 and September 2021. The most common research topics were blood pressure prediction (38%), hypertension (22%), cardiovascular outcomes (6%), blood pressure variability (5%), treatment response (5%), and real-time blood pressure estimation (5%). The reporting quality of the articles was variable. Only 46% of articles described the study population or derivation cohort. Most articles (81%) reported at least 1 performance measure, but only 40% presented any measures of calibration. Compliance with ethics, patient privacy, and data security regulations were mentioned in 30 (48%) of the articles. Only 14% used geographically or temporally distinct validation data sets. Algorithmic bias was not addressed in any of the articles, with only 6 of them acknowledging risk of bias. Conclusions Recent ML research on hypertension is limited to exploratory research and has significant shortcomings in reporting quality, model validation, and algorithmic bias. Our analysis identifies areas for improvement that will help pave the way for the realization of the potential of ML in hypertension and facilitate its adoption.

Interpretable Machine Learning Techniques in ECG-Based Heart Disease Classification: A Systematic Review

Heart disease is one of the leading causes of mortality throughout the world. Among the different heart diagnosis techniques, an electrocardiogram (ECG) is the least expensive non-invasive procedure. However, the following are challenges: the scarcity of medical experts, the complexity of ECG interpretations, the manifestation similarities of heart disease in ECG signals, and heart disease comorbidity. Machine learning algorithms are viable alternatives to the traditional diagnoses of heart disease from ECG signals. However, the black box nature of complex machine learning algorithms and the difficulty in explaining a model's outcomes are obstacles for medical practitioners in having confidence in machine learning models. This observation paves the way for interpretable machine learning (IML) models as diagnostic tools that can build a physician's trust and provide evidence-based diagnoses. Therefore, in this systematic literature review, we studied and analyzed the research landscape in interpretable machine learning techniques by focusing on heart disease diagnosis from an ECG signal. In this regard, the contribution of our work is manifold; first, we present an elaborate discussion on interpretable machine learning techniques. In addition, we identify and characterize ECG signal recording datasets that are readily available for machine learning-based tasks. Furthermore, we identify the progress that has been achieved in ECG signal interpretation using IML techniques. Finally, we discuss the limitations and challenges of IML techniques in interpreting ECG signals.

Artificial intelligence-based opportunistic screening for the detection of arterial hypertension through ECG signals

OBJECTIVES

Hypertension is a major risk factor for cardiovascular disease (CVD), which often escapes the diagnosis or should be confirmed by several office visits. The ECG is one of the most widely used diagnostic tools and could be of paramount importance in patients' initial evaluation.

METHODS

We used machine learning techniques based on clinical parameters and features derived from the ECG, to detect hypertension in a population without CVD. We enrolled 1091 individuals who were classified as hypertensive or normotensive, and trained a Random Forest model, to detect the existence of hypertension. We then calculated the values for the Shapley additive explanations (SHAP), a sophisticated feature importance analysis, to interpret each feature's role in the Random Forest's results.

RESULTS

Our Random Forest model was able to distinguish hypertensive from normotensive patients with accuracy 84.2%, specificity 78.0%, sensitivity 84.0% and area under the receiver-operating curve 0.89, using a decision threshold of 0.6. Age, BMI, BMI-adjusted Cornell criteria (BMI multiplied by RaVL+SV 3 ), R wave amplitude in aVL and BMI-modified Sokolow-Lyon voltage (BMI divided by SV 1 +RV 5 ), were the most important anthropometric and ECG-derived features in terms of the success of our model.

CONCLUSION

Our machine learning algorithm is effective in the detection of hypertension in patients using ECG-derived and basic anthropometric criteria. Our findings open new horizon in the detection of many undiagnosed hypertensive individuals who have an increased CVD risk.

A Powerful Paradigm for Cardiovascular Risk Stratification Using Multiclass, Multi-Label, and Ensemble-Based Machine Learning Paradigms: A Narrative Review

Background and Motivation: Cardiovascular disease (CVD) causes the highest mortality globally. With escalating healthcare costs, early non-invasive CVD risk assessment is vital. Conventional methods have shown poor performance compared to more recent and fast-evolving Artificial Intelligence (AI) methods. The proposed study reviews the three most recent paradigms for CVD risk assessment, namely multiclass, multi-label, and ensemble-based methods in (i) office-based and (ii) stress-test laboratories. Methods: A total of 265 CVD-based studies were selected using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) model. Due to its popularity and recent development, the study analyzed the above three paradigms using machine learning (ML) frameworks. We review comprehensively these three methods using attributes, such as architecture, applications, pro-and-cons, scientific validation, clinical evaluation, and AI risk-of-bias (RoB) in the CVD framework. These ML techniques were then extended under mobile and cloud-based infrastructure. Findings: Most popular biomarkers used were office-based, laboratory-based, image-based phenotypes, and medication usage. Surrogate carotid scanning for coronary artery risk prediction had shown promising results. Ground truth (GT) selection for AI-based training along with scientific and clinical validation is very important for CVD stratification to avoid RoB. It was observed that the most popular classification paradigm is multiclass followed by the ensemble, and multi-label. The use of deep learning techniques in CVD risk stratification is in a very early stage of development. Mobile and cloud-based AI technologies are more likely to be the future. Conclusions: AI-based methods for CVD risk assessment are most promising and successful. Choice of GT is most vital in AI-based models to prevent the RoB. The amalgamation of image-based strategies with conventional risk factors provides the highest stability when using the three CVD paradigms in non-cloud and cloud-based frameworks.

Explainable machine learning predictions to support personalized cardiology strategies

Aims

A widely practiced intervention to modify cardiac health, the effect of physical activity on older adults is likely heterogeneous. While machine learning (ML) models that combine various systemic signals may aid in predictive modelling, the inability to rationalize predictions at a patient personalized level is a major shortcoming in the current field of ML.

Methods and results

We applied a novel methodology, SHapley Additive exPlanations (SHAP), on a dataset of older adults n = 86 (mean age 72 ± 4 years) whose physical activity levels were studied alongside changes in their left ventricular (LV) structure. SHAP was tested to provide intelligible visualization on the magnitude of the impact of the features in their physical activity levels on their LV structure. As proof of concept, using repeated K-cross-validation on the train set (n = 68), we found the Random Forest Regressor with the most optimal hyperparameters, which achieved the lowest mean squared error. With the trained model, we evaluated its performance by reporting its mean absolute error and plotting the correlation on the test set (n = 18). Based on collective force plot, individually numbered patients are indicated on the horizontal axis, and each bandwidth implies the magnitude (i.e. effect) of physical parameters (higher in red; lower in blue) towards prediction of their LV structure.

Conclusions

As a tool that identified specific features in physical activity that predicted cardiac structure on a per patient level, our findings support a role for explainable ML to be incorporated into personalized cardiology strategies.