Cross-sectional angle prediction of lipid-rich and calcified tissue on computed tomography angiography images.

PURPOSE: The assessment of vulnerable plaque characteristics and distribution is important to stratify cardiovascular risk in a patient. Computed tomography angiography (CTA) offers a promising alternative to invasive imaging but is limited by the fact that the range of Hounsfield units (HU) in lipid-rich areas overlaps with the HU range in fibrotic tissue and that the HU range of calcified plaques overlaps with the contrast within the contrast-filled lumen. This paper is to investigate whether lipid-rich and calcified plaques can be detected more accurately on cross-sectional CTA images using deep learning methodology. METHODS: Two deep learning (DL) approaches are proposed, a 2.5D Dense U-Net and 2.5D Mask-RCNN, which separately perform the cross-sectional plaque detection in the Cartesian and polar domain. The spread-out view is used to evaluate and show the prediction result of the plaque regions. The accuracy and F1-score are calculated on a lesion level for the DL and conventional plaque detection methods. RESULTS: For the lipid-rich plaques, the median and mean values of the F1-score calculated by the two proposed DL methods on 91 lesions were approximately 6 and 3 times higher than those of the conventional method. For the calcified plaques, the F1-score of the proposed methods was comparable to those of the conventional method. The median F1-score of the Dense U-Net-based method was 3% higher than that of the conventional method. CONCLUSION: The two methods proposed in this paper contribute to finer cross-sectional predictions of lipid-rich and calcified plaques compared to studies focusing only on longitudinal prediction. The angular prediction performance of the proposed methods outperforms the convincing conventional method for lipid-rich plaque and is comparable for calcified plaque.

A novel alignment procedure to assess calcified coronary plaques in histopathology, post-mortem computed tomography angiography and optical coherence tomography.

PURPOSE: Improve mapping and registration of longitudinal view on histopathology vessels in a three-dimensional alignment procedure for postmortem quantitative coronary plaque analyses. This new procedure is applied and results shown using calcified coronary plaque analyses within post-mortem computed tomography angiography (PMCTA), optical coherence tomography (OCT) and the gold standard of histopathology. RESULTS: In total, 338 annotated histopathology images were included, 166 PMCTA transversal images and 285 OCT images were aligned in the comparison. The results from the comparison using the alignment procedure showed overall that the calcified plaques seem to be overestimated by PMCTA and underestimated by OCT. CONCLUSIONS: The 3D fusion approach, aligning the images of PMCTA, OCT and histopathology as gold standard allowed for a slice-based comparison of the different modalities. The results showed that PMCTA overestimates the calcified plaques while OCT underestimates these, compared to histopathology.

A model-guided method for improving coronary artery tree extractions from CCTA images.

PURPOSE: Automatically extracted coronary artery trees (CATs) from coronary computed tomography angiography images could contain incorrect extractions which require manual corrections before they can be used in clinical practice. A model-guided method for improving the extracted CAT is described to automatically detect potential incorrect extractions and improve them. METHODS: The proposed method is a coarse-to-fine approach. A coarse improvement is first applied on all vessels in the extracted CAT, and then a fine improvement is applied only on vessels with higher clinical significance. Based upon a decision tree, the proposed method automatically and iteratively performs improvement operations for the entire extracted CAT until it meets the stop criteria. The improvement in the extraction quality obtained by the proposed method is measured using a scoring system. 18 datasets were used to determine optimal values for the parameters involved in the model-guided method and 122 datasets were used for evaluation. RESULTS: Compared to the initial automatic extractions, the proposed method improves the CATs for 122 datasets from an average quality score of 87 ± 6 to 93 ± 4. The developed method is able to run within 2 min on a typical workstation. The difference in extraction quality after automatic improvement is negatively correlated with the initial extraction quality (R = - 0.694, P < 0.001). CONCLUSION: Without deteriorating the initially extracted CATs, the presented method automatically detects incorrect extractions and improves the CATs to an average quality score of 93 guided by anatomical statistical models.

Impact of diabetes on coronary artery plaque volume by coronary CT angiography and subsequent adverse cardiac events.

BACKGROUND: To investigate the impact of diabetes on coronary artery total plaque volume (TPV) and adverse events in long-term follow-up. METHODS: One-hundred-and-eight diabetic patients were matched to 324 non-diabetic patients, with respect to age, sex, body-mass index, hypertension, smoking habits, LDL and HDL cholesterol, family history for CAD as well as aspirin and statin medication. In all patients, TPV was quantified from coronary CT angiographies (CTA) using dedicated software. All-cause mortality, acute coronary syndrome and late revascularisation (>90 days) served as combined endpoint. RESULTS: Patients were followed for 5.6 years. The endpoint occurred in 18 (16.7%) diabetic and 26 (8.0%) non-diabetic patients (odds ratio 2.3, p = 0.03). Diabetic patients had significantly higher TPV than non-diabetic patients (55.1 mm³ [IQR: 6.2 and 220.4 mm³] vs. 24.9 mm³ [IQR: 0 and 166.7 mm³], p = 0.02). A TPV threshold of 110.5 mm³ provided good separation of diabetic and non-diabetic patients at higher and lower risk for adverse events. Noteworthy, diabetic and non-diabetic patients with a TPV<110.5 mm³ had comparable outcome (hazard ratio: 1.3, p = 0.59), while diabetic patients with TPV>110.5 mm³ had significantly higher incidence of adverse events (hazard ratio 2.3, p = 0.03) compared to non-diabetic patients with TPV>110.5 mm³. There was incremental prognostic value in diabetic and non-diabetic patients over the Framingham Risk Score (Integrated Discrimination Improvement: 0.052 and 0.012, p for both <0.05). CONCLUSION: Diabetes is associated with significantly higher TPV, which is independent of other CAD risk factors. Quantification of TPV improves the identification of diabetic patients at higher risk for future adverse events.

Comparative Effectiveness of CT-Derived Atherosclerotic Plaque Metrics for Predicting Myocardial Ischemia.

OBJECTIVES: This study sought to investigate the performance of various cardiac computed tomography (CT)-derived atherosclerotic plaque metrics for predicting provocable myocardial ischemia. BACKGROUND: The association of coronary arterial diameter stenosis with myocardial ischemia is only modest, but cardiac CT provides several other, readily available atherosclerosis metrics, which may have incremental value. METHODS: The study analyzed 873 nonstented coronary arteries and their myocardial perfusion territories in 356 patients (mean 62 years of age) enrolled in the CORE320 (Coronary Artery Evaluation using 320-row Multidetector Computed Tomography Angiography and Myocardial Perfusion) study. Myocardial perfusion defects in static CT perfusion imaging were graded at rest and after adenosine in 13 myocardial segments using a 4-point scale. The summed difference score was calculated by subtracting the summed rest score from the summed stress score. Reversible ischemia was defined as summed difference score ≥1. In a sensitivity analysis, results were also provided using single-photon emission computed tomography (SPECT) as the reference standard. Vessel based predictor variables included maximum percent diameter stenosis, lesion length, coronary calcium score, maximum cross-sectional calcium arc, percent atheroma volume (PAV), low-attenuation atheroma volume, positive (external) vascular remodeling, and subjective impression of "vulnerable plaque." The study used logistic regression models to assess the association of plaque metrics with myocardial ischemia. RESULTS: In univariate analysis, all plaque metrics were associated with reversible ischemia. In the adjusted logistic model, only maximum percent diameter stenosis (1.26; 95% confidence interval: 1.15 to 1.38) remained an independent predictor. With SPECT as outcome variable, PAV and "vulnerable" plaque remained predictive after adjustment. In vessels with intermediate stenosis (40% to 70%), no single metric had clinically meaningful incremental value. CONCLUSIONS: Various plaque metrics obtained by cardiac CT predict provocable myocardial ischemia by CT perfusion imaging through their association with maximum percent stenosis, while none had significant incremental value. With SPECT as reference standard, PAV and "vulnerable plaque" remained predictors of ischemia after adjustment but the predictive value added to stenosis assessment alone was small.

Diagnostic performance of hyperaemic myocardial blood flow index obtained by dynamic computed tomography: does it predict functionally significant coronary lesions?

AIMS: The severity of coronary artery narrowing is a poor predictor of functional significance, in particular in intermediate coronary lesions (30-70% diameter narrowing). The aim of this work was to compare the performance of a quantitative hyperaemic myocardial blood flow (MBF) index derived from adenosine dynamic computed tomography perfusion (CTP) imaging with that of visual CT coronary angiography (CTCA) and semi-automatic quantitative CT (QCT) in the detection of functionally significant coronary lesions in patients with stable chest pain. METHODS AND RESULTS: CTCA and CTP were performed in 80 patients (210 analysable coronary vessels) referred to invasive coronary angiography (ICA). The MBF index (mL/100 mL/min) was computed using a model-based parametric deconvolution method. The diagnostic performance of the MBF index in detecting functionally significant coronary lesions was compared with visual CTCA and QCT. Coronary lesions with invasive fractional flow reserve of ≤0.75 were defined as functionally significant. The optimal cut-off value of the MBF index to detect functionally significant coronary lesions was 78 mL/100 mL/min. On a vessel-territory level, the MBF index had a larger area under the curve (0.95; 95% confidence interval [95% CI]: 0.92-0.98) compared with visual CTCA (0.85; 95% CI: 0.79-0.91) and QCT (0.89; 95% CI: 0.84-0.93) (both P-values <0.001). In the analysis restricted to intermediate coronary lesions, the specificity of visual CTCA (69%) and QCT (77%) could be improved by the subsequent use of the MBF index (89%). CONCLUSION: In this proof-of-principle study, the MBF index performed better than visual CTCA and QCT in the identification of functionally significant coronary lesions. The MBF index had additional value beyond CTCA anatomy in intermediate coronary lesions. This may have a potential to support patient management.

Automated quantification of stenosis severity on 64-slice CT: a comparison with quantitative coronary angiography.

OBJECTIVES: This study sought to demonstrate the feasibility of a dedicated algorithm for automated quantification of stenosis severity on multislice computed tomography in comparison with quantitative coronary angiography (QCA). BACKGROUND: Limited information is available on quantification of coronary stenosis, and previous attempts using semiautomated approaches have been suboptimal. METHODS: In patients who had undergone 64-slice computed tomography and invasive coronary angiography, the most severe lesion on QCA was quantified per coronary artery using quantitative coronary computed tomography (QCCTA) software. Additionally, visual grading of stenosis severity using a binary approach (50% stenosis as a cutoff) was performed. Diameter stenosis (percentage) was obtained from detected lumen contours at the minimal lumen area, and corresponding reference diameter values were obtained from an automatic trend analysis of the vessel areas within the artery. RESULTS: One hundred patients (53 men; 59.8 +/- 8.0 years) were evaluated, and 282 (94%) vessels were analyzed. Good correlations for diameter stenosis were observed for vessel-based (n = 282; r = 0.83; p < 0.01) and patient-based (n = 93; r = 0.86; p < 0.01) analyses. Mean differences between QCCTA and QCA were -3.0% +/- 12.3% and -6.2% +/- 12.4%. Furthermore, good agreement was observed between QCCTA and QCA for semiquantitative assessment of diameter stenosis (accuracy of 95%). Diagnostic accuracy for assessment of > or =50% diameter stenosis was higher using QCCTA compared with visual analysis (95% vs. 87%; p = 0.08). Moreover, a significantly higher positive predictive value was observed with QCCTA when compared with visual analysis (100% vs. 78%; p < 0.05). Although the visual approach showed a reduced diagnostic accuracy for data sets with moderate image quality, QCCTA performed equally well in patients with moderate or good image quality. However, in data sets with good image quality, QCCTA tended to have a reduced sensitivity compared with visual analysis. CONCLUSIONS: Good correlations were found for quantification of stenosis severity between QCCTA and QCA. QCCTA showed an improved positive predictive value when compared with visual analysis.