Feasibility and prognostic role of machine learning-based FFRCT in patients with stent implantation.

OBJECTIVES: To investigate the feasibility and prognostic implications of coronary CT angiography (CCTA) derived fractional flow reserve (FFRCT) in patients who have undergone stents implantation. METHODS: Firstly, the feasibility of FFRCT in stented vessels was validated. The diagnostic performance of FFRCT in identifying hemodynamically in-stent restenosis (ISR) in 33 patients with invasive FFR ≤ 0.88 as reference standard, intra-group correlation coefficient (ICC) between FFRCT and FFR was calculated. Secondly, prognostic value was assessed with 115 patients with serial CCTA scans after PCI. Stent characteristics (location, diameter, length, etc.), CCTA measurements (minimum lumen diameter [MLD], minimum lumen area [MLA], ISR), and FFRCT measurements (FFRCT, ΔFFRCT, ΔFFRCT/stent length) both at baseline and follow-up were recorded. Longitudinal analysis included changes of MLD, MLA, ISR, and FFRCT. The primary endpoint was major adverse cardiovascular events (MACE). RESULTS: Per-patient accuracy of FFRCT was 0.85 in identifying hemodynamically ISR. FFRCT had a good correlation with FFR (ICC = 0.84). 15.7% (18/115) developed MACE during 25 months since follow-up CCTA. Lasso regression identified age and follow-up ΔFFRCT/length as candidate variables. In the Cox proportional hazards model, age (hazard ratio [HR], 1.102 [95% CI, 1.032-1.177]; p = 0.004) and follow-up ΔFFRCT/length (HR, 1.014 [95% CI, 1.006-1.023]; p = 0.001) were independently associated with MACE (c-index = 0.856). Time-dependent ROC analysis showed AUC was 0.787 (95% CI, 0.594-0.980) at 25 months to predict adverse outcome. After bootstrap validation with 1000 resamplings, the bias-corrected c-index was 0.846. CONCLUSIONS: Noninvasive ML-based FFRCT is feasible in patients following stents implantation and shows prognostic value in predicting adverse events after stents implantation in low-moderate risk patients. KEY POINTS: • Machine-learning-based FFRCT is feasible to evaluate the functional significance of in-stent restenosis in patients with stent implantation. • Follow-up △FFRCT along with the stent length might have prognostic implication in patients with stent implantation and low-to-moderate risk after 2 years follow-up. The prognostic role of FFRCT in patients with moderate-to-high or high risk needs to be further studied. • FFRCT might refine the clinical pathway of patients with stent implantation to invasive catheterization.

Thyroid gland delineation in noncontrast-enhanced CTs using deep convolutional neural networks.

The purpose of this study is to develop a deep learning method for thyroid delineation with high accuracy, efficiency, and robustness in noncontrast-enhanced head and neck CTs. The cross-sectional analysis consisted of six tests, including randomized cross-validation and hold-out experiments, tests of prediction accuracy between cancer and benign and cross-gender analysis were performed to evaluate the proposed deep-learning-based performance method. CT images of 1977 patients with suspected thyroid carcinoma were retrospectively investigated. The automatically segmented thyroid gland volume was compared against physician-approved clinical contours using metrics, the Pearson correlation and Bland-Altman analysis. Quantitative metrics included: the Dice similarity coefficient (DSC), sensitivity, specificity, Jaccard index (JAC), Hausdorff distance (HD), mean surface distance (MSD), residual mean square distance (RMSD) and the center of mass distance (CMD). The robustness of the proposed method was further tested using the nonparametric Kruskal-Wallis test to assess the equality of distribution of DSC values. The proposed method's accuracy remained high through all the tests, with the median DSC, JAC, sensitivity and specificity higher than 0.913, 0.839, 0.856 and 0.979, respectively. The proposed method also resulted in median MSD, RMSD, HD and CMD, of less than 0.31 mm, 0.48 mm, 2.06 mm and 0.50 mm, respectively. The MSD and RMSD were 0.40 ± 0.29 mm and 0.70 ± 0.46 mm, respectively. Concurrent testing of the proposed method with 3D U-Net and V-Net showed that the proposed method had significantly improved performance. The proposed deep-learning method achieved accurate and robust performance through six cross-sectional analysis tests.

Automatic quantification of myocardium and pericardial fat from coronary computed tomography angiography: a multicenter study.

OBJECTIVES: To develop a deep learning-based method for simultaneous myocardium and pericardial fat quantification from coronary computed tomography angiography (CCTA) for the diagnosis and treatment of cardiovascular disease (CVD). METHODS: We retrospectively identified CCTA data obtained between May 2008 and July 2018 in a multicenter (six centers) CVD study. The proposed method was evaluated on 422 patients' data by two studies. The first overall study involves training model on CVD patients and testing on non-CVD patients, as well as training on non-CVD patients and testing on CVD patients. The second study was performed using the leave-center-out approach. The method performance was evaluated using Dice similarity coefficient (DSC), Jaccard index (JAC), 95% Hausdorff distance (HD95), mean surface distance (MSD), residual mean square distance (RMSD), and the center of mass distance (CMD). The robustness of the proposed method was tested using the nonparametric Kruskal-Wallis test and post hoc test to assess the equality of distribution of DSC values among different tests. RESULTS: The automatic segmentation achieved a strong correlation with contour (ICC and R > 0.97, p value < 0.001 throughout all tests). The accuracy of the proposed method remained high through all the tests, with the median DSC higher than 0.88 for pericardial fat and 0.96 for myocardium. The proposed method also resulted in mean MSD, RMSD, HD95, and CMD of less than 1.36 mm for pericardial fat and 1.00 mm for myocardium. CONCLUSIONS: The proposed deep learning-based segmentation method enables accurate simultaneous quantification of myocardium and pericardial fat in a multicenter study. KEY POINTS: • Deep learning-based myocardium and pericardial fat segmentation method tested on 422 patients' coronary computed tomography angiography in a multicenter study. • The proposed method provides segmentations with high volumetric accuracy (ICC and R > 0.97, p value < 0.001) and similar shape as manual annotation by experienced radiologists (median Dice similarity coefficient ≥ 0.88 for pericardial fat and 0.96 for myocardium).

Prognostic implication of CT-FFR based functional SYNTAX score in patients with de novo three-vessel disease.

AIMS: This study was aimed at investigating whether a machine learning (ML)-based coronary computed tomographic angiography (CCTA) derived fractional flow reserve (CT-FFR) SYNTAX score (SS), 'Functional SYNTAX score' (FSSCTA), would predict clinical outcome in patients with three-vessel coronary artery disease (CAD). METHODS AND RESULTS: The SS based on CCTA (SSCTA) and ICA (SSICA) were retrospectively collected in 227 consecutive patients with three-vessel CAD. FSSCTA was calculated by combining the anatomical data with functional data derived from a ML-based CT-FFR assessment. The ability of each score system to predict major adverse cardiac events (MACE) was compared. The difference between revascularization strategies directed by the anatomical SS and FSSCTA was also assessed. Two hundred and twenty-seven patients were divided into two groups according to the SSCTA cut-off value of 22. After determining FSSCTA for each patient, 22.9% of patients (52/227) were reclassified to a low-risk group (FSSCTA ≤ 22). In the low- vs. intermediate-to-high (>22) FSSCTA group, MACE occurred in 3.2% (4/125) vs. 34.3% (35/102), respectively (P < 0.001). The independent predictors of MACE were FSSCTA (OR = 1.21, P = 0.001) and diabetes (OR = 2.35, P = 0.048). FSSCTA demonstrated a better predictive accuracy for MACE compared with SSCTA (AUC: 0.81 vs. 0.75, P = 0.01) and SSICA (0.81 vs. 0.75, P < 0.001). After FSSCTA was revealed, 52 patients initially referred for CABG based on SSCTA would have been changed to PCI. CONCLUSION: Recalculating SS by incorporating lesion-specific ischaemia as determined by ML-based CT-FFR is a better predictor of MACE in patients with three-vessel CAD. Additionally, the use of FSSCTA may alter selected revascularization strategies in these patients.

Automatic segmentation and quantification of epicardial adipose tissue from coronary computed tomography angiography.

Epicardial adipose tissue (EAT) is a visceral fat deposit, that's known for its association with factors, such as obesity, diabetes mellitus, age, and hypertension. Segmentation of the EAT in a fast and reproducible way is important for the interpretation of its role as an independent risk marker intricate. However, EAT has a variable distribution, and various diseases may affect the volume of the EAT, which can increase the complexity of the already time-consuming manual segmentation work. We propose a 3D deep attention U-Net method to automatically segment the EAT from coronary computed tomography angiography (CCTA). Five-fold cross-validation and hold-out experiments were used to evaluate the proposed method through a retrospective investigation of 200 patients. The automatically segmented EAT volume was compared with physician-approved clinical contours. Quantitative metrics used were the Dice similarity coefficient (DSC), sensitivity, specificity, Jaccard index (JAC), Hausdorff distance (HD), mean surface distance (MSD), residual mean square distance (RMSD), and the center of mass distance (CMD). For cross-validation, the median DSC, sensitivity, and specificity were 92.7%, 91.1%, and 95.1%, respectively, with JAC, HD, CMD, MSD, and RMSD are 82.9% ± 8.8%, 3.77 ± 1.86 mm, 1.98 ± 1.50 mm, 0.37 ± 0.24 mm, and 0.65 ± 0.37 mm, respectively. For the hold-out test, the accuracy of the proposed method remained high. We developed a novel deep learning-based approach for the automated segmentation of the EAT on CCTA images. We demonstrated the high accuracy of the proposed learning-based segmentation method through comparison with ground truth contour of 200 clinical patient cases using 8 quantitative metrics, Pearson correlation, and Bland-Altman analysis. Our automatic EAT segmentation results show the potential of the proposed method to be used in computer-aided diagnosis of coronary artery diseases (CADs) in clinical settings.