Deep Learning for Cardiovascular Risk Stratification

Enhancing the diagnosis of functionally relevant coronary artery disease with machine learning

Functionally relevant coronary artery disease (fCAD) can result in premature death or nonfatal acute myocardial infarction. Its early detection is a fundamentally important task in medicine. Classical detection approaches suffer from limited diagnostic accuracy or expose patients to possibly harmful radiation. Here we show how machine learning (ML) can outperform cardiologists in predicting the presence of stress-induced fCAD in terms of area under the receiver operating characteristic (AUROC: 0.71 vs. 0.64, p = 4.0E-13). We present two ML approaches, the first using eight static clinical variables, whereas the second leverages electrocardiogram signals from exercise stress testing. At a target post-test probability for fCAD of <15%, ML facilitates a potential reduction of imaging procedures by 15-17% compared to the cardiologist's judgement. Predictive performance is validated on an internal temporal data split as well as externally. We also show that combining clinical judgement with conventional ML and deep learning using logistic regression results in a mean AUROC of 0.74.

Artificial Intelligence, Computational Simulations, and Extended Reality in Cardiovascular Interventions

Artificial intelligence, computational simulations, and extended reality, among other 21st century computational technologies, are changing the health care system. To collectively highlight the most recent advances and benefits of artificial intelligence, computational simulations, and extended reality in cardiovascular therapies, we coined the abbreviation AISER. The review particularly focuses on the following applications of AISER: 1) preprocedural planning and clinical decision making; 2) virtual clinical trials, and cardiovascular device research, development, and regulatory approval; and 3) education and training of interventional health care professionals and medical technology innovators. We also discuss the obstacles and constraints associated with the application of AISER technologies, as well as the proposed solutions. Interventional health care professionals, computer scientists, biomedical engineers, experts in bioinformatics and visualization, the device industry, ethics committees, and regulatory agencies are expected to streamline the use of AISER technologies in cardiovascular interventions and medicine in general.

Contrast-Enhanced Ultrasound Feasibility in Assessing Carotid Plaque Vulnerability-Narrative Review

The risk assessment for carotid atherosclerotic lesions involves not only determining the degree of stenosis but also plaque morphology and its composition. Recently, carotid contrast-enhanced ultrasound (CEUS) has gained importance for evaluating vulnerable plaques. This review explores CEUS's utility in detecting carotid plaque surface irregularities and ulcerations as well as intraplaque neovascularization and its alignment with histology. Initial indications suggest that CEUS might have the potential to anticipate cerebrovascular incidents. Nevertheless, there is a need for extensive, multicenter prospective studies that explore the relationships between CEUS observations and patient clinical outcomes in cases of carotid atherosclerotic disease.

ECG-guided non-invasive estimation of pulmonary congestion in patients with heart failure

Quantifying hemodynamic severity in patients with heart failure (HF) is an integral part of clinical care. A key indicator of hemodynamic severity is the mean Pulmonary Capillary Wedge Pressure (mPCWP), which is ideally measured invasively. Accurate non-invasive estimates of the mPCWP in patients with heart failure would help identify individuals at the greatest risk of a HF exacerbation. We developed a deep learning model, HFNet, that uses the 12-lead electrocardiogram (ECG) together with age and sex to identify when the mPCWP > 18 mmHg in patients who have a prior diagnosis of HF. The model was developed using retrospective data from the Massachusetts General Hospital and evaluated on both an internal test set and an independent external validation set, from another institution. We developed an uncertainty score that identifies when model performance is likely to be poor, thereby helping clinicians gauge when to trust a given model prediction. HFNet AUROC for the task of estimating mPCWP > 18 mmHg was 0.8 [Formula: see text] 0.01 and 0.[Formula: see text] 0.01 on the internal and external datasets, respectively. The AUROC on predictions with the highest uncertainty are 0.50 [Formula: see text] 0.02 (internal) and 0.[Formula: see text] 0.04 (external), while the AUROC on predictions with the lowest uncertainty were 0.86 ± 0.01 (internal) and 0.82 ± 0.01 (external). Using estimates of the prevalence of mPCWP > 18 mmHg in patients with reduced ventricular function, and a decision threshold corresponding to an 80% sensitivity, the calculated positive predictive value (PPV) is 0.[Formula: see text] 0.01when the corresponding chest x-ray (CXR) is consistent with interstitial edema HF. When the CXR is not consistent with interstitial edema, the estimated PPV is 0.[Formula: see text] 0.02, again at an 80% sensitivity threshold. HFNet can accurately predict elevated mPCWP in patients with HF using the 12-lead ECG and age/sex. The method also identifies cohorts in which the model is more/less likely to produce accurate outputs.

Predicting outcomes in patients with aortic stenosis using machine learning: the Aortic Stenosis Risk (ASteRisk) score

Objective

To use echocardiographic and clinical features to develop an explainable clinical risk prediction model in patients with aortic stenosis (AS), including those with low-gradient AS (LGAS), using machine learning (ML).

Methods

In 1130 patients with moderate or severe AS, we used bootstrap lasso regression (BLR), an ML method, to identify echocardiographic and clinical features important for predicting the combined outcome of all-cause mortality or aortic valve replacement (AVR) within 5 years after the initial echocardiogram. A separate hold out set, from a different centre (n=540), was used to test the generality of the model. We also evaluated model performance with respect to each outcome separately and in different subgroups, including patients with LGAS.

Results

Out of 69 available variables, 26 features were identified as predictive by BLR and expert knowledge was used to further reduce this set to 9 easily available and input features without loss of efficacy. A ridge logistic regression model constructed using these features had an area under the receiver operating characteristic curve (AUC) of 0.74 for the combined outcome of mortality/AVR. The model reliably identified patients at high risk of death in years 2–5 (HRs ≥2.0, upper vs other quartiles, for years 2–5, p<0.05, p=not significant in year 1) and was also predictive in the cohort with LGAS (n=383, HRs≥3.3, p<0.05). The model performed similarly well in the independent hold out set (AUC 0.78, HR ≥2.5 in years 1–5, p<0.05).

Conclusion

In two separate longitudinal databases, ML identified prognostic features and produced an algorithm that predicts outcome for up to 5 years of follow-up in patients with AS, including patients with LGAS. Our algorithm, the Aortic Stenosis Risk (ASteRisk) score, is available online for public use.

A Deep Learning Model for Inferring Elevated Pulmonary Capillary Wedge Pressures From the 12-Lead Electrocardiogram

Background

Central hemodynamic parameters are typically measured via pulmonary artery catherization-an invasive procedure that involves some risk to the patient and is not routinely available in all settings.

Objectives

This study sought to develop a noninvasive method to identify elevated mean pulmonary capillary wedge pressure (mPCWP).

Methods

We leveraged data from 248,955 clinical records at the Massachusetts General Hospital to develop a deep learning model that can infer when the mPCWP >15 mmHg using the 12-lead electrocardiogram (ECG). Of these data, 242,216 records were used to pre-train a model that generates useful ECG representations. The remaining 6,739 records contain encounters with direct measurements of the mPCWP. Eighty percent of these data were used for model development and testing (5,390), and the remaining records comprise a holdout set (1,349) that was used to evaluate the model. We developed an associated unreliability score that identifies when model predictions are likely to be untrustworthy.

Results

The model achieves an area under the receiver operating characteristic curve (AUC) of 0.80 ± 0.02 (test set) and 0.79 ± 0.01 (holdout set). Model performance varies as a function of the unreliability, where patients with high unreliability scores correspond to a subgroup where model performance is poor: for example, patients in the holdout set with unreliability scores in the highest decile have a reduced AUC of 0.70 ± 0.06.

Conclusions

The mPCWP can be inferred from the ECG, and the reliability of this inference can be measured. When invasive monitoring cannot be expeditiously performed, deep learning models may provide information that can inform clinical care.

Prediction and risk stratification from hospital discharge records based on Hierarchical sLDA

BACKGROUND

The greatly accelerated development of information technology has conveniently provided adoption for risk stratification, which means more beneficial for both patients and clinicians. Risk stratification offers accurate individualized prevention and therapeutic decision making etc. Hospital discharge records (HDRs) routinely include accurate conclusions of diagnoses of the patients. For this reason, in this paper, we propose an improved model for risk stratification in a supervised fashion by exploring HDRs about coronary heart disease (CHD).

METHODS

We introduced an improved four-layer supervised latent Dirichlet allocation (sLDA) approach called Hierarchical sLDA model, which categorized patient features in HDRs as patient feature-value pairs in one-hot way according to clinical guidelines for lab test of CHD. To address the data missing and imbalance problem, RFs and SMOTE methods are used respectively. After TF-IDF processing of datasets, variational Bayes expectation-maximization method and generalized linear model were used to recognize the latent clinical state of a patient, i.e., risk stratification, as well as to predict CHD. Accuracy, macro-F1, training and testing time performance were used to evaluate the performance of our model.

RESULTS

According to the characteristics of our datasets, i.e., patient feature-value pairs, we construct a supervised topic model by adding one more Dirichlet distribution hyperparameter to sLDA. Compared with established supervised algorithm Multi-class sLDA model, we demonstrate that our proposed approach enhances training time by 59.74% and testing time by 25.58% but almost no loss of average prediction accuracy on our datasets.

CONCLUSIONS

A model for risk stratification and prediction of CHD based on sLDA model was proposed. Experimental results show that Hierarchical sLDA model we proposed is competitive in time performance and accuracy. Hierarchical processing of patient features can significantly improve the disadvantages of low efficiency and time-consuming Gibbs sampling of sLDA model.

The roles of predictors in cardiovascular risk models - a question of modeling culture?

BACKGROUND

While machine learning (ML) algorithms may predict cardiovascular outcomes more accurately than statistical models, their result is usually not representable by a transparent formula. Hence, it is often unclear how specific values of predictors lead to the predictions. We aimed to demonstrate with graphical tools how predictor-risk relations in cardiovascular risk prediction models fitted by ML algorithms and by statistical approaches may differ, and how sample size affects the stability of the estimated relations.

METHODS

We reanalyzed data from a large registry of 1.5 million participants in a national health screening program. Three data analysts developed analytical strategies to predict cardiovascular events within 1 year from health screening. This was done for the full data set and with gradually reduced sample sizes, and each data analyst followed their favorite modeling approach. Predictor-risk relations were visualized by partial dependence and individual conditional expectation plots.

RESULTS

When comparing the modeling algorithms, we found some similarities between these visualizations but also occasional divergence. The smaller the sample size, the more the predictor-risk relation depended on the modeling algorithm used, and also sampling variability played an increased role. Predictive performance was similar if the models were derived on the full data set, whereas smaller sample sizes favored simpler models.

CONCLUSION

Predictor-risk relations from ML models may differ from those obtained by statistical models, even with large sample sizes. Hence, predictors may assume different roles in risk prediction models. As long as sample size is sufficient, predictive accuracy is not largely affected by the choice of algorithm.