Coronary CT angiography derived features for predicting an abnormal pet myocardial perfusion imaging: a machine learning approach

Background

Computed Tomography Coronary Angiography (CTCA) is an effective non-invasive imaging modality for anatomo-functional assessment of coronary artery disease (CAD). Machine learning (ML) algorithms allow extraction and process of useful information from multidimensional spaces for evaluation of coronary lesions.

Purpose

To investigate the ability of ML to integrate computational fluid dynamics (CFD) derived parameters with quantitative plaque burden, plaque morphology and anatomical characteristics for predicting impaired myocardial flow reserve by PET myocardial perfusion imaging (MPI).

Methods

49 patients (29 male, mean age 65.3±6.3 years) with intermediate pre-test likelihood of CAD who underwent CTCA and PET-MPI were included. PET was considered positive when >1 contiguous segment demonstrated Myocardial flow reserve (MFR) ≤2.5 mL/g/min for 15O-water or ≤2.0 for 13N-ammonia respectively. CDF derived parameters such as a previously validated CT-FFR surrogate, virtual functional assessment index (vFAI), segmental endothelial shear stress (ESS), as well as anatomical and plaque characteristics were assessed. k-nearest neighbor (k-NN), support vector machines (SVM) and feedforward neural networks (FF-NN) were implemented. ML was internally validated using 5-fold cross validation, repeated 100 times. Using sequential forward selection (SFS), the 5 highest rank features based on appearances in each classification scheme were selected and following exhaustive search (ES) the best features combinations were identified. Each classifier's performance was evaluated using an area-under-receiver operating characteristic curve (AUC) analysis.

Results

85 coronary segments were analyzed and 28 features derived from CTCA were extracted. The features ranking for every classifier are depicted in Figure 1. k-NN using a combination only of ESS in the proximal (ESSprox) and distal segment achieved an AUC=0.78 (Sens=0.71, Spec=0.77, p<0.05) for predicting a positive PET result. Combining ESSprox with burden fibrofatty tissue and non-calcified plaque burden, SVM achieved an AUC=0.75 (Sens=0.74, Spec=0.67, p<0.05) whilst for FF-NN, the corresponding AUC was 0.79 (Sens=0.76, Spec=0.7, p<0.05) using ESSprox, vFAI and % Fibrofatty volume. Among the best features combinations, ESSprox was the most consistent one achieving an AUC=0.75 (Sens=0.66, Spec=0.73, p<0.05) for k-NN, AUC=0.73 (Sens=0.58, Spec=0.59, p<0.05), for SVM and an AUC=0.73 (Sens=0.63, Spec=0.62, p<0.05) for FF-NN respectively.

Conclusion

ML analysis is feasible for predicting abnormal MFR by PET using a combination of CFD derived parameters, anatomical and morphological features. ESSprox was present in every combination of best features. As a single characteristic was a moderate predictor of impaired MFR, whilst in combination with plaque characteristics and CFD derived features resulted in improved sensitivity and specificity.

Figure 1

Funding Acknowledgement

Type of funding source: Public grant(s) – EU funding. Main funding source(s): This research is co-financed by Greece and the European Union (European Social Fund-ESF) through the Operational Programme “Human, Resources Development, Education and Lifelong Learning 2014-2020” in the context of the project “Assessment of coronary atherosclerosis: a new complete, anatomo-functional, morphological and biomechanical approach” and from p-Med GR 5002802

Elevated TG/HDL-C ratio is an independent predictor of outcome and it is associated with CAD progression in patients with stable coronary artery disease

Background

Elevated TG/HDL-C ratio is associated with CVD outcomes in high-risk populations presenting for coronary angiography, but studies were limited in gender-specific populations or in pts with ACS.

Purpose

Aim of this study was to evaluate the prognostic role of TG/HDL-C levels and their association with CAD progression in pts with suspected stable CAD.

Methods

TG/HDL-C ratio was calculated in 545 pts (60±9yrs,330males) with symptoms of stable CAD enrolled in the EVINCI study. 490 pts underwent coronary CTA to assess the presence of CAD (>50%stenosis) and entered a clinical follow up (4.5±0.9yrs). The CVD outcome measure included all cause mortality, non fatal MI, hospitalization for unstable angina or HF. After 6±1yrs, during the SMARTool study, a second CTA was obtained in 171 EVINCI pts and a CTA risk score (based on plaque extent, severity, composition, and location) was calculated at enrolment and at follow up to assess CAD progression (ΔCTA score).

Results

Pts were divided according to TG/HDL-C quartiles: IQ (<1.32), IIQ (1.32–2.03), IIIQ (2.04–3.33), and IVQ (>3.33). As reported in Table, the frequency of male, diabetes, metabolic syndrome and obesity increased among quartiles. Glucidic biomarkers progressively increased from quartile I to IV, while LDL-C decreased. The prevalence of obstructive CAD at CTA did not differ among groups. The CVD endpoint occurred in 7% of pts. At multivariable analyses, high TG/HDL-C ratio (IVQ) was associated with the outcome endpoint independently from presence of obstructive CAD and treatment (HR 3.477, 95% CI 1.181–10.239, P=0.0237). CTA score was significantly higher in pts in IVQ compared to IQ at both SMARTool enrolment and follow up (Figure1). A significantly higher ΔCTA score was observed in pts in III-IVQ compared with those in I-IIQ (Figure2).

Conclusion

Elevated TG/HDL-C ratio is an independent predictor of outcome and it is associated with CAD progression in patients with stable CAD.

Funding Acknowledgement

Type of funding source: Public grant(s) – EU funding. Main funding source(s): “EValuation of INtegrated Cardiac Imaging” - EVINCI [GA number: 222915]; “Simulation Modeling of coronary ARTery disease: a tool for clinical decision support - SMARTool” [GA number: 689068]

Relationship between endothelial shear stress, plaque burden and stenosis severity and their comparative performance in predicting impaired coronary vasodilation by pet myocardial perfusion imaging

Background

Advances in CTCA imaging enable assessment of coronary plaque burden, a predictor of myocardial perfusion abnormalities and more recently, with the use of computational fluid dynamics (CFD) of endothelial shear stress (ESS), an established contributor to atherosclerotic plaque development and progression.

Purpose

To investigate the relationship of local endothelial shear stress (ESS) and plaque burden (PB) between them and with stenosis severity as well as their comparative performance in predicting impaired coronary vasodilating capability assessed by PET myocardial perfusion imaging (MPI).

Methods

49 patients (29 males, mean age 65.3±6.3 years, intermediate pre-test likelihood of coronary artery disease, CAD), who underwent PET-MPI with 15O-water or 13N-ammonia and CTCA were included. PET was considered abnormal when >1 contiguous segment showed both stress Myocardial Blood Flow ≤2.3 mL/g/min and Myocardial Flow Reserve ≤2.5 for 15O-water or <1.79 mL/g/min and ≤2.0 for 13N-ammonia respectively. On CTCA, stenosis (sten) severity was classified as: <30%, 31–50%, 51–70% and 71–90%. CFD were applied to every vessel, assuming a mean pressure of 100 mmHg as the inlet boundary condition and a coronary velocity profile of 1 ml/sec as the outlet. ESS was calculated for the full length of a stenosis (total), as well as in the proximal (prox), minimum lumen area (MLA) and distal (dist) stenotic segments. Atherosclerotic PB was defined as lesion plaque volume/lesion vessel volume ×100.

Results

85 coronary vessels were evaluated. There was a positive correlation between ESS and PB (r(total)=0.544, r(prox)=0.528, r(MLA)=0.529, r(dist)=0.474, p<0.001 for all). All ESS indices and PB increased progressively with stenosis severity compared to segments with a <30% stenosis (p≤0.004 for all comparisons). ESS indices and PB were also higher in lesions demonstrating impaired vasodilating capacity compared to those without (p≤0.02 for all comparisons, figure 1). All ESS indices performed equally with PB and sten >50% in predicting an abnormal PET MPI, (AUC: from 0.682 to 0.780, p-diff >0.5 for all comparisons). The pairwise combination of sten >50% with the ESS segments, except the distal one, increased the predictive ability of the model over stenosis alone (AUC (sten >50% + ESS(total)) = 0.80, AUC (sten >50% + ESS(prox)) = 0.797, AUC (sten >50% + ESS(MLA)) = 0.822, p-diff ≤0.01 for all comparisons, AUC (sten >50% + ESS(dist)) = 0.768, p-diff=0.07).

Conclusion

There is a low to moderate positive association between lesion plaque burden and ESS indices. Like PB, ESS increases progressively with stenosis severity and is higher in lesions paired with abnormal PET results. ESS is a moderate predictor of impaired vasodilating capability, performing equally with PB and stenosis severity. The addition of ESS to stenosis severity can improve prediction of an abnormal PET result.

Figure 1

Funding Acknowledgement

Type of funding source: Public grant(s) – EU funding. Main funding source(s): This study is co-financed by Greece and the European Union (European Social Fund-ESF) through the Operational Programme “Human Resources Development, Education and Lifelong Learning 2014-2020” in the context of the project “Assessment of coronary atherosclerosis: a new complete, anatomo-functional, morphological and biomechanical approach” and from p-Med GR 5002802

Sex differences in the natural history of plaque progression by serial coronary computed tomography angiography

Background

Sex related differences exist for coronary artery disease (CAD). Women tend to be older when presenting with CAD and have lower rates of obstructive disease. Invasive intravascular ultrasound studies have shown differences in plaque composition between males and females. However, these studies were performed in a high risk population needing invasive imaging. Coronary computed tomography angiography (CTA) allows for a fast and non-invasive quantification of CAD in low risk patients. Sex differences and quantitative analysis of plaque progression and changes in plaque composition have not been studied intensively.

Purpose

To evaluate the role of sex on long term plaque progression and on the change of plaque composition in a population with low-intermediate risk.

Methods

Patients that received a coronary CTA were prospectively included in the SMARTool study to receive a follow-up coronary CTA. In total, 275 patients from 5 European countries were recruited in 7 centers. Baseline and follow-up coronary CTA were quantitative analyzed on a per-lesion basis using dedicated software package. Patients without coronary plaques at follow-up or with uninterpretable coronary CTA results were excluded. Total plaque volume and compositional volumes, calcified or non-calcified (defined as fibrous, fibro-fatty or necrotic core), were normalized using the vessel volume to calculate a percentage atheroma volume (PAV). Lesions between males and females were compared using linear mixed models. We further classified patients into age groups <55 and ≥55 years to evaluate the influence of menopause on plaque progression.

Results

In total, 211 patients were included in this analysis, 146 (69%) were male and 65 (31%) were female. Mean interscan period was 6.2±1.4 years. Females were older (64±7 vs 61±8 years; p<0.001), had higher HDL levels (56±15 vs 49±15 mg/dL; p=0.003) and presented more often with atypical chest pain (62 vs 38%; p=0.017). Males had 434 plaque sites and females 156. On a per-lesion analysis females had less fibro-fatty PAV compared to males (β −1.3±0.4%; p<0.001), no other differences were seen (p>0.05). When stratifying the patients in above and below 55 years old, females still had less fibro-fatty PAV compared to males in both age groups (p<0.05). However, females in the age group <55 years showed more regression of fibrous PAV compared to males (β −0.8±0.3% per year; p=0.002) and non-calcified plaque PAV (β −0.7±0.3% per year; p=0.027) (Figure).

Conclusions

Males have larger fibro-fatty PAV compared to females, however the rate of change did not differ. Younger women showed more regression of fibrous PAV and non-calcified PAV compared to males. No differences in the rate of plaque progression or plaque composition changes were seen between males and females in the older age group.

Figure plaque progression and sex diff

Funding Acknowledgement

Type of funding source: Public grant(s) – EU funding. Main funding source(s): EU H2020 research and innovation program

Simulation of atherosclerotic plaque growth using computational biomechanics and patient-specific data

Atherosclerosis is the one of the major causes of mortality worldwide, urging the need for prevention strategies. In this work, a novel computational model is developed, which is used for simulation of plaque growth to 94 realistic 3D reconstructed coronary arteries. This model considers several factors of the atherosclerotic process even mechanical factors such as the effect of endothelial shear stress, responsible for the initiation of atherosclerosis, and biological factors such as the accumulation of low and high density lipoproteins (LDL and HDL), monocytes, macrophages, cytokines, nitric oxide and formation of foams cells or proliferation of contractile and synthetic smooth muscle cells (SMCs). The model is validated using the serial imaging of CTCA comparing the simulated geometries with the real follow-up arteries. Additionally, we examine the predictive capability of the model to identify regions prone of disease progression. The results presented good correlation between the simulated lumen area (P < 0.0001), plaque area (P < 0.0001) and plaque burden (P < 0.0001) with the realistic ones. Finally, disease progression is achieved with 80% accuracy with many of the computational results being independent predictors.

Site specific prediction of PCI stenting based on imaging and biomechanics data using gradient boosting tree ensembles

Cardiovascular diseases are nowadays considered as the main cause of morbidity and mortality worldwide. Coronary Artery Disease (CAD), the most typical form of cardiovascular disease is diagnosed by a variety of imaging modalities, both invasive and non-invasive, which involve either risk implications or high cost. Therefore, several attempts have been undertaken to early diagnose and predict either the high CAD risk patients or the cardiovascular events, implementing machine learning techniques. The purpose of this study is to present a classification scheme for the prediction of Percutaneous Coronary Intervention (PCI) stenting placement, using image-based data. The proposed classification model is a gradient boosting classifier, incorporated into a class imbalance handling technique, the Easy ensemble scheme and aims to classify coronary segments into high CAD risk and low CAD risk, based on their PCI placement. Through this study, we investigate the importance of image based features, concluding that the combination of the coronary degree of stenosis and the fractional flow reserve achieves accuracy 78%.

Impact of Clinical Characteristics and Statins on Coronary Plaque Progression by Serial Computed Tomography Angiography

Background

Progression of coronary artery disease using serial coronary computed tomography angiography (CTA) is of clinical interest. Our primary aim was to prospectively assess the impact of clinical characteristics and statin use on quantitatively assessed coronary plaque progression in a low-risk study population during long-term follow-up.

Methods

Patients who previously underwent coronary CTA for suspected coronary artery disease were prospectively included to undergo follow-up coronary CTA. The primary end point was coronary artery disease progression, defined as the absolute annual increase in total, calcified, and noncalcified plaque volume by quantitative CTA analysis.

Results

In total, 202 patients underwent serial coronary CTA with a mean interscan period of 6.2±1.4 years. On a per-plaque basis, increasing age (β=0.070; P =0.058) and hypertension (β=1.380; P =0.075) were nonsignificantly associated with annual total plaque progression. Male sex (β=1.676; P =0.009), diabetes mellitus (β=1.725; P =0.012), and statin use (β=1.498; P =0.046) showed an independent association with annual progression of calcified plaque. While hypertension (β=2.259; P =0.015) was an independent determinant of noncalcified plaque progression, statin use (β=−2.178; P =0.050) was borderline significantly associated with a reduced progression of noncalcified plaque.

Conclusions

Statin use was associated with an increased progression of calcified coronary plaque and a reduced progression of noncalcified coronary plaque, potentially reflecting calcification of the noncalcified plaque component. Whereas hypertension was the only modifiable risk factor predictive of noncalcified plaque progression, diabetes mellitus mainly led to an increase in calcified plaque. These findings could yield the need for intensified preventive treatment of patients with diabetes mellitus and hypertension to slow and stabilize coronary artery disease progression and improve clinical outcome.

The effect of error propagation in the 3D reconstruction of coronary segments using CTCA images on crucial hemodynamic parameters

The development of 3D reconstruction methods of the coronary vasculature has gained substantial ground during the past years. The accurate 3D reconstruction is of utmost importance because the propagation of errors caused by either equipment calibration errors, human errors or other error sources can seriously affect the computation of critical hemodynamic parameters such as Endothelial Shear Stress, intracoronary pressures etc. In this work, we present a study on how the 3D reconstruction error can affect the subsequent blood flow simulations in 3D coronary arterial models. Eight arterial segments were reconstructed, creating the control models and were then modified in order to create an underestimated and an overestimated model of the same segment using a 5% error. Cross-sectional ESS values, as well as, smartFFR values were calculated to examine the effect of the reconstruction error. As it was expected, the underestimated models presented with higher ESS values and lower smartFFR values, whereas the overestimated models presented with lower ESS values and higher smartFFR values, respectively.

P2702Assessment of endothelial shear stress and functional significance of coronary lesions by computed tomography coronary angiography (CTCA) and computational fluid dynamics: a comparison with PET

Background

The feasibility of assessing endothelial shear stress (ESS) in coronary lesions by non-invasive imaging and its potential role in different clinical settings has been recently explored. However, the relationship of ESS with functional indices derived by computed tomography coronary angiography (CTCA) and its value in predicting perfusion changes by quantitative positron emission tomography (PET) downstream stenotic vessels has not been assessed.

Purpose

To investigate the feasibility of calculating local ESS, its relationship with stenosis severity as well as with virtual functional assessment index (vFAI), and the comparative performance of the two parameters for predicting impaired coronary vasodilating capability in terms both of stress myocardial blood flow (MBF) and myocardial flow reserve (MFR) in patients submitted to CTCA.

Methods

Thirty-two patients (23 male-9 female, mean age 65.6±7.2 years) with intermediate pre-test likelihood of coronary artery disease (CAD), who were enrolled in the EVINCI and SMARTool projects, and had undergone CTCA with vFAI and PET myocardial perfusion imaging with 15 O-water or 13 N-ammonia were included in the study. PET was considered positive when >1 contiguous segments showed both stress MBF ≤2.3 mL/g/min and MFR ≤2.5 for 15 O-water or ≤1.79 mL/g/min and ≤2.0, for 13 N-ammonia respectively. A vFAI threshold of 0.85 was used as predictor of impaired vasodilating capability. ESS computation was based on a mean aortic pressure of 100 mmHg for the inlet and a mean blood flow at rest of 0.00105 kg/s for the outlet. ESS was calculated (Pa) in the full length of the stenosis and the mean value was obtained.

Results

Hybrid imaging analysis was performed in CTCA and PET datasets. 51 coronary segments were assessed. There were 27 lesions with stenosis 31–50% and 24 lesions with stenosis 51–70%. ESS was higher in the latter (20.4, IQ: 11.4–32.1 vs. 10.4, IQ: 5.5–15.7, p=0.04). Similarly, ESS was higher in stenoses with impaired vasodilating capacity compared to those without, although this difference did not reach statistical significance (22.8, IQ: 13.2–35.1 vs. 10.6, IQ: 5.7–22.1, p=0.10). The ROC curve analysis for prediction of both abnormal stress MBF and MFR followed the same pattern (AUC=0.668, 95% confidence interval (CI): 0.490–0.810, p=0.11).On the other hand, there was a moderate negative correlation (r=−0.41, p=0.004) between ESS and vFAI and the former was lower in stenoses with vFAI >0.85 compared to those below this threshold (7.35, IQ: 3.2–13.9 vs. 19.1, IQ: 14.1–32.8, p=0.012). vFAI was a good predictor of coronary flow capacity (AUC=0.737, CI: 0.58–0.85, p=0.02).

Conclusion

Calculation of ESS is feasible in CTCA datasets. ESS was related with stenosis severity and there was a trend to be higher in lesions with impaired coronary vasodilating capability. ESS is modestly related with vFAI and may also be an additional predictor of impaired regional myocardial flow obtained by PET imaging.

Acknowledgement/Funding

This study has received funding from the EU H2020 research and innovation programme under grant agreement No 689068 and from p-Med GR 5002802

P6167Low leptin plasma levels are associated with progression of coronary atherosclerosis in patients with stable coronary artery disease from the SMARTool Study

Background

Leptin is an adipokine involved in energy homeostasis and has been related with established vascular risk factors. However, studies on the association of leptin plasma levels with coronary artery disease (CAD) have yielded conflicting results.

Purpose

Aim of the present study was to evaluate the association between leptin plasma levels and presence, severity and progression of coronary atherosclerosis in patients with suspected stable CAD.

Methods

In a cohort of 257 patients with symptoms of stable CAD enrolled in the SMARTool study, coronary computed tomography angiography (CTA), plasma leptin levels and clinical and bio-humoral CAD risk profile (including glucose, lipid and inflammation variables) were obtained at enrolment and after 6±1yrs of follow-up. Sixty-four patients were revascularized and the remaining 193 represent the population for the present study. CTA findings were categorised as no-minimal CAD (<30% stenosis), non-obstructive CAD (30%-50% stenosis) and obstructive CAD (≥50% stenosis in at least one major coronary vessel). A CTA risk score (based on plaque extent, severity, composition, and location) was calculated at baseline and at follow-up to assess coronary atherosclerotic burden and its progression (Δ CTA score≥5).

Results

CTA findings showed obstructive CAD in 11% of patients at baseline and in 15% at follow-up (p<0.0001). CTA risk score, was 8.03±7.80 at baseline and increased to 10.33±8.17 at follow-up (p<0.0001) with CAD progression in 20% of patients. Leptin plasma levels were inversely related with CTA findings both at baseline and follow-up (Figure). In a Cox model, baseline plasma leptin was an independent predictor of CAD progression, after adjustment for clinical risk factors, biomarkers, and treatment (HR 0.572, 95% CI 0.393–0.834, P=0.0037).

Figure 1

Conclusion

Plasma leptin is inversely associated with coronary atherosclerotic burden and disease progression in patients with stable CAD. This association is independent of known factors affecting leptin levels. These results could prompt further investigations on the pathophysiological mechanisms of this association.

Acknowledgement/Funding

EU H2020 research and innovation program under grant agreement No 689068

A three-dimensional quantification of calcified and non-calcified plaques in coronary arteries based on computed tomography coronary angiography images: Comparison with expert's annotations and virtual histology intravascular ultrasound

The detection, quantification and characterization of coronary atherosclerotic plaques has a major effect on the diagnosis and treatment of coronary artery disease (CAD). Different studies have reported and evaluated the noninvasive ability of Computed Tomography Coronary Angiography (CTCA) to identify coronary plaque features. The identification of calcified plaques (CP) and non-calcified plaques (NCP) using CTCA has been extensively studied in cardiovascular research. However, NCP detection remains a challenging problem in CTCA imaging, due to the similar intensity values of NCP compared to the perivascular tissue, which surrounds the vasculature. In this work, we present a novel methodology for the identification of the plaque burden of the coronary artery and the volumetric quantification of CP and NCP utilizing CTCA images and we compare the findings with virtual histology intravascular ultrasound (VH-IVUS) and manual expert's annotations. Bland-Altman analyses were employed to assess the agreement between the presented methodology and VH-IVUS. The assessment of the plaque volume, the lesion length and the plaque area in 18 coronary lesions indicated excellent correlation with VH-IVUS. More specifically, for the CP lesions the correlation of plaque volume, lesion length and plaque area was 0.93, 0.84 and 0.85, respectively, whereas the correlation of plaque volume, lesion length and plaque area for the NCP lesions was 0.92, 0.95 and 0.81, respectively. In addition to this, the segmentation of the lumen, CP and NCP in 1350 CTCA slices indicated that the mean value of DICE coefficient is 0.72, 0.7 and 0.62, whereas the mean HD value is 1.95, 1.74 and 1.95, for the lumen, CP and NCP, respectively.

Integrative and comparative genomic analyses identify clinically relevant pulmonary carcinoid groups and unveil the supra-carcinoids

The worldwide incidence of pulmonary carcinoids is increasing, but little is known about their molecular characteristics. Through machine learning and multi-omics factor analysis, we compare and contrast the genomic profiles of 116 pulmonary carcinoids (including 35 atypical), 75 large-cell neuroendocrine carcinomas (LCNEC), and 66 small-cell lung cancers. Here we report that the integrative analyses on 257 lung neuroendocrine neoplasms stratify atypical carcinoids into two prognostic groups with a 10-year overall survival of 88% and 27%, respectively. We identify therapeutically relevant molecular groups of pulmonary carcinoids, suggesting DLL3 and the immune system as candidate therapeutic targets; we confirm the value of OTP expression levels for the prognosis and diagnosis of these diseases, and we unveil the group of supra-carcinoids. This group comprises samples with carcinoid-like morphology yet the molecular and clinical features of the deadly LCNEC, further supporting the previously proposed molecular link between the low- and high-grade lung neuroendocrine neoplasms.

Characterization of functionally significant coronary artery disease by a coronary computed tomography angiography-based index: a comparison with positron emission tomography

Aims

To test the hypothesis that virtual functional assessment index (vFAI) is related with regional flow parameters derived by quantitative positron emission tomography (PET) and can be used to assess abnormal vasodilating capability in coronary vessels with stenotic lesions at coronary computed tomography angiography (CCTA).

Methods and results

vFAI, stress myocardial blood flow (MBF), and myocardial flow reserve (MFR) were assessed in 78 patients (mean age 62.2 ± 7.7 years) with intermediate pre-test likelihood of coronary artery disease (CAD). Coronary stenoses ≥50% were considered angiographically significant. PET was considered positive for significant CAD, when more than one contiguous segments showed stress MBF ≤2.3 mL/g/min for 15O-water or <1.79 mL/g/min for 13N-ammonia. MFR thresholds were ≤2.5 and ≤2.0, respectively. vFAI was lower in vessels with abnormal stress MBF (0.76 ± 0.10 vs. 0.89 ± 0.07, P < 0.001) or MFR (0.80 ± 0.10 vs. 0.89 ± 0.07, P < 0.001). vFAI had an accuracy of 78.6% and 75% in unmasking abnormal stress MBF and MFR in 15O-water and 82.7% and 71.2% in 13N-ammonia studies, respectively. Addition of vFAI to anatomical CCTA data increased the ability for predicting abnormal stress MBF and MFR in 15O-water studies [AUCccta + vfai = 0.866, 95% confidence interval (CI) 0.783–0.949; P = 0.013 and AUCccta + vfai = 0.737, 95% CI 0.648–0.825; P = 0.007, respectively]. An incremental value was also demonstrated for prediction of stress MBF (AUCccta + vfai = 0.887, 95% CI 0.799–0.974; P = 0.001) in 13N-ammonia studies. A similar trend was recorded for MFR (AUCccta + vfai = 0.780, 95% CI 0.632–0.929; P = 0.13).

Conclusion

vFAI identifies accurately the presence of impaired vasodilating capability. In combination with anatomical data, vFAI enhances the diagnostic performance of CCTA.

Numerical simulation of stent deployment within patient-specific artery and its validation against clinical data

BACKGROUND AND OBJECTIVE

One of the most widely adopted endovascular treatment procedures is the stent implantation. The effectiveness of the treatment depends on the appropriate stent expansion. However, it is difficult to accurately predict the outcome of such an endovascular intervention. Numerical simulations represent a useful tool to study the complex behavior of the stent during deployment. This study presents a numerical model capable of simulating this process.

METHODS

The numerical model consists of three parts: modeling of stent expansion, modeling the interaction of the stent with the arterial wall and the deformation of the arterial wall. The model is able to predict the shapes of both stent and arterial wall during the entire deployment process. Simulations are performed using patient-specific clinical data that ensures more realistic results.

RESULTS

The numerical simulations of stent deployment are performed using the extracted geometry of the coronary arteries of two patients. The obtained results are validated against clinical data from the follow up examination and both quantitative and qualitative analysis of the results is presented. The areas of several slices of the arterial wall are calculated for all the three states (before, after and follow up) and the standard error of the area when comparing simulation and follow up examination is 5.27% for patient #1 and 4.5% for patient #2.

CONCLUSIONS

The final goal of numerical simulations in stent deployment should be to provide a clinical tool that is capable of reliably predicting the treatment outcome. This study showed through the good agreement of results of the numerical simulations and clinical data that the presented numerical model represents a step towards this final goal. These simulations can also provide valuable information about distribution of forces and stress in the arterial wall that can improve pre-operative planning and treatment optimization.

A Machine Learning Approach for the Prediction of the Progression of Cardiovascular Disease based on Clinical and Non-Invasive Imaging Data

Nowadays, cardiovascular diseases are very common and are considered as the main cause of morbidity and mortality worldwide. Coronary Artery Disease (CAD), the most typical cardiovascular disease is diagnosed by a variety of medical imaging modalities, which involve costs and complications. Therefore, several attempts have been undertaken to early diagnose and predict CAD status and progression through machine learning approaches. The purpose of this study is to present a machine learning technique for the prediction of CAD, using image-based data and clinical data. We investigate the effect of vascular anatomical features of the three coronary arteries on the graduation of CAD. A classification model is built to predict the future status of CAD, including cases of "no CAD" patients, "non-obstructive CAD" patients and "obstructive CAD" patients. The best accuracy was achieved by the implementation of a tree-based classifier, J48 classifier, after a ranking feature selection methodology. The majority of the selected features are the vessel geometry derived features, among the traditional risk factors. The combination of geometrical risk factors with the conventional ones constitutes a novel scheme for the CAD prediction.

Thermal Balance and Dialysis Hypotension

Many Studies Have Confirmed Our Original Observation That Dialysate T Set At About 35° C Affords A Better Hemodynamic Protection Than The Standard Dialysate T Of 37-38° C. In This Review We Present Some New Data On The Hemodynamic Mechanism Of The Protective Effect Of Cold Dialysis On Blood Pressure. The Study Was Based On Serial Assessment Of The Percent Changes Occurring During Dialysis Treatment In Estimated Stroke Volume (Aortic Blood Flow Determined By Doppler Echocardiography), Blood Volume (Hemoglobinometry), Arterial Pressure (Dynamap), And Heart Rate (Ecg), From Which Cardiac Output (Co) Indexes And Total Peripheral Vascular Resistances (Tpvr) Were Derived. Of The 14 Pts Studied, 7 Showed A Drop In Mean Arterial Pressure (Map) Of 25° Or Greater During Standard Dialysis (Unstable Patients). Compared With The 7 Patients Having More Stable Intradialysis Map, Unstable Pts Showed Greater Reduction In Co Which Was Disproportionately Greater Than The Reduction In Blood Volume, And A Paradoxical Decrease In Tpvr, The Difference Being Highly Significant (P ≤ 0.01 For Both Changes). When Crossed-Over To Cold Dialysis, Along With A Significantly Lower Reduction In Map (P ≤ 0.01) The Unstable Pts Showed A Lower Decrease In Co Which Paralleled The Reduction In Blood Volume, And An Increase In Tpvr. These Changes Were Highly Significant (P ≤ 0.01). Data Suggest That Dialysis Hypotension Is Characterized By An Impaired Venous Return, Probably Due To The Peripheral Blood Pooling (Increased Ratio Between The ‘Unstressed’ And ‘Stressed’ Blood Volume) Associated With The Decrease In Tpvr. Exposure Of Extracorporeal Blood To Cold Dialysate Favours The Venous Return To The Heart By Increasing Tpvr And The ‘Stressed’ Blood Volume.

Percutaneous cardiac support during myocardial infarction drastically reduces mortality: perspectives from a swine model

BACKGROUND/AIMS

Acute myocardial infarction (AMI) with cardiogenic shock (CS) remains the leading cause of in-hospital death in acute coronary syndromes. In the AMI-CS pig model we tested the efficacy of temporary percutaneous cardiorespiratory assist device (PCRA) in rescuing the failing heart and reducing early mortality.

METHODS

In open-chest pigs we induced AMI by proximal left anterior descending coronary artery (LAD) ligation. Eight animals without PCRA (C group) were compared with 12 animals otherwise treated with PCRA (T group), starting approximately at 60 minutes post-occlusion and lasting 120-180 minutes. In 3 animals of the T group, regional myocardial oxygen content was also imaged by two-dimensional near infrared spectroscopy (2D-NIRS) with and without PCRA, before and after LAD reperfusion.

RESULTS

All animals without PCRA died despite unrelenting resuscitation maneuvers (120 minutes average survival time). Conversely, animals treated with PCRA showed a reduction in life-threatening arrhythmia and maintenance of aortic pressure, allowing interruption of PCRA in all cases early in the experiments, with sound hemodynamics at the end of the observation period. During LAD occlusion, NIRS showed severe de-oxygenation of the LAD territory that improved with PCRA. After PCRA suspension and LAD reperfusion, the residual de-oxygenated area proved to be smaller than the initial risk area.

CONCLUSIONS

In AMI, PCRA initiated during advanced CS drastically reduced early mortality from 100% to 0% in a 4-5 hour observation period. PCRA promoted oxygenation of the ischemic area during LAD occlusion. Results support the use of PCRA as first line of treatment in AMI-CS, improving myocardial rescue and short-term survival.