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.

Abdominal aortic calcium and geometry in patients with essential hypertension.

PURPOSE: Abdominal aorta calcium (AAC) burden and dilatation are associated with an increased risk of mortality. The purpose of this study was to investigate determinants of AAC and abdominal aorta size in patients with essential hypertension. MATERIALS AND METHODS: Patients with uncomplicated essential hypertension who had undergone non-enhanced abdominal CT to rule out secondary hypertension in addition to biological test were recruited between 2010 and 2018. A semi-automatic system was designed to estimate the aortic size (diameter, length, volume) and quantify the AAC from mesenteric artery to bifurcation using the Agatston score. Determinants of aortic size and those related to AAC were searched for using uni- and multivariables analyses. RESULTS: Among 293 randomly selected patients with hypertension (age 52 ± 11 [SD] years) included, 23% had resistant hypertension. Mean abdominal aorta diameter was 20.1 ± 2.1 (SD) mm. Eight (3%) patients had abdominal aorta aneurysm ≥ 30 mm and 58 (20%) had dilated abdominal aorta ≥ 27 mm. Median AAC score was 38 and calcifications were detected in the infra- and supra-renal abdominal aortic portions in 59% and 26% of the patients, respectively. After adjustment for age, male sex and body surface area, abdominal aorta diameter was positively associated with diastolic blood pressure (P = 0.0019). Smoking was the single variable associated with calcified abdominal aorta (P < 0.001) after adjustment for cofactors. In patients with calcifications of abdominal aorta, the score increased with smoking history (P < 0.001), statins treatment (P < 0.01), greater number of anti-hypertensive drugs (P < 0.01), larger abdominal aorta (P < 0.05) and greater systolic blood pressure (P < 0.05). Patients with resistant hypertension had more AAC in the supra-renal abdominal aorta portion than those without resistant hypertension (P < 0.01). CONCLUSION: In patients with essential hypertension, abdominal aorta dilation is related with diastolic blood pressure while AAC is associated with smoking history and resistant hypertension when located to the supra-renal abdominal aorta portion.

Thoracic Aorta Calcium Detection and Quantification Using Convolutional Neural Networks in a Large Cohort of Intermediate-Risk Patients.

Arterial calcification is an independent predictor of cardiovascular disease (CVD) events whereas thoracic aorta calcium (TAC) detection might anticipate extracoronary outcomes. In this work, we trained six convolutional neural networks (CNNs) to detect aortic calcifications and to automate the TAC score assessment in intermediate CVD risk patients. Cardiac computed tomography images from 1415 patients were analyzed together with their aortic geometry previously assessed. Orthogonal patches centered in each aortic candidate lesion were reconstructed and a dataset with 19,790 images (61% positives) was built. Three single-input 2D CNNs were trained using axial, coronal and sagittal patches together with two multi-input 2.5D CNNs combining the orthogonal patches and identifying their best regional combination (BRC) in terms of lesion location. Aortic calcifications were concentrated in the descending (66%) and aortic arch (26%) portions. The BRC of axial patches to detect ascending or aortic arch lesions and sagittal images for the descending portion had the best performance: 0.954 F1-Score, 98.4% sensitivity, 87% of the subjects correctly classified in their TAC category and an average false positive TAC score per patient of 30. A CNN that combined axial and sagittal patches depending on the candidate aortic location ensured an accurate TAC score prediction.

Comprehensive assessment of local and regional aortic stiffness in patients with tricuspid or bicuspid aortic valve aortopathy using magnetic resonance imaging.

BACKGROUND: We aimed to provide a comprehensive aortic stiffness description using magnetic resonance imaging (MRI) in patients with ascending thoracic aorta aneurysm and tricuspid (TAV-ATAA) or bicuspid (BAV) aortic valve. METHODS: This case-control study included 18 TAV-ATAA and 19 BAV patients, with no aortic valve stenosis/severe regurgitation, who were 1:1 age-, gender- and central blood pressures (BP)-matched to healthy volunteers. Each underwent simultaneous aortic MRI and BP measurements. 3D anatomical MRI provided aortic diameters. Stiffness indices included: regional ascending (AA) and descending (DA) aorta pulse wave velocity (PWV) from 4D flow MRI; local AA and DA strain, distensibility and theoretical Bramwell-Hill (BH) model-based PWV, as well as regional arch PWV from 2D flow MRI. RESULTS: Patient groups had significantly higher maximal AA diameter (median[interquartile range], TAV-ATAA: 47.5[42.0-51.3]mm, BAV: 45.0[41.0-47.0]mm) than their respective controls (29.1[26.8-31.8] and 28.1[26.0-32.0]mm, p < 0.0001), while BP were similar (p ≥ 0.25). Stiffness indices were significantly associated with age (ρ ≥ 0.33), mean BP (arch PWV: ρ = 0.25, p = 0.05; DA distensibility: ρ = -0.30, p = 0.02) or AA diameter (arch PWV: ρ = 0.28, p = 0.03; DA PWV: ρ = 0.32, p = 0.009). None of them, however, was significantly different between TAV-ATAA or BAV patients and their matched controls. Finally, while direct PWV measures were significantly correlated to BH-PWV estimates in controls (ρ ≥ 0.40), associations were non-significant in TAV-ATAA and BAV groups (p ≥ 0.18). CONCLUSIONS: The overlap of MRI-derived aortic stiffness indices between patients with TAV or BAV aortopathy and matched controls highlights another heterogeneous feature of aortopathy, and suggests the urgent need for more sensitive indices which might help better discriminate such diseases.

Regional assessment of vascular morphology and hemodynamics: methodology and evaluation for abdominal aortic aneurysms after endovascular repair.

Abdominal aortic aneurysm is a deadly disease that can be treated with different endovascular devices that will distinctly alter the aortic morphology. Computational methods can be used to understand the effect of anatomical changes on aortic hemodynamics. We propose a standardized method to assess morphological and hemodynamic changes of the abdominal aorta through the longitudinal axis of the vessel. Patient-specific CFD simulations were used to quantify these changes for two different endografts before and after surgery. Differences in cross-sectional area, blood pressure, peak blood velocity, wall shear stress, and retrograde blood flow were accurately evidenced with the proposed methodology.

Association of calcium density in the thoracic aorta with risk factors and clinical events.

OBJECTIVES: In the ascending aorta, calcification density was independently and inversely associated with cardiovascular disease (CVD) risk prediction. Until now, the density of thoracic aorta calcium (TAC) was estimated as the Agatston score divided by the calcium area (DAG). We thought to analyze TAC density in a full Hounsfield unit (HU) range and to study its association with TAC volume, traditional risk factors, and CVD events. METHODS: Non-enhanced CT images of 1426 patients at intermediate risk were retrospectively reviewed. A calcium density score was estimated as the average of the maximum HU attenuation in all calcified plaques of the entire thoracic aorta (DAV). RESULTS: During a mean 4.0 years follow-up, there were 26 events for a total of 674 patients with TAC > 0. TAC volume and DAV were positively correlated (R = 0.72). The median DAV value was 457 HU (IQ 323-603 HU) and was exponentially related to DAG (R = 0.86). DAV was inversely associated with systolic pressure (p < 0.05), pulse pressure (p < 0.01), hypertension (p < 0.05), and 10-year FRS (p < 0.001) after adjusting for TAC volume. When TAC volume and DAV were included in a logistic model, a significant improvement was shown in CVD risk estimation beyond coronary artery calcium (CAC) (AUC = 0.768 vs 0.814, p < 0.05). In multivariable Cox models, TAC volume and DAV showed an independent association with CVD. CONCLUSIONS: In intermediate risk patients, TAC density was inversely associated with several risk factors after adjustment for TAC volume. A significant improvement was observed over CAC when TAC volume and density were added into the risk prediction model. KEY POINTS: • Calcifications in the aorta can be non-invasively assessed using CT images • A higher calcium score is associated with a higher cardiovascular risk • Measuring the calcifications size and the density separately can improve the risk prediction.

Automatic correction of background phase offset in 4D-flow of great vessels and of the heart in MRI using a third-order surface model.

OBJECTIVE: To evaluate an automatic correction method for velocity offset errors in cardiac 4D-flow acquisitions. MATERIALS AND METHODS: Velocity offset correction was done in a plane-by-plane scheme and compared to a volumetric approach. Stationary regions were automatically detected. In vitro experiments were conducted in a phantom using two orientations and two encoding velocities (Venc). First- to third-order models were fit to the time-averaged images of the three velocity components. In vivo experiments included realistic ROIs in a volunteer superimposed to a phantom. In 15 volunteers, blood flow volume of the proximal and distal descending aorta, of the pulmonary artery (Qp) and the ascending aorta (Qs) was compared. RESULTS: Offset errors were reduced after correction with a third-order model, yielding residual phantom velocities below 0.6 cm/s and 0.4% of Venc. The plane-by-plane correction method was more effective than the volumetric approach. Mean velocities through superimposed ROIs of a volunteer vs phantom were highly correlated (r2 = 0.96). The significant difference between proximal and distal descending aortic flows was decreased after correction from 8.1 to - 1.4 ml (p < 0.001) and Qp/Qs reduced from 1.08 ± 0.09 to 1.01 ± 0.05. DISCUSSION: An automatic third-order model corrected velocity offset errors in 4D-flow acquisitions, achieving acceptable levels for clinical applications.

Effects on Aortoiliac Fluid Dynamics After Endovascular Sealing of Abdominal Aneurysms.

OBJECTIVES:: To evaluate the effects on aortoiliac fluid dynamics after the implantation of an endograft based on endovascular aneurysm sealing (EVAS) versus endovascular aneurysm repair (EVAR) strategy. METHODS:: An adaptive geometrical deformable model was used for aortic lumen segmentation in 8 patients before and after the surgery. Abdominal aneurysms were treated with an endograft based on the EVAS system (Nellix, n = 4) and with a device based on an anatomical fixation technology (n = 4). Pressure, blood velocity, and wall shear stress (WSS) were estimated at different aortic regions using computational fluid dynamics methods. Physiologic inlet/outlet flow values at the abdominal aorta, the celiac trunk, and the mesenteric and the renal arteries were set. Pressure references were set at iliac arteries outlet. RESULTS:: Maximum aneurysm sizes were similar for both groups in the preoperative scans. The lumen area was lower after EVAR ( P < .05) and EVAS ( P < .01) compared to preoperative aortic lumen sizes. Pressure increase was higher in the proximal abdominal aorta after EVAS compared to EVAR (2.3 ± 0.3 mm Hg vs 0.9 ± 0.3 mm Hg, P < .001). Peak blood velocities inside the endografts were 3-fold higher for EVAS compared to EVAR (54 ± 5 cm/s vs 17 ± 4 cm/s, P < .01). Velocities at the iliac arteries also remained higher for EVAS (38 ± 4 cm/s vs 24 ± 4 cm/s, P < .05). Peak WSS at the iliac arteries remained higher for EVAS compared to EVAR group ( P < .05). CONCLUSION:: The significant modification of the aortic bifurcation anatomy after EVAS alters aortoiliac fluid dynamics, showing a pressure impact at the renal arteries level and an acceleration of the blood velocity at the iliac region with a concomitant increase in peak WSS.

Association of thoracic aorta calcium and non cardiac vascular events in cardiac disease-free individuals.

OBJECTIVE: Thoracic aorta calcium (TAC) is measurable on the same computed tomography (CT) scan as coronary artery calcium (CAC) but has still unclear clinical value. We assessed TAC and CAC relations with non-cardiac vascular events history in a cohort of subjects at risk for cardiovascular disease. METHODS: We analyzed retrospectively 1000 consecutive subjects having undergone CAC detection by non-contrast multi-slice CT with measurement field longer than usual in order to measure total TAC including aortic arch calcium. We also determined partial TAC restricted to ascending and descending thoracic aorta sites by removing arch calcium from total TAC. Calcium deposits were measured with a custom made software using Agatston score. RESULTS: Compared with the rest of the cohort, the 30 subjects with non-cardiac vascular event history had higher median values [95% CI] of total TAC (282 [28-1809] vs 39 [0-333], p < 0.01) and partial TAC (4 [0-284] vs 0 [0-5], p < 0.01) but no different value of CAC (73 [0-284] vs 16 [0-148]). Odds ratio [95% CI] of having non-cardiac vascular event per 1-SD increase in log-transformed calcium value was significant for total TAC but not for CAC, if total TAC and CAC were entered separately (1.56 [1.12-2.24], p < 0.01 and 1.13 [0.86-1.50], respectively) or together (1.57 [1.10-2.32], p < 0.01 and 0.98 [0.73-1.32], respectively) in the logistic adjusted model. CONCLUSION: TAC assessment simultaneous with CAC detection provides complementary information on the extra coronary component of cardiovascular risk beyond CAC's coronary risk prediction. Further studies are required to prospectively confirm this result.

Deformable Surface Model for the Evaluation of Abdominal Aortic Aneurysms Treated with an Endovascular Sealing System.

Rupture of abdominal aortic aneurysms (AAA) is responsible for 1-3% of all deaths among the elderly population in developed countries. A novel endograft proposes an endovascular aneurysm sealing (EVAS) system that isolates the aneurysm wall from blood flow using a polymer-filled endobag that surrounds two balloon-expandable stents. The volume of injected polymer is determined by monitoring the endobag pressure but the final AAA expansion remains unknown. We conceived and developed a fully deformable surface model for the comparison of pre-operative sac lumen size and final endobag size (measured using a follow-up scan) with the volume of injected polymer. Computed tomography images were acquired for eight patients. Aneurysms were manually and automatically segmented twice by the same observer. The injected polymer volume resulted 9% higher than the aneurysm pre-operative lumen size (p < 0.05), and 11% lower than the final follow-up endobag volume (p < 0.01). The automated method required minimal user interaction; it was fast and used a single set of parameters for all subjects. Intra-observer and manual vs. automated variability of measured volumes were 0.35 ± 2.11 and 0.07 ± 3.04 mL, respectively. Deformable surface models were used to quantify AAA size and showed that EVAS system devices tended to expand the sac lumen size.

Aging impact on thoracic aorta 3D morphometry in intermediate-risk subjects: looking beyond coronary arteries with non-contrast cardiac CT.

An increasing number of intermediate risk asymptomatic subjects benefit from measures of atherosclerosis burden like coronary artery calcification studies with non-contrast heart computed tomography (CT). However, additional information can be derived from these studies, looking beyond the coronary arteries and without exposing the patients to further radiation. We report a semi-automatic method that objectively assesses ascending, arch and descending aorta dimension and shape from non-contrast CT datasets to investigate the effect of aging on thoracic aorta geometry. First, the segmentation process identifies the vessel centerline coordinates following a toroidal path for the curvilinear portion and axial planes for descending aorta. Then, reconstructing oblique planes orthogonal to the centerline direction, it iteratively fits circles inside the vessel cross-section. Finally, regional thoracic aorta dimensions (diameter, volume and length) and shape (vessel curvature and tortuosity) are calculated. A population of 200 normotensive men was recruited. Length, mean diameter and volume differed by 1.2 cm, 0.13 cm and 21 cm(3) per decade of life, respectively. Aortic shape uncoiled with aging, reducing its tortuosity and increasing its radius of curvature. The arch was the most affected segment. In conclusion, non-contrast cardiac CT imaging can be successfully employed to assess thoracic aorta 3D morphometry.