Fully automatic coronary calcium scoring in non-ECG-gated low-dose chest CT: comparison with ECG-gated cardiac CT

Performance of fully automated deep-learning-based coronary artery calcium scoring in ECG-gated calcium CT and non-gated low-dose chest CT

OBJECTIVES

This study aimed to validate the agreement and diagnostic performance of a deep-learning-based coronary artery calcium scoring (DL-CACS) system for ECG-gated and non-gated low-dose chest CT (LDCT) across multivendor datasets.

MATERIALS AND METHODS

In this retrospective study, datasets from Seoul National University Hospital (SNUH, 652 paired ECG-gated and non-gated CT scans) and the Stanford public dataset (425 ECG-gated and 199 non-gated CT scans) were analyzed. Agreement metrics included intraclass correlation coefficient (ICC), coefficient of determination (R²), and categorical agreement (κ). Diagnostic performance was assessed using categorical accuracy and the area under the receiver operating characteristic curve (AUROC).

RESULTS

DL-CACS demonstrated excellent performance for ECG-gated CT in both datasets (SNUH: R² = 0.995, ICC = 0.997, κ = 0.97, AUROC = 0.99; Stanford: R² = 0.989, ICC = 0.990, κ = 0.97, AUROC = 0.99). For non-gated CT using manual LDCT CAC scores as a reference, performance was similarly high (R² = 0.988, ICC = 0.994, κ = 0.96, AUROC = 0.98-0.99). When using ECG-gated CT scores as the reference, performance for non-gated CT was slightly lower but remained robust (SNUH: R² = 0.948, ICC = 0.968, κ = 0.88, AUROC = 0.98-0.99; Stanford: R² = 0.949, ICC = 0.948, κ = 0.71, AUROC = 0.89-0.98).

CONCLUSION

DL-CACS provides a reliable and automated solution for CACS, potentially reducing workload while maintaining robust performance in both ECG-gated and non-gated CT settings.

KEY POINTS

Question How accurate and reliable is deep-learning-based coronary artery calcium scoring (DL-CACS) in ECG-gated CT and non-gated low-dose chest CT (LDCT) across multivendor datasets? Findings DL-CACS showed near-perfect performance for ECG-gated CT. For non-gated LDCT, performance was excellent using manual scores as the reference and lower but reliable when using ECG-gated CT scores. Clinical relevance DL-CACS provides a reliable and automated solution for CACS, potentially reducing workload and improving diagnostic workflow. It supports cardiovascular risk stratification and broader clinical adoption, especially in settings where ECG-gated CT is unavailable.

Action plan for an international imaging framework for implementation of global low-dose CT screening for lung cancer

Reduction in lung cancer mortality is achievable through low dose computed tomography (LDCT) screening in high-risk individuals. Many countries are progressing from local LDCT screening studies to national screening programs. Implementation of effective large-scale screening programs is complex and requires a multi-disciplinary approach. A recent overview of the technical aspects of implementing high quality LDCT for screening resulted from the inaugural international expert meeting of the Alliance for Global Implementation of Lung and Cardiac Early Disease Detection and Treatment (AGILE). This covers the most important aspects of the CT imaging process: standardisation in CT image acquisition and interpretation, CT protocol management, technology developments and minimal requirements, integration of lung cancer biomarkers, and the role of AI in CT lung nodule detection, segmentation, and classification, and related data security issues.

A systematic review on the impact of artificial intelligence on electrocardiograms in cardiology

BACKGROUND

Artificial intelligence (AI) has revolutionized numerous industries, enhancing efficiency, scalability, and insight generation. In cardiology, particularly through electrocardiogram (ECG) analysis, AI has the potential to improve diagnostic accuracy and reduce the time needed for diagnosis. This systematic review explores the integration of AI, machine learning (ML), and deep learning (DL) in ECG analysis, focusing on their impact on predictive diagnostics and treatment support in cardiology.

METHODS

A systematic literature review was conducted following the PRISMA 2020 framework, using four high-impact databases to identify studies from 2014 to -2024. The inclusion criteria included English-language journal articles and research papers that focused on AI applications in cardiology, specifically ECG analysis. Records were screened, duplicates were removed, and final selections were made on the basis of their relevance to AI-ECG integration for cardiac health.

RESULTS

The review included 46 studies that met the inclusion criteria, covering diverse AI models such as CNNs, RNNs, and hybrid models. These models were applied to ECG data to detect and predict heart conditions such as arrhythmia, myocardial infarction, and heart failure. These findings indicate that AI-driven ECG analysis improves diagnostic accuracy and provides significant support for early diagnosis and personalized treatment.

CONCLUSIONS

AI technologies, especially ML and DL, enhance ECG-based cardiology diagnostics by increasing accuracy, reducing diagnosis time, and supporting timely interventions and personalized care. Continued research in this area is essential to refine algorithms and integrate AI tools into clinical practice for improved patient outcomes in cardiology.

Updates on CAD risk assessment: using the coronary artery calcium score in combination with traditional risk factors

BACKGROUND

Coronary artery disease (CAD) is the third leading cause of death worldwide, so prevention and early diagnosis play important roles to reduce mortality and morbidity. Traditional risk-score assessments were used to find the at-risk patients in order to prevent or early treatment of CAD. Adding imaging data to traditional risk-score systems will able us to find these patients more confidently and reduce the probable mismanagements.

MAIN TEXT

Measuring the vascular calcification by coronary artery calcium (CAC) score can prepare valuable data for this purpose. Using CAC became more popular in recent years. The most applicable method to evaluate CAC is Agatston scoring using computed tomography (CT) scanning. Patients are classified into several subgroups: no evidence of CAD (score 0), mild CAD (score 1-10), minimal CAD (score 11-100), moderate CAD (score 101-400), and severe CAD (score > 400) and higher than1000 as the extreme risk of CVD events.

CONCLUSIONS

CAC assessment was recommended in the patients older than 40 years old with CAD risk factors, the ones with stable angina, borderline-to-intermediate-risk group, etc. According to the results of the CAC the patients may be candidate for further evaluation for needing revascularization, medical treatment, or routine follow-up. Adding artificial intelligence (AI) to CAC will prepare more data and can increase the reliability of our approach to the patients promising a bright future to improve this technology.

Automatic identification of coronary stent in coronary calcium scoring CT using deep learning

Coronary artery calcium (CAC) scoring CT is a useful tool for screening coronary artery disease and for cardiovascular risk stratification. However, its efficacy in patients with coronary stents, who had pre-existing coronary artery disease, remains uncertain. Historically, CAC CT scans of these patients have been manually excluded from the CAC scoring process, even though most of the CAC scoring process is now fully automated. Therefore, we hypothesized that automating the filtering of patients with coronary stents using artificial intelligence could streamline the entire CAC workflow, eliminating the need for manual intervention. Consequently, we aimed to develop and evaluate a deep learning-based coronary stent filtering algorithm (StentFilter) in CAC scoring CT scans using a multicenter CAC dataset. We developed StentFilter comprising two main processes: stent identification and false-positive reduction. Development utilized 108 non-enhanced echocardiography-gated CAC scans (including 74 with manually labeled stents), and for false positive reduction, 2063 CAC scans with significant coronary calcium (average Agatston score: 523.8) but no stents were utilized. StentFilter's performance was evaluated on two independent internal test sets (Asan cohort- and 2; n = 355 and 396; one without coronary stents) and two external test sets from different institutions (n = 105 and 62), each with manually labeled stents. We calculated the per-patient sensitivity, specificity, and false-positive rate of StentFilter. StentFilter demonstrated a high overall per-patient sensitivity of 98.8% (511/517 cases with stents) and a false-positive rate of 0.022 (20/918). Notably, the false-positive ratio was significantly lower in the dataset containing stents (Asan cohort-1; 0.008 [3/355]) compared with the dataset without stents (Asan cohort-2; 0.043 [17/396], p = 0.008). All false-positive identifications were attributed to dense coronary calcifications, with no false positives identified in extracoronary locations. The automated StentFilter accurately distinguished coronary stents from pre-existing coronary calcifications. This approach holds potential as a filter within a fully automated CAC scoring workflow, streamlining the process efficiently.

Novel risk score for predicting acute cardiovascular and cerebrovascular events after chest radiotherapy in patients with breast or lung cancer

AIMS

Radiation therapy (RT) is an integral component of cancer therapy but associated with adverse events. Our goal was to establish risk prediction models for major adverse cardiovascular and cerebrovascular events (MACCE) after chest RT.

METHODS AND RESULTS

A retrospective study of lung/breast cancer patients who had chest RT with planning CT at Mayo Clinic between 01/2010 and 01/2014. Predictive models were developed based on weighted independent predictors using a derivation (406 lung and 711 breast cancer) and validation cohort (179 lung and 234 breast cancer). Patient characteristics, pre-RT CT for coronary artery calcification (CAC), and post-RT MACCE data were reviewed. Post-RT MACCE occurred in 6.1 and 5.6% in the derivation and validation cohort over a mean follow-up of 42 ± 13 months. Post-therapy model (C2AD2) included CAC (two points), MACCE history (two points), age ≥74 (three points), DM (two points), and mean heart radiation dose ≥ 850 mGy (two points), and pre-therapy model (C2AD) included post-therapy model parameters minus mean heart radiation dose. Both models stratified patients into three risk groups: low (0-2), intermediate (3-5), and high (≥6). Post-RT MACCE across these groups were 2.7, 8.9, and 19.8% in the derivation, and 3.9, 6.6, and 16.4% in the validation cohort for post-therapy model (C2AD2) and 2.8, 9.2, and 20.4% in the derivation and 3.7, 9.2, and 13.2% in the validation cohort for pre-therapy model. Both models showed statistically significant graded survival outcome.

CONCLUSION

Post-therapy (C2AD2) and pre-therapy (C2AD) models are simple, easy to use and effective tools to stratify breast and lung cancer patients undergoing chest radiation for post-RT MACCE.

Artificial Intelligence in Coronary Artery Calcium Scoring

Cardiovascular disease (CVD), particularly coronary heart disease (CHD), is the leading cause of death in the US, with a high economic impact. Coronary artery calcium (CAC) is a known marker for CHD and a useful tool for estimating the risk of atherosclerotic cardiovascular disease (ASCVD). Although CACS is recommended for informing the decision to initiate statin therapy, the current standard requires a dedicated CT protocol, which is time-intensive and contributes to radiation exposure. Non-dedicated CT protocols can be taken advantage of to visualize calcium and reduce overall cost and radiation exposure; however, they mainly provide visual estimates of coronary calcium and have disadvantages such as motion artifacts. Artificial intelligence is a growing field involving software that independently performs human-level tasks, and is well suited for improving CACS efficiency and repurposing non-dedicated CT for calcium scoring. We present a review of the current studies on automated CACS across various CT protocols and discuss consideration points in clinical application and some barriers to implementation.

Recent advances in artificial intelligence for cardiac CT: Enhancing diagnosis and prognosis prediction

Recent advances in artificial intelligence (AI) for cardiac computed tomography (CT) have shown great potential in enhancing diagnosis and prognosis prediction in patients with cardiovascular disease. Deep learning, a type of machine learning, has revolutionized radiology by enabling automatic feature extraction and learning from large datasets, particularly in image-based applications. Thus, AI-driven techniques have enabled a faster analysis of cardiac CT examinations than when they are analyzed by humans, while maintaining reproducibility. However, further research and validation are required to fully assess the diagnostic performance, radiation dose-reduction capabilities, and clinical correctness of these AI-driven techniques in cardiac CT. This review article presents recent advances of AI in the field of cardiac CT, including deep-learning-based image reconstruction, coronary artery motion correction, automatic calcium scoring, automatic epicardial fat measurement, coronary artery stenosis diagnosis, fractional flow reserve prediction, and prognosis prediction, analyzes current limitations of these techniques and discusses future challenges.

Automated Coronary Artery Calcium and Quantitative Emphysema in Lung Cancer Screening: Association With Mortality, Lung Cancer Incidence, and Airflow Obstruction

PURPOSE

To assess automated coronary artery calcium (CAC) and quantitative emphysema (percentage of low attenuation areas [%LAA]) for predicting mortality and lung cancer (LC) incidence in LC screening. To explore correlations between %LAA, CAC, and forced expiratory value in 1 second (FEV 1 ) and the discriminative ability of %LAA for airflow obstruction.

MATERIALS AND METHODS

Baseline low-dose computed tomography scans of the BioMILD trial were analyzed using an artificial intelligence software. Univariate and multivariate analyses were performed to estimate the predictive value of %LAA and CAC. Harrell C -statistic and time-dependent area under the curve (AUC) were reported for 3 nested models (Model survey : age, sex, pack-years; Model survey-LDCT : Model survey plus %LAA plus CAC; Model final : Model survey-LDCT plus selected confounders). The correlations between %LAA, CAC, and FEV 1 and the discriminative ability of %LAA for airflow obstruction were tested using the Pearson correlation coefficient and AUC-receiver operating characteristic curve, respectively.

RESULTS

A total of 4098 volunteers were enrolled. %LAA and CAC independently predicted 6-year all-cause (Model final hazard ratio [HR], 1.14 per %LAA interquartile range [IQR] increase [95% CI, 1.05-1.23], 2.13 for CAC ≥400 [95% CI, 1.36-3.28]), noncancer (Model final HR, 1.25 per %LAA IQR increase [95% CI, 1.11-1.37], 3.22 for CAC ≥400 [95%CI, 1.62-6.39]), and cardiovascular (Model final HR, 1.25 per %LAA IQR increase [95% CI, 1.00-1.46], 4.66 for CAC ≥400, [95% CI, 1.80-12.58]) mortality, with an increase in concordance probability in Model survey-LDCT compared with Model survey ( P <0.05). No significant association with LC incidence was found after adjustments. Both biomarkers negatively correlated with FEV 1 ( P <0.01). %LAA identified airflow obstruction with a moderate discriminative ability (AUC, 0.738).

CONCLUSIONS

Automated CAC and %LAA added prognostic information to age, sex, and pack-years for predicting mortality but not LC incidence in an LC screening setting. Both biomarkers negatively correlated with FEV 1 , with %LAA enabling the identification of airflow obstruction with moderate discriminative ability.

A Modified Length-Based Grading Method for Assessing Coronary Artery Calcium Severity on Non-Electrocardiogram-Gated Chest Computed Tomography: A Multiple-Observer Study

OBJECTIVE

To validate a simplified ordinal scoring method, referred to as modified length-based grading, for assessing coronary artery calcium (CAC) severity on non-electrocardiogram (ECG)-gated chest computed tomography (CT).

MATERIALS AND METHODS

This retrospective study enrolled 120 patients (mean age ± standard deviation [SD], 63.1 ± 14.5 years; male, 64) who underwent both non-ECG-gated chest CT and ECG-gated cardiac CT between January 2011 and December 2021. Six radiologists independently assessed CAC severity on chest CT using two scoring methods (visual assessment and modified length-based grading) and categorized the results as none, mild, moderate, or severe. The CAC category on cardiac CT assessed using the Agatston score was used as the reference standard. Agreement among the six observers for CAC category classification was assessed using Fleiss kappa statistics. Agreement between CAC categories on chest CT obtained using either method and the Agatston score categories on cardiac CT was assessed using Cohen's kappa. The time taken to evaluate CAC grading was compared between the observers and two grading methods.

RESULTS

For differentiation of the four CAC categories, interobserver agreement was moderate for visual assessment (Fleiss kappa, 0.553 [95% confidence interval {CI}: 0.496-0.610]) and good for modified length-based grading (Fleiss kappa, 0.695 [95% CI: 0.636-0.754]). The modified length-based grading demonstrated better agreement with the reference standard categorization with cardiac CT than visual assessment (Cohen's kappa, 0.565 [95% CI: 0.511-0.619 for visual assessment vs. 0.695 [95% CI: 0.638-0.752] for modified length-based grading). The overall time for evaluating CAC grading was slightly shorter in visual assessment (mean ± SD, 41.8 ± 38.9 s) than in modified length-based grading (43.5 ± 33.2 s) (P < 0.001).

CONCLUSION

The modified length-based grading worked well for evaluating CAC on non-ECG-gated chest CT with better interobserver agreement and agreement with cardiac CT than visual assessment.