Risk predicting for acute coronary syndrome based on machine learning model with kinetic plaque features from serial coronary computed tomography angiography

Prognostic value of serial coronary computed tomography angiography-derived perivascular fat-attenuation index and plaque volume in patients with suspected coronary artery disease

AIMS

To investigate the prognostic value of serial coronary computed tomography angiography (CCTA) derived plaque information, fractional flow reserve (CT-FFR), and perivascular fat-attenuation index (FAI) on major adverse cardiac events (MACE) in patients with suspected coronary artery disease.

MATERIALS AND METHODS

A total of 252 patients who underwent serial CCTA between January 2018 and December 2021 and were followed until June 2022. MACE were recorded. The analysis indexes included percent diameter stenosis (%DS), lesion length, plaque volume, CT-FFR, and FAI, with an emphasis on their changes between the baseline and follow-up CCTAs. Multivariate regression analysis were employed to identify independent risk factors for MACE.

RESULTS

After a median follow-up of 48-month, MACE occurred in 32 patients (12.7%). Patients with MACE displayed more severe stenosis, longer lesions, and larger plaque volumes in both baseline and follow-up CCTAs compared with no-MACE patients (all P<0.05). Patients with MACE displayed more severe stenosis, longer lesion, and larger plaque volume in both baseline and follow-up CCTAs compared with no-MACE patients. In addition, MACE patients also showed lower CT-FFR and higher △CT-FFR. Although FAI was significantly higher in MACE patients at baseline CCTA, FAI was notably increased in MACE patients, and decreased in the no-MACE patients (all P<0.05). Logistic regression analysis showed that ΔFAI, %DS, and plaque volume were independent predictors of MACE, with ΔFAI being the most significant (OR: 16.725, P<0.000). A multivariable model showed a significantly improved C-index of 0.903 (95% confidence interval: 0.836-0.970) for MACE prediction, when compared with single index alone.

CONCLUSIONS

Serial CCTA-derived ΔFAI, %DS, and plaque volume are crucial independent predictors of MACE in patients with suspected coronary artery disease, highlighting the importance of CCTA in patient risk stratification and prognostic assessment.

Epidemiology and current management of cardiovascular disease in China

The Annual Report on Cardiovascular Health and Diseases in China (2022) intricate landscape of cardiovascular health in China. This is the fourth section of the report with a specific focus on epidemiology and current management of cardiovascular disease (CVD) in China. This section of the report highlights the epidemiological trends of CVD in China. It reveal a concerning rise in prevalence, with approximately 330 million affected individuals, including significant numbers with stroke, coronary artery disease (CAD), heart failure, and other conditions. CVD stands as the primary cause of mortality among both urban and rural populations, accounting for nearly half of all deaths in 2020. Mortality rates are notably higher in rural areas compared to urban centers since 2009. While age-standardized mortality rates have decreased, the absolute number of CVD deaths has increased, primarily due to population aging. Ischemic heart disease, hemorrhagic and ischemic strokes are the leading causes of CVD-related deaths. Notably, the burden of atherosclerotic cardiovascular disease has risen substantially, with atherosclerotic cardiovascular disease-related deaths increasing from 1990 to 2016. The incidence of ischemic stroke and ischemic heart disease has shown similar increasing trends over the past three decades. CAD mortality, particularly acute myocardial infarction, has been on the rise, with higher mortality rates observed in rural areas since 2016. The prevalence of CAD has increased significantly, with over 11 million patients identified in 2013. Studies assessing hospital performance in managing acute coronary syndrome reveal gaps in adherence to guideline-recommended strategies, with disparities in care quality across hospitals. However, initiatives like the China Patient-centered Evaluative Assessment of Cardiac Events (PEACE)-Retrospective AMI Study and the Improving Care for Cardiovascular Disease in China-Acute Coronary Syndrome (CCC-ACS) project aim to improve patient outcomes through enhanced care protocols. Moreover, advancements in medical technology, such as quantitative flow ratio-guided lesion selection during percutaneous coronary intervention, show promise in improving clinical outcomes for patients undergoing intervention.

Machine learning-based identification of symptomatic carotid atherosclerotic plaques with dual-energy computed tomography angiography

OBJECTIVE

This study aimed to develop and validate a machine learning model incorporating both dual-energy computed tomography (DECT) angiography quantitative parameters and clinically relevant risk factors for the identification of symptomatic carotid plaques to prevent acute cerebrovascular events.

METHODS

The data of 180 patients with carotid atherosclerosis plaques were analysed from January 2017 to December 2021; 110 patients (64.03±9.58 years old, 20 women, 90 men) were allocated to the symptomatic group, and 70 patients (64.70±9.89 years old, 50 women, 20 men) were allocated to the asymptomatic group. Overall, five machine learning models using the XGBoost algorithm, based on different CT and clinical features, were developed in the training cohort. The performances of all five models were assessed in the testing cohort using receiver operating characteristic curves, accuracy, recall rate, and F1 score.

RESULTS

The shapley additive explanation (SHAP) value ranking showed fat fraction (FF) as the highest among all CT and clinical features and normalised iodine density (NID) as the 10th. The model based on the top 10 features from the SHAP measurement showed optimal performance (area under the curve [AUC] .885, accuracy .833, recall rate .933, F1 score .861), compared with the other four models based on conventional CT features (AUC .588, accuracy .593, recall rate .767, F1 score .676), DECT features (AUC .685, accuracy .648, recall rate .667, F1 score .678), conventional CT and DECT features (AUC .819, accuracy .740, recall rate .867, F1 score .788), and all CT and clinical features (AUC .878, accuracy .833, recall rate .867, F1 score .852).

CONCLUSION

FF and NID can serve as useful imaging markers of symptomatic carotid plaques. This tree-based machine learning model incorporating both DECT and clinical features could potentially comprise a non-invasive method for identification of symptomatic carotid plaques to guide clinical treatment strategies.

Deep learning-based prediction for significant coronary artery stenosis on coronary computed tomography angiography in asymptomatic populations

Background

Although coronary computed tomography angiography (CCTA) is currently utilized as the frontline test to accurately diagnose coronary artery disease (CAD) in clinical practice, there are still debates regarding its use as a screening tool for the asymptomatic population. Using deep learning (DL), we sought to develop a prediction model for significant coronary artery stenosis on CCTA and identify the individuals who would benefit from undergoing CCTA among apparently healthy asymptomatic adults.

Methods

We retrospectively reviewed 11,180 individuals who underwent CCTA as part of routine health check-ups between 2012 and 2019. The main outcome was the presence of coronary artery stenosis of ≥70% on CCTA. We developed a prediction model using machine learning (ML), including DL. Its performance was compared with pretest probabilities, including the pooled cohort equation (PCE), CAD consortium, and updated Diamond-Forrester (UDF) scores.

Results

In the cohort of 11,180 apparently healthy asymptomatic individuals (mean age 56.1 years; men 69.8%), 516 (4.6%) presented with significant coronary artery stenosis on CCTA. Among the ML methods employed, a neural network with multi-task learning (19 selected features), one of the DL methods, was selected due to its superior performance, with an area under the curve (AUC) of 0.782 and a high diagnostic accuracy of 71.6%. Our DL-based model demonstrated a better prediction than the PCE (AUC, 0.719), CAD consortium score (AUC, 0.696), and UDF score (AUC, 0.705). Age, sex, HbA1c, and HDL cholesterol were highly ranked features. Personal education and monthly income levels were also included as important features of the model.

Conclusion

We successfully developed the neural network with multi-task learning for the detection of CCTA-derived stenosis of ≥70% in asymptomatic populations. Our findings suggest that this model may provide more precise indications for the use of CCTA as a screening tool to identify individuals at a higher risk, even in asymptomatic populations, in clinical practice.

Clinical value of perivascular fat attenuation index and computed tomography derived fractional flow reserve in identification of culprit lesion of subsequent acute coronary syndrome

Purpose

To explore the potential of perivascular fat attenuation index (FAI) and coronary computed tomography angiography (CCTA) derived fractional flow reserve (CT-FFR) in the identification of culprit lesion leading to subsequent acute coronary syndrome (ACS).

Methods

Thirty patients with documented ACS event who underwent invasive coronary angiography (ICA) from February 2019 to February 2021 and had received CCTA in the previous 6 months were collected retrospectively. 40 patients with stable angina pectoris (SAP) were matched as control group according to sex, age and risk factors. The study population has a mean age of 59.3 ± 12.3 years, with a male prevalence of 81.4%. The plaque characteristics, perivascular fat attenuation index (FAI), and coronary computed tomography angiography-derived fractional flow reserve (CT-FFR) of 32 culprit lesions and 30 non-culprit lesions in ACS patients and 40 highest-grade stenosis lesions in SAP patients were statistically analyzed.

Results

FAI around culprit lesions was increased significantly (-72.4 ± 3.2 HU vs. -79.0 ± 7.7 HU, vs. -80.4 ± 7.0HU, all p < 0.001) and CT-FFR was decreased for culprit lesions of ACS patients [0.7(0.1) vs. 0.8(0.1), vs.0.8(0.1), p < 0.001] compared to other lesions. According to multivariate analysis, diameter stenosis (DS), FAI, and CT-FFR were significant predictors for identification of the culprit lesion. The integration model of DS, FAI, and CT-FFR showed the significantly highest area under the curve (AUC) of 0.917, compared with other single predictors (all p < 0.05).

Conclusions

This study proposes a novel integrated prediction model of DS, FAI, and CT-FFR that enhances the diagnostic accuracy of traditional CCTA for identifying culprit lesions that trigger ACS. Furthermore, this model also provides improved risk stratification for patients and offers valuable insights for predicting future cardiovascular events.

Impact of coronary computed tomography angiography-derived fractional flow reserve based on deep learning on clinical management

Background

To examine the value of coronary computed tomography angiography (CCTA)-derived fractional flow reserve based on deep learning (DL-FFRCT) on clinical practice and analyze the limitations of the application of DL-FFRCT.

Methods

This is an observational, retrospective, single-center study. Patients with suspected coronary artery disease (CAD) were enrolled. The patients underwent invasive coronary angiography (ICA) examination within 1 months after CCTA examination. And quantitative coronary angiography (QCA) was performed to evaluate the area stenosis rate. The CCTA data of these patients were retrospectively analyzed to calculate the FFRCT value.

Results

A total of 485 lesions of coronary arteries in 229 patients were included in the analysis. Of the lesions, 275 (56.7%) were ICA-positive, and 210 (43.3%) were FFRCT-positive. The discordance rate of the risk stratification of FFRCT for ICA-positive lesions was 33.1% (91) and that for ICA-negative lesions was 12.4% (26). 14.6% (7/48) patients with mild to moderate coronary stenosis in ICA have functional ischemia according to FFRCT positive indications. In addition, hemodynamic analysis of severely calcified, occluded, or small (< 2 mm in diameter) coronary arteries by DL-FFRCT is not so reliable.

Conclusion

This study revealed that most patients with ICA negative did not require further invasive FFR. Besides, some patients with mild to moderate coronary stenosis in ICA may also have functional ischemia. However, for severely calcified, occluded, or small coronary arteries, treatment strategy should be selected based on ICA in combination with clinical practice.

Leveraging Machine Learning Techniques to Forecast Chronic Total Occlusion before Coronary Angiography

BACKGROUND

Chronic total occlusion (CTO) remains the most challenging procedure in coronary artery disease (CAD) for interventional cardiology. Although some clinical risk factors for CAD have been identified, there is no personalized prognosis test available to confidently identify patients at high or low risk for CTO CAD. This investigation aimed to use a machine learning algorithm for clinical features from clinical routine to develop a precision medicine tool to predict CTO before CAG.

METHODS

Data from 1473 CAD patients were obtained, including 1105 in the training cohort and 368 in the testing cohort. The baseline clinical characteristics were collected. Univariate and multivariate logistic regression analyses were conducted to identify independent risk factors that impact the diagnosis of CTO. A CTO predicting model was established and validated based on the independent predictors using a machine learning algorithm. The area under the curve (AUC) was used to evaluate the model.

RESULTS

The CTO prediction model was developed with the training cohort using the machine learning algorithm. Eight variables were confirmed as 'important': gender (male), neutrophil percentage (NE%), hematocrit (HCT), total cholesterol (TC), high-density lipoprotein cholesterol (HDL), ejection fraction (EF), troponin I (TnI), and N-terminal pro-B-type natriuretic peptide (NT-proBNP). The model achieved good concordance indices of 0.724 and 0.719 in the training and testing cohorts, respectively.

CONCLUSIONS

An easy-to-use tool to predict CTO in patients with CAD was developed and validated. More research with larger cohorts are warranted to improve the prediction model, which can support clinician decisions on the early discerning CTO in CAD patients.

Coronary Computed Tomography Angiography-derived Fractional Flow Reserve: An Expert Consensus Document of Chinese Society of Radiology

Invasive fractional flow reserve (FFR) measured by a pressure wire is a reference standard for evaluating functional stenosis in coronary artery disease. Coronary computed tomography angiography-derived fractional flow reserve (CT-FFR) uses advanced computational analysis methods to noninvasively obtain FFR results from a single conventional coronary computed tomography angiography data to evaluate the hemodynamic significance of coronary artery disease. More and more evidence has found good correlation between the results of noninvasive CT-FFR and invasive FFR. CT-FFR has proven its potential in optimizing patient management, improving risk stratification and prognosis, and reducing total health care costs. However, there is still a lack of standardized interpretation of CT-FFR technology in real-world clinical settings. This expert consensus introduces the principle, workflow, and interpretation of CT-FFR; summarizes the state-of-the-art application of CT-FFR; and provides suggestions and recommendations for the application of CT-FFR with the aim of promoting the standardized application of CT-FFR in clinical practice.

Predicting major adverse cardiac events based on multi-parameter coronary computed tomography angiography

OBJECTIVE

To build a nomogram model to improve the prediction of major adverse cardiac events (MACE) using multi-parameter coronary computed tomography angiography (CCTA).

METHODS

All patients underwent CCTA. Those who developed MACE 90 days later but within 2 years between January 2008 and December 2018 were retrospectively enrolled as MACE group, while those without MACE were 1:1 propensity score matched in the control group. CCTA stenosis, plaque qualitative-quantitative characteristics, and fractional flow reserve derived from computed tomography angiography (FFRct) were analyzed and compared between the two groups. The independent risk factors for predicting MACE were obtained through univariate and multivariate regression analysis, after which multi-parameter models were built to predict MACE. Finally, the nomogram for predicting MACE was created using the independent risk factors from multivariate regression analysis.

RESULTS

A total of 483 vessels in 260 patients were successfully analyzed. The combination of CCTA stenosis, plaque qualitative-quantitative characteristics, and FFRct (AUC = 0.922, P<0.001) showed a higher predictive value compared to CCTA stenosis alone, FFRct alone, plaque qualitative-quantitative characteristics alone, CCTA stenosis combined with plaque qualitative-quantitative characteristics, and CCTA stenosis combined with FFRct (all P <0.001). Independent risk factors were CCTA stenosis ≥50%, low attenuation plaque, positive remodeling, napkin ring sign, lipid plaque volume proportion, and FFRct. Subsequently, a nomogram was created using these independent risk factors.

CONCLUSIONS

The multi-parameter CCTA model has improved performance in predicting MACE. Nomogram for predicting MACE, which includes these factors, represents a practical and easy-to-use method in the clinical setting. This article is protected by copyright. All rights reserved.