Cardiac Computed Tomography Imaging Research Articles

Predictors of debulking devices use by coronary artery calcification classification using cardiac CT

Abstract Background Multi-slice computed tomography (CT) is useful for the non-invasive detection and quantification of coronary calcium burden. However, a definitive parameter based on CT imaging for predicting the need for debulking devices, such as rotational atherectomy (RA) and orbital atherectomy (OAS), has yet to be established. Purpose We aimed to develop a calcification risk score using cardiac CT to predict the use of debulking devices during coronary intervention. Methods In this retrospective, multi-center, observational study in Japan, patients who underwent percutaneous coronary intervention (PCI) were screened. Inclusion criteria were (1) cardiac CT within 6 months prior to PCI, (2) moderate or severe coronary calcification detected on CT, and (3) de novo lesions in a native coronary artery. A total of 349 lesions in 343 patients from 13 hospitals, there were 124 lesions that were treated with debulking devices and 225 lesions without debulking devices. Cardiac CT images were analyzed for circumferential degree, longitudinal length, intracoronary distribution of calcification, peak density, mean density, vessel bend and presence of bifurcation lesions. Data from a random two-thirds (the derivation cohort, n=247) were used to develop a new risk prediction model, which was then tested in the remaining one-third (the validation cohort, n=102). Odds ratios and the risk prediction model were developed using a logistic regression model. The predicted risk was translated into a simplified scoring system (Figure 1A). Results RA was used in 111 lesions and OAS in 23 lesions. Factors selected in the model included circumferential degree (≥270°: 3 points), bifurcation (2 points), longitudinal length (≥8mm: 1 point), and mean density (≥600 HU: 1 point) (Figure 1B). This model demonstrated good discrimination (c-statistic=0.772) and calibration (Hosmer–Lemeshow χ²-statistic=0.885; P=0.64) in the validation cohort (Figure 2). Conclusions The new prediction model based on cardiac CT performed well in predicting the need for debulking devices. This could serve as an objective tool to support a clinical decision to use debulking devices even in cases where intravascular imaging is not applicable.Study flow and Cardiac CT calcium scoreDiscrimination and Calibration

Significant impact of a novel whole heart motion correction algorithm on cardiac computed tomography and aortic valve calcium scores

Abstract Introduction A new whole-heart motion-correction algorithm has been available in cardiac computed tomography (CT), and it has already revealed the higher diagnostic accuracy of significant coronary artery stenosis in cases with higher heart rate at the CT scan. Recently, aortic valve calcium scores (AVCS) are often measured for the decision-making of surgical treatment of aortic valve stenosis. However, the measurement of AVCS of the aortic valve has potential risks of motion artifact of the aortic valve. Therefore, we would like to know the clinical impact of a new whole-heart motion correction algorithm for accurately measuring aortic valve AVCS without motion artifacts. Purpose The purpose of this study is to evaluate the impact of a new whole-heart motion-correction algorithm on non-contrast cardiac CT images and the AVCS in trans-catheter aortic valve regurgitation (TAVR) candidates with severe aortic valve stenosis. Materials and methods Forty-two consecutive TAVR candidates (83 ± 7 years old, 21 males) who underwent ECG-gated CT using 256-row detector CT from November 2023 to February 2024 at our hospital were analyzed. The best phase non-contrast cardiac images were automatically selected, and they were reconstructed with both the standard (STD) algorithm and the new whole-heart motion-correction algorithm. AVCS was calculated using semi-automatic commercially available software. The aortic valve's image noise was also measured. Results Non-contrast best cardiac phase images were successfully reconstructed using the new motion correction algorithm in 36 patients (86%). The best cardiac phase was mainly the diastolic phase in 29 patients (81%). The overall Agatston score and calcified volume of the aortic valve in SSF2 were 1779 ± 896, and 1403 ± 689, respectively, and they were lower than 2195 ± 1132 and 1656 ± 851 of STD (all P < 0.01). Conclusion The new whole-heart motion-correction algorithm worked in almost all TAVR candidates, and revealed lower AVCS and shows a potential for increasing AVCS accuracy.

Deep learning enables the fully automated quantification of high-resolution cardiac CT images: human in loop for model development

Abstract Introduction Cardiac computed tomography (CCT) images provide vast spatial and temporal information. Segmentation of the structure of interest is necessary to provide quantitative information. However, manual segmentation of cardiac substructures is highly time-consuming. Objective This study aims to develop a robust deep-learning segmentation model for cardiac substructure segmentation in 3D CCT images. Method We initially developed a 3D U-Net architecture, incorporating extensive augmentation, for the segmentation of the Left Ventricle Myocardium (LVM), the blood cavities of the Left and Right Ventricles (LV, RV), and the Left and Right Atria (LA, RA). This model utilized CT images from 20 patients who had available manual segmentation. Subsequently, we applied this initial model to datasets from two centers, including different generations of scanners. We then visually inspected the segmentation, selected the highest-quality segmentations, and modify it if necessary, and excluded the low-quality data. We continued fine-tuning the model with the new dataset until we reached a count of 456 patients. As there were multiple images from one patient during this model's development, we generated datasets to develop a new model using the entire new dataset to avoid any data leakage in the reported metrics. The model was developed using data from Center 1 (5-fold cross-validation) and then evaluated on an external test set from Center 2 (126 patients) and also 4D images from Center 1. Various metrics were calculated to assess the model's performance. Results In the internal test set, utilizing 5-fold CV, the model demonstrated high segmentation accuracy, achieving an overall Dice coefficient of 0.98, with individual scores of 0.96 for LVM, 0.98 for LV, 0.98 for RV, 0.99 for RA, and 0.98 for LA. We assessed the automatic quantification capability of the segmentation model from 256 patients, resulting in an R2 value greater than 0.92 for both LV end-diastolic and systolic volumes. In contrast, the model maintained robust performance on the external dataset, achieving an overall Dice coefficient of 0.96, with individual scores of 0.95 for LVM, 0.98 for LV, 0.97 for RV, 0.98 for RA, and 0.94 for LA. Conclusion We provided a large cohort of ground truth segmentations for different cardiac CT substructures using a human-in-the-loop strategy. Subsequently, we developed a deep learning model for highly accurate automated segmentation of these substructures. This model allows for fast and fully automatic segmentation and quantification of 3D and 4D cardiac CT images, from which various quantitative metrics can be extracted from different cardiac substructures.

Computer-Assisted Algorithm for Quantification of Fibrosis by Native Cardiac CT: A Pilot Study.

Background/Objectives: Recent advances in artificial intelligence, particularly in cardiac imaging, can potentially enhance patients' diagnosis and prognosis and identify novel imaging markers. We propose an automated, computer-aided algorithm utilizing native cardiac computed tomography (CT) imaging to identify myocardial fibrosis. This study aims to evaluate its performance compared to CMR markers of fibrosis in a cohort of patients diagnosed with breast cancer. Methods: The study included patients diagnosed with early HER2+ breast cancer, who presented LV dysfunction (LVEF < 50%) and myocardial fibrosis detected on CMR at the time of diagnosis. The patients were also evaluated by cardiac CT, and the extracted images were processed for the implementation of the automatic, computer-assisted algorithm, which marked as fibrosis every pixel that fell within the range of 60-90 HU. The percentage of pixels with fibrosis was subsequently compared with CMR parameters. Results: A total of eight patients (n = 8) were included in the study. High positive correlations between the algorithm's result and the ECV fraction (r = 0.59, p = 0.126) and native T1 (r = 0.6, p = 0.112) were observed, and a very high positive correlation with LGE of the LV(g) and the LV-LGE/LV mass percentage (r = 0.77, p = 0.025; r = 0.81, p = 0.015). A very high negative correlation was found with GLS (r = -0.77, p = 0.026). The algorithm presented an intraclass correlation coefficient of 1 (95% CI 0.99-1), p < 0.001. Conclusions: The present pilot study proposes a novel promising imaging marker for myocardial fibrosis, generated by an automatic algorithm based on native cardiac CT images.

The role of the right atrial appendage and right atrium in post-radiofrequency ablation recurrence in different types of atrial fibrillation

To quantitatively evaluate the relationship between the anatomical parameters of the right atrium and the recurrence of atrial fibrillation (AF) after radiofrequency ablation, considering different types of AF, utilizing 256-slice spiral computed tomography (CT). A total of 297 patients with AF who underwent initial radiofrequency ablation were enrolled, divided into the paroxysmal atrial fibrillation (PaAF) group (n=230) and the persistent atrial fibrillation (PeAF) group (n=67). Subsequently, patients in each group were further stratified into recurrent and non-recurrent subgroups. In addition, 100 healthy outpatients were selected as the normal group. All patients underwent preoperative cardiac CT (CCT) examination. The volumes of the right atrium (RA), right atrial appendage (RAA), and left atrial (LA), RAA height, the length, short diameter, perimeter, and area of the RAA base, anteroposterior diameter of the RA, tricuspid annulus diameter, crista terminalis, and inferior vena cavotricuspid isthmus (CTI) on CCT images were measured. In both the PaAF group and the PeAF group, except for the crista terminalis thickness, the other measured parameters were greater than those in the normal group, and recurrent patients exhibited larger RAA base, crista terminalis and LA volume. Recurrent patients with PeAF presented larger RAVI, while recurrent patients with PaAF did not. The short diameter of the RAA base was an independent predictor of recurrence in patients with PaAF (p=0.001), while the height of the RAA, thickness of the crista terminalis, and hypertension were independent predictors of recurrence in PeAF (p<0.05). The ROC curve was used to analysis the predictive model in PaAF and PeAF group, the corresponding sensitivity and specificity were 0.604 and 0.864 in PaAF group, respectively (AUC=0.840, P=0.001), in PeAF group, the corresponding sensitivity and specificity were 0.967 and 0.892, respectively (AUC=0.959, P=0.001). The short diameter of RAA base > 22.15 mm had the highest predictive value for recurrence in PaAF patients, with a sensitivity of 0.887, and a specificity of 0.520 (AUC: 0.743, p=0.001). The RAA height > 28.95 mm has the highest predictive value for recurrence in PeAF, with a sensitivity of 0.633, and a specificity of 0.865 (AUC: 0.816, p=0.001). Recurrent patients both in the PaAF and PeAF groups demonstrated larger RAA base and RA structural parameters. Compared to patients with PaAF, recurrent patients with PeAF presented larger RA volume.

Angle Dependence of Electrode Lead-Related Artifacts in Single- and Dual-Energy Cardiac ECG-Gated CT Scanning-A Phantom Study.

Background: The electrodes of implantable cardiac devices (ICDs) may cause significant problems in cardiac computed tomography (CT) because they are a source of artifacts that obscure surrounding structures and possible pathology. There are a few million patients currently with ICDs, and some of these patients will require cardiac imaging due to coronary artery disease or problems with ICDs. Modern CT scanners can reduce some of the metal artifacts because of MAR software, but in some vendors, it does not work with ECG gating. Introduced in 2008, dual-energy CT scanners can generate virtual monoenergetic images (VMIs), which are much less susceptible to metal artifacts than standard CT images. Objective: This study aimed to evaluate if dual-energy CT can reduce metal artifacts caused by ICD leads by using VMIs. The second objective was to determine how the angle between the electrode and the plane of imaging affects the severity of the artifacts in three planes of imaging. Methods: A 3D-printed model was constructed to obtain a 0-90-degree field at 5-degree intervals between the electrode and each of the planes: axial, coronal, and sagittal. This electrode was scanned in dual-energy and single-energy protocols. VMIs with an energy of 40-140 keV with 10 keV intervals were reconstructed. The length of the two most extended artifacts originating from the tip of the electrode and 2 cm above it-at the point where the thick metallic defibrillating portion of the electrode begins-was measured. Results: For the sagittal plane, these observations were similar for both points of the ICDs that were used as the reference location. VMIs with an energy over 80 keV produce images with fewer artifacts than similar images obtained in the single-energy scanning mode. Conclusions: Virtual monoenergetic imaging techniques may reduce streak artifacts arising from ICD electrodes and improve the quality of the image. Increasing the angle of the electrode as well as the imaging plane can reduce artifacts. The angle between the electrode and the beam of X-rays can be increased by tilting the gantry of the scanner or lifting the upper body of the patient.

MSMHSA-DeepLab V3+: An Effective Multi-Scale, Multi-Head Self-Attention Network for Dual-Modality Cardiac Medical Image Segmentation.

The automatic segmentation of cardiac computed tomography (CT) and magnetic resonance imaging (MRI) plays a pivotal role in the prevention and treatment of cardiovascular diseases. In this study, we propose an efficient network based on the multi-scale, multi-head self-attention (MSMHSA) mechanism. The incorporation of this mechanism enables us to achieve larger receptive fields, facilitating the accurate segmentation of whole heart structures in both CT and MRI images. Within this network, features extracted from the shallow feature extraction network undergo a MHSA mechanism that closely aligns with human vision, resulting in the extraction of contextual semantic information more comprehensively and accurately. To improve the precision of cardiac substructure segmentation across varying sizes, our proposed method introduces three MHSA networks at distinct scales. This approach allows for fine-tuning the accuracy of micro-object segmentation by adapting the size of the segmented images. The efficacy of our method is rigorously validated on the Multi-Modality Whole Heart Segmentation (MM-WHS) Challenge 2017 dataset, demonstrating competitive results and the accurate segmentation of seven cardiac substructures in both cardiac CT and MRI images. Through comparative experiments with advanced transformer-based models, our study provides compelling evidence that despite the remarkable achievements of transformer-based models, the fusion of CNN models and self-attention remains a simple yet highly effective approach for dual-modality whole heart segmentation.

Association between left atrial wall thickness and residual potential after first-pass pulmonary vein isolation using very high-power short-duration ablation in patients with atrial fibrillation

Abstract Background In this study, we evaluated efficacy, efficiency, and safety of pulmonary vein isolation (PVI) applying vHPSD ablation strategy in patients with atrial fibrillation (AF) and explored the association between left atrial wall thickness (LAWT) and the residual potential after first-pass PVI. Methods Between April 2, 2023 and October 6, 2023, drug refractory symptomatic AF patients who underwent AF ablation using vHPSD ablation strategy with QDOT MICRO catheter were prospectively enrolled. All patients took cardiac computed tomography (CT) a day before procedure. Using CT image, LAWT map was created. PV segments were categorized into prespecified 14 segments, and LAWT of each segment, segments with residual potential (RP) after first-pass PVI and early reconnection after 20-minute waiting after PVI were evaluated. Total procedure time, ablation time for PVI, and occurrence of procedure-related complication were collected. Results A total of 40 patients were included (mean age 64 years, 45% of persistent AF). Additional ablation other than PV was performed in 48% of patients, total procedure time was 77 minutes, and total PVI time was 9.9 minutes. With vHPSD PVI, all PVs first pass isolation rate was 60%; in terms of per PV, RSPV first-pass isolation was 88%, followed by LSPV and LIPV 85%, RIPV showed lowest first-pass isolation rate, 75% (Figure A). Most common RP sites was right carina, followed by left carina (Figure A). Compared to historical comparator, vHPSD PVI showed shorter procedure time, shorter ablation time for PVI, with numerically lower but statistically comparable first-pass PVI rate (Figure B). In total 560 PV segments in 40 patients, RSPV anterior, left carina, and right carina showed higher prevalence to be thick PV segments (over 60% of patients), followed by RPSV posterior, LSPV posterior, and RIPV anterior (over 40% of patients). Looking into ablation parameter detail, mean contact force was 10g, mean temperature was 47°C, and mean impedance drop was 8.4Ω. Compared to segments without RP, segments with RP showed thicker mean left atrial wall thickness grades, lower minimum contact force, higher maximum temperature, and lower minimum impedance drop. According to the left atrial wall thickness, thick PV segments showed higher prevalence of RP after first-pass PVI with vHPSD ablation. In segments with LAWT grade 3 or more, the RP rate was 9.1%. This thickness means thicker than 1.5 to 2.0 mm of thickness (Figure C). In 10% of patients (n=4) had audible or tactile steam-pops, visible char was confirmed in 5% of patients (n=2). There were no clinical complications. Conclusion PVI using vHPSD achieved shorter procedure time, shorter ablation time for PVI, and comparable first-pass PVI rate compared to historical comparator using ablation-index guided conventional power ablation. Thick segments had higher chance to have residual potential after first-pass PVI with vHPSD.

Predicting the response to cardiac resynchronization therapy using in silico heart models: pilot study of comparison between in silico and real cardiac resynchronization therapy outcomes

Abstract Background Cardiac resynchronization therapy (CRT) is treatment for patients with heart failure and left ventricular (LV) dyssynchrony. The objective of this study is to evaluate the feasibility of an in silico electromechanical heart model to predict CRT outcomes. Methods The three-dimensional heart geometries of 11 patients (age, 64.0±10.7 years, 4 men) with CRT were constructed from cardiac computed tomography images. We coupled cardiac electrical excitation and mechanical contraction with vascular hemodynamics using a lumped parameter model. We predicted the LV end-diastolic and end-systolic volumes and LV ejection fraction (LVEF) after CRT implantation using the in silico CRT simulation model. We compared the in silico CRT outcomes with the real CRT outcomes. Results Two patients were in silico CRT responders, and those 2 patients were also real CRT responders. Nine patients were in silico CRT non-responders, 8 who were real CRT non-responders, and 1 who was a real CRT responder. The in silico CRT outcomes agreed with the real CRT outcomes in 10 of 11 patients. The positive and negative predictive values and accuracy of the in silico CRT model were 100%, 88.9%, and 90.9%, respectively. Conclusion The in silico CRT simulation model is feasible for predicting real CRT outcomes.In silico CRT simulation model generatioAn example (patient No. 2) of LV (black

Automatic 3D left atrial strain extraction framework on cardiac computed tomography

Background and objectiveStrain analysis provides insights into myocardial function and cardiac condition evaluation. However, the anatomical characteristics of left atrium (LA) inherently limit LA strain analysis when using echocardiography. Cardiac computed tomography (CT) with its superior spatial resolution, has become critical for in-depth evaluation of LA function. Recent studies have explored the feasibility of CT-derived strain; however, they relied on manually selected regions of interest (ROIs) and mainly focused on left ventricle (LV). This study aimed to propose a first-of-its-kind fully automatic deep learning (DL)-based framework for three-dimensional (3D) LA strain extraction on cardiac CT. MethodsA total of 111 patients undergoing ECG-gated contrast-enhanced CT for evaluating subclinical atrial fibrillation (AF) were enrolled in this study. We developed a 3D strain extraction framework on cardiac CT images, containing a 2.5D GN-U-Net network for LA segmentation, axis-oriented 3D view extraction, and LA strain measure. The segmentation accuracy was evaluated using Dice similarity coefficient (DSC). The model-extracted LA volumes and emptying fraction (EF) were compared with ground-truth measurements using intraclass correlation coefficient (ICC), correlation coefficient (r), and Bland-Altman plot (B-A). The automatically extracted LA strains were evaluated against the LA strains measured from 2D echocardiograms. We utilized this framework to gauge the effect of AF burden on LA strain, employing the atrial high rate episode (AHRE) burden as the measurement parameter. ResultsThe GN-U-Net LA segmentation network achieved a DSC score of 0.9603 on the test set. The framework-extracted LA estimates demonstrated excellent ICCs of 0.949 (95 % CI: 0.93–0.97) for minimal LA volume, 0.904 (95 % CI: 0.86–0.93) for maximal LA volume, and 0.902 (95 % CI: 0.86–0.93) for EF, compared with expert measurements. The framework-extracted LA strains demonstrated moderate agreement with the LA strains based on 2D echocardiography (ICCs >0.703). Patients with AHRE > 6 min had significantly lower global strain and LAEF, as extracted by the framework than those with AHRE ≤ 6 min. ConclusionThe promising results highlighted the feasibility and clinical usefulness of automatically extracting 3D LA strain from CT images using a DL-based framework. This tool could provide a 3D-based alternative to echocardiography for assessing LA function.

Diagnostic performance of unenhanced electrocardiogram-gated cardiac CT for detecting myocardial edema.

To assess the diagnostic performance of unenhanced electrocardiogram (ECG)-gated cardiac computed tomography (CT) for detecting myocardial edema, using MRI T2 mapping as the reference standard. This retrospective study protocol was approved by our institutional review board, which waived the requirement for written informed consent. Between December 2017 to February 2019, consecutive patients who had undergone T2 mapping for myocardial tissue characterization were identified. We excluded patients who did not undergo unenhanced ECG-gated cardiac CT within 3 months from MRI T2 mapping or who had poor CT image quality. All patients underwent unenhanced ECG-gated cardiac CT with an axial scan using a third-generation, 320 × 0.5 mm detector-row CT unit. Two radiologists together drew regions of interest (ROIs) in the interventricular septum on the unenhanced ECG-gated cardiac CT images. Using T2 mapping as the reference standard, the diagnostic performance of unenhanced cardiac CT for detecting myocardial edema was evaluated by using the area under the receiver operating characteristic curve with sensitivity and specificity. Youden index was used to find an optimal sensitivity-specificity cutoff point. A cardiovascular radiologist independently performed the measurements, and interobserver reliability was assessed using intraclass correlation coefficients for CT value measurements. A P value of <.05 was considered statistically significant. We included 257 patients who had undergone MRI T2 mapping. Of the 257 patients, 35 patients underwent unenhanced ECG-gated cardiac CT. One patient was excluded from the study because of poor CT image quality. Finally, 34 patients (23 men; age 64.7 ± 14.6 years) comprised our study group. Using T2 mapping, we identified myocardial edema in 19 patients. Mean CT and T2 values for 34 patients were 46.3 ± 2.7 Hounsfield unit and 49.0 ± 4.9 ms, respectively. Mean CT values moderately correlated with mean T2 values (Rho = -0.41; P < .05). Mean CT values provided a sensitivity of 63.2% and a specificity of 93.3% for detecting myocardial edema, with a cutoff value of ≤45.0 Hounsfield unit (area under the receiver operating characteristic curve = 0.77; P < .01). Inter-observer reproducibility in measuring mean CT values was excellent (intraclass correlation coefficient = 0.93; [95% confidence interval: 0.86, 0.96]). Myocardial edema could be detected by CT value of myocardium in unenhanced ECG-gated cardiac CT.

CCT‘S CRYSTAL BALL: PREDICTING CARDIOVASCULAR RISK IN AORTIC VALVE STENOSIS PATIENTS UNDERGOING VALVE REPLACEMENT

Abstract Background and Aims Currently, computed tomography cardiac imaging (CTCI) plays a fundamental role in the pre–procedural planning of degenerative–calcific aortic valve stenosis(AS) correction. However, its prognostic role in assessing mortality and major adverse cardiovascular events (MACE) has not been extensively explored. CTCI allows for the evaluation of coronary artery calcification(CAC), mitral annular calcifications (MAC), aortic valve calcifications (CVA), and thoracic aorta calcifications (CAT). The aim of our study was to assess whether the presence and extent of CAC, MAC, CVA, and CAT could influence non–cardiovascular mortality (NCM), cardiovascular mortality (CM), and MACE in patients undergoing valve replacement(VR). Materials and Methods We collected retrospectivelly clinical and instrumental data and reviewed CTCI using a semi–quantitative approach. CACs were classified according to the visual 2022 Coronary Artery Disease–Reporting and Data System. MACs were defined following the Guerrero et al. study, CVAs according to the Win Registry and CATs were assessed in the ascending aorta, aortic arch, and descending aorta according to the Vos classification. Results 244 patients(59% male, mean age 80.1±6.6 years), with 91% undergoing transfemoral VR, were enrolled(2016–2018). CM was significantly higher in patients with hypertension,cerebro/peripheral vascular disease, stage IV–V renal insufficiency and history of atrial fibrillation (p=0.012,p&amp;lt;0.001,p=0.005, p=0.034 and p=0.024, respectively). CM was also more pronounced in patients with STS SCORE≥6.7±5.2 and EuroSCOREII≥ 5.5±5.8(p=0.016 and p=0.037). Extensive CAC(p=0.019), a MAC score&amp;gt;7(p=0.051), moderate and severe AVC (p=0.029 and p=0.053) and TAC&amp;gt;270° in aortic arch(p=0.019) were linked to CM. In multivariate analysis, a high MAC score and the presence of moderate AVC persisted as risk factors for CM. The presence of a MAC score &amp;gt;7(p=0.006) and the extension of descending TAC &amp;gt;1.5 mm (p=0.034) were associated with an increased risk of NCM.The presence of extensive TAC (≥270°) measured in the aortic arch and descending aorta were associated with MACE (p=0.034 and p=0.0504 respectively). Conclusions CTCI emerges as a crucial tool for evaluating prognostic risk in patients undergoing VR for AS. This type of evaluation can assist clinicians in pre–procedural planning and patient management, allowing for better identification of high–risk patients and optimization of therapeutic strategies

A Novel Data Augmentation Method for Radiomics Analysis Using Image Perturbations

Radiomics extracts hundreds of features from medical images to quantitively characterize a region of interest (ROI). When applying radiomics, imbalanced or small dataset issues are commonly addressed using under or over-sampling, the latter being applied directly to the extracted features. Aim of this study is to propose a novel balancing and data augmentation technique by applying perturbations (erosion, dilation, contour randomization) to the ROI in cardiac computed tomography images. From the perturbed ROIs, radiomic features are extracted, thus creating additional samples. This approach was tested addressing the clinical problem of distinguishing cardiac amyloidosis (CA) from aortic stenosis (AS) and hypertrophic cardiomyopathy (HCM). Twenty-one CA, thirty-two AS and twenty-one HCM patients were included in the study. From each original and perturbed ROI, 107 radiomic features were extracted. The CA-AS dataset was balanced using the perturbation-based method along with random over-sampling, adaptive synthetic (ADASYN) and the synthetic minority oversampling technique (SMOTE). The same methods were tested to perform data augmentation dealing with CA and HCM. Features were submitted to robustness, redundancy, and relevance analysis testing five feature selection methods (p-value, least absolute shrinkage and selection operator (LASSO), semi-supervised LASSO, principal component analysis (PCA), semi-supervised PCA). Support vector machine performed the classification tasks, and its performance were evaluated by means of a 10-fold cross-validation. The perturbation-based approach provided the best performances in terms of f1 score and balanced accuracy in both CA-AS (f1 score: 80%, AUC: 0.91) and CA-HCM (f1 score: 86%, AUC: 0.92) classifications. These results suggest that ROI perturbations represent a powerful approach to address both data balancing and augmentation issues.

Hybrid U-Net and Swin-transformer network for limited-angle cardiac computed tomography

Objective. Cardiac computed tomography (CT) is widely used for diagnosis of cardiovascular disease, the leading cause of morbidity and mortality in the world. Diagnostic performance depends strongly on the temporal resolution of the CT images. To image the beating heart, one can reduce the scanning time by acquiring limited-angle projections. However, this leads to increased image noise and limited-angle-related artifacts. The goal of this paper is to reconstruct high quality cardiac CT images from limited-angle projections. Approach. The ability to reconstruct high quality images from limited-angle projections is highly desirable and remains a major challenge. With the development of deep learning networks, such as U-Net and transformer networks, progresses have been reached on image reconstruction and processing. Here we propose a hybrid model based on the U-Net and Swin-transformer (U-Swin) networks. The U-Net has the potential to restore structural information due to missing projection data and related artifacts, then the Swin-transformer can gather a detailed global feature distribution. Main results. Using synthetic XCAT and clinical cardiac COCA datasets, we demonstrate that our proposed method outperforms the state-of-the-art deep learning-based methods. Significance. It has a great potential to freeze the beating heart with a higher temporal resolution.

Screening, risk stratification and CT imaging for assessment of coronary artery disease

Early treatment of cardiovascular risk factors and characterization of coronary plaques is essential to collect prognostic information about coronary artery disease (CAD) and prevent cardiovascular events. Discussion of the most important risk factors of CAD, basic diagnostic of CAD, prevention, and prognostic factors of CAD with focus on cardiac computed tomography (CT). Prevalence and prognostic impact of CAD risk factors; description of specific assessment of risk profiles; estimation of pretest probability; conventional prevention of CAD; prognostic assessment of CAD using the Calcium Scoring and coronary CT angiography. Assessment of risk profiles and estimation of pretest probability for obstructive coronary stenosis necessitates athorough evaluation of medical history and laboratory values. The composition and extent of calcified and noncalcified plaques in CT exams based on the criteria of the Coronary Artery Disease-Reporting and Data System give important prognostic information about the risk of cardiovascular events, which increases with high plaque burden and vice versa. Initial imaging with CT for evaluation of CAD leads to areduction of invasive coronary angiographies and catheter-associated complications. Besides early detection of cardiovascular risk factors, the additional assessment of plaque burden and significant stenosis in CT gives further prognostic information to facilitate effective therapies to prevent cardiovascular events and in the case of low plaque burden avoid invasive coronary angiography. However, systmatic screening using Calcium Scoring is not established yet due to insufficient data, although it could potentially be used for an early risk stratification in patients with multiple risk factors.

CAR and CSTR Cardiac Computed Tomography (CT) Practice Guidelines: Part 2-Non-Coronary Imaging.

The cardiac computed tomography (CT) practice guidelines provide an updated review of the technological improvements since the publication of the first Canadian Association of Radiologists (CAR) cardiac CT practice guidelines in 2009. An overview of the current evidence supporting the use of cardiac CT in the most common clinical scenarios, standards of practice to optimize patient preparation and safety as well as image quality are described. Coronary CT angiography (CCTA) is the focus of Part I. In Part II, an overview of cardiac CT for non-coronary indications that include valvular and pericardial imaging, tumour and mass evaluation, pulmonary vein imaging, and imaging of congenital heart disease for diagnosis and treatment monitoring are discussed. The guidelines are intended to be relevant for community hospitals and large academic centres with established cardiac CT imaging programs.

Automatic thoracic aorta calcium quantification using deep learning in non-contrast ECG-gated CT images

Thoracic aorta calcium (TAC) can be assessed from cardiac computed tomography (CT) studies to improve cardiovascular risk prediction. The aim of this study was to develop a fully automatic system to detect TAC and to evaluate its performance for classifying the patients into four TAC risk categories. The method started by segmenting the thoracic aorta, combining three UNets trained with axial, sagittal and coronal CT images. Afterwards, the surrounding lesion candidates were classified using three combined convolutional neural networks (CNNs) trained with orthogonal patches. Image datasets included 1190 non-enhanced ECG-gated cardiac CT studies from a cohort of cardiovascular patients (age 57 ± 9 years, 80% men, 65% TAC > 0). In the test set (N = 119), the combination of UNets was able to successfully segment the thoracic aorta with a mean volume difference of 0.3 ± 11.7 ml (<6%) and a median Dice coefficient of 0.947. The combined CNNs accurately classified the lesion candidates and 87% of the patients (N = 104) were accurately placed in their corresponding risk categories (Kappa = 0.826, ICC = 0.9915). TAC measurement can be estimated automatically from cardiac CT images using UNets to isolate the thoracic aorta and CNNs to classify calcified lesions.

Comparison of left atrial appendage measurements between conventional transesophageal echocardiography and "Virtual TEE" reconstructed from computed tomography for pre-procedural planning of device closure.

For pre-procedural planning of left atrial appendage (LAA) closure, sizing is crucial. Although transesophageal echocardiography (TEE) is a standard modality, cardiac computed tomography (CT) is also widely used. The virtual TEE (V-TEE) that our group developed enables us to reconstruct images similar to TEE images from CT images. The software should be helpful to understand and plan the procedure strategy. Accordingly, we investigated the utility of V-TEE. Sixty-six patients at 4 participating sites who completed both CT and TEE prior to LAA closure were included. The LAA diameter at the landing zone (LZ) for WATCHMAN™ device implantation was statistically compared at 0°, 45°, 90°, and 135° between V-TEE and TEE. Among 66 cases, only 3 cases were excluded due to poor imaging quality, and 63 cases were analyzed. The device LZ diameters based on V-TEE were strongly correlated with those based on TEE, despite the significantly greater diameter based on V-TEE with mean differences of 2.4 to 3.0mm (all of them: P < 0.001). The discordances (V-TEE/TEE ratio) at most angles were significantly larger in the elliptical LAAs. V-TEE provides a valuable method for the evaluation of the LAA diameters. V-TEE-based measurements were larger than conventional TEE-based measurements, especially in cases of elliptical LAAs. The assessment by V-TEE has the potential benefit of ensuring proper device sizing regardless of the LAA morphology.

Cardiac Computed Tomography Angiography in Infants and Young Children Without Sedation.

Cardiac computed tomography (CT) imaging plays a pivotal role in the diagnosis and management of infants and young children with congenital heart disease (CHD). While the benefits of CT imaging are well-established, the challenge lies in adapting these procedures to the unique requirements of infants and young children. Traditionally, sedation has been a common practice to ensure cooperation and motion control during imaging. However, using sedation introduces its challenges including potential risks, limitations, and cost implications. In this study, authors explore the feasibility, safety, and diagnostic accuracy of unsedated cardiac CT examinations in infants and young children. This study proves cardiac CT can be performed in India without sedation using simple restraining techniques. This approach aligns with the cultural and familial dynamics prevalent in the country and holds the potential to address economic and infrastructure challenges.

Higher serum phosphate within the normal range is associated with the development of calcified aortic valve disease.

Despite the essential role of ectopic osteogenic calcium-phosphate metabolism in the development of calcific aortic valve disease (CAVD), the implications of high serum phosphate levels in CAVD development are not fully understood. Asymptomatic individuals who underwent health screening using serial cardiac computed tomography (CT) and echocardiography were selected from a multicenter registry. CAVD was identified and quantified on CT images using the aortic valve calcification (AVC) score. The associations between initial serum phosphate levels and the presence of baseline CAVD, development of new CAVD, and the AVC score progression rate were investigated using multivariable regression models. A total of 736 individuals were selected for analysis, and the median interscan duration was 36.4 months. On initial CT, 83 (13.7%) participants had baseline CAVD, while 52 (7.0%) individuals developed new CAVD during follow-up. Serum phosphate levels were not associated with a higher probability of baseline CAVD but were predictive of newly developed CAVD (odds ratio per 1 mg/dl, 1.05; 95% confidence interval, 1.01-1.10; p = 0.02). Higher phosphate levels were also associated with a faster AVC score progression in those with baseline CAVD (regression coefficient per 1 mg/dl, 15.55 Agatston units/year; 95% confidence interval, 6.02-25.07; p < 0.01), an association which remained significant when the analysis was extended to include newly developed CAVD. Even slight elevations in serum phosphate are associated with accelerated CAVD progression from an early stage. Further studies are needed to investigate whether the regulation of phosphate metabolism can slow the progression of CAVD to aortic stenosis.