AI-based volumetric six-tissue body composition quantification from CT cardiac attenuation scans enhances mortality prediction: multicenter study.

Background: Computed tomography attenuation correction (CTAC) scans are routinely obtained during cardiac perfusion imaging, but currently only utilized for attenuation correction and visual calcium estimation. We aimed to develop a novel artificial intelligence (AI)-based approach to obtain volumetric measurements of chest body composition from CTAC scans and evaluate these measures for all-cause mortality (ACM) risk stratification. Methods: We applied AI-based segmentation and image-processing techniques on CTAC scans from a large international image-based registry (four sites), to define chest rib cage and multiple tissues. Volumetric measures of bone, skeletal muscle (SM), subcutaneous, intramuscular (IMAT), visceral (VAT), and epicardial (EAT) adipose tissues were quantified between automatically-identified T5 and T11 vertebrae. The independent prognostic value of volumetric attenuation, and indexed volumes were evaluated for predicting ACM, adjusting for established risk factors and 18 other body compositions measures via Cox regression models and Kaplan-Meier curves. Findings: End-to-end processing time was <2 minutes/scan with no user interaction. Of 9918 patients studied, 5451(55%) were male. During median 2.5 years follow-up, 610 (6.2%) patients died. High VAT, EAT and IMAT attenuation were associated with increased ACM risk (adjusted hazard ratio (HR) [95% confidence interval] for VAT: 2.39 [1.92, 2.96], p<0.0001; EAT: 1.55 [1.26, 1.90], p<0.0001; IMAT: 1.30 [1.06, 1.60], p=0.0124). Patients with high bone attenuation were at lower risk of death as compared to subjects with lower bone attenuation (adjusted HR 0.77 [0.62, 0.95], p=0.0159). Likewise, high SM volume index was associated with a lower risk of death (adjusted HR 0.56 [0.44, 0.71], p<0.0001). Interpretations: CTAC scans obtained routinely during cardiac perfusion imaging contain important volumetric body composition biomarkers which can be automatically measured and offer important additional prognostic value.

Holistic AI analysis of hybrid cardiac perfusion images for mortality prediction.

Background: While low-dose computed tomography scans are traditionally used for attenuation correction in hybrid myocardial perfusion imaging (MPI), they also contain additional anatomic and pathologic information not utilized in clinical assessment. We seek to uncover the full potential of these scans utilizing a holistic artificial intelligence (AI)-driven image framework for image assessment. Methods: Patients with SPECT/CT MPI from 4 REFINE SPECT registry sites were studied. A multi-structure model segmented 33 structures and quantified 15 radiomics features for each on CT attenuation correction (CTAC) scans. Coronary artery calcium and epicardial adipose tissue scores were obtained from separate deep-learning models. Normal standard quantitative MPI features were derived by clinical software. Extreme Gradient Boosting derived all-cause mortality risk scores from SPECT, CT, stress test, and clinical features utilizing a 10-fold cross-validation regimen to separate training from testing data. The performance of the models for the prediction of all-cause mortality was evaluated using area under the receiver-operating characteristic curves (AUCs). Results: Of 10,480 patients, 5,745 (54.8%) were male, and median age was 65 (interquartile range [IQR] 57-73) years. During the median follow-up of 2.9 years (1.6-4.0), 651 (6.2%) patients died. The AUC for mortality prediction of the model (combining CTAC, MPI, and clinical data) was 0.80 (95% confidence interval [0.74-0.87]), which was higher than that of an AI CTAC model (0.78 [0.71-0.85]), and AI hybrid model (0.79 [0.72-0.86]) incorporating CTAC and MPI data (p<0.001 for all). Conclusion: In patients with normal perfusion, the comprehensive model (0.76 [0.65-0.86]) had significantly better performance than the AI CTAC (0.72 [0.61-0.83]) and AI hybrid (0.73 [0.62-0.84]) models (p<0.001, for all).CTAC significantly enhances AI risk stratification with MPI SPECT/CT beyond its primary role - attenuation correction. A comprehensive multimodality approach can significantly improve mortality prediction compared to MPI information alone in patients undergoing cardiac SPECT/CT.

Deep Learning-Enabled Quantification of 99mTc-Pyrophosphate SPECT/CT for Cardiac Amyloidosis.

Transthyretin cardiac amyloidosis (ATTR CA) is increasingly recognized as a cause of heart failure in older patients, with 99mTc-pyrophosphate imaging frequently used to establish the diagnosis. Visual interpretation of SPECT images is the gold standard for interpretation but is inherently subjective. Manual quantitation of SPECT myocardial 99mTc-pyrophosphate activity is time-consuming and not performed clinically. We evaluated a deep learning approach for fully automated volumetric quantitation of 99mTc-pyrophosphate using segmentation of coregistered anatomic structures from CT attenuation maps. Methods: Patients who underwent SPECT/CT 99mTc-pyrophosphate imaging for suspected ATTR CA were included. Diagnosis of ATTR CA was determined using standard criteria. Cardiac chambers and myocardium were segmented from CT attenuation maps using a foundational deep learning model and then applied to attenuation-corrected SPECT images to quantify radiotracer activity. We evaluated the diagnostic accuracy of target-to-background ratio (TBR), cardiac pyrophosphate activity (CPA), and volume of involvement (VOI) using the area under the receiver operating characteristic curve (AUC). We then evaluated associations with the composite outcome of cardiovascular death or heart failure hospitalization. Results: In total, 299 patients were included (median age, 76 y), with ATTR CA diagnosed in 83 (27.8%) patients. CPA (AUC, 0.989; 95% CI, 0.974-1.00) and VOI (AUC, 0.988; 95% CI, 0.973-1.00) had the highest prediction performance for ATTR CA. The next highest AUC was for TBR (AUC, 0.979; 95% CI, 0.964-0.995). The AUC for CPA was significantly higher than that for heart-to-contralateral ratio (AUC, 0.975; 95% CI, 0.952-0.998; P = 0.046). Twenty-three patients with ATTR CA experienced cardiovascular death or heart failure hospitalization. All methods for establishing TBR, CPA, and VOI were associated with an increased risk of events after adjustment for age, with hazard ratios ranging from 1.41 to 1.84 per SD increase. Conclusion: Deep learning segmentation of coregistered CT attenuation maps is not affected by the pattern of radiotracer uptake and allows for fully automatic quantification of hot-spot SPECT imaging such as 99mTc-pyrophosphate. This approach can be used to accurately identify patients with ATTR CA and may play a role in risk prediction.

Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography.

Chest computed tomography is one of the most common diagnostic tests, with 15 million scans performed annually in the United States. Coronary calcium can be visualized on these scans, but other measures of cardiac risk such as atrial and ventricular volumes have classically required administration of contrast. Here we show that a fully automated pipeline, incorporating two artificial intelligence models, automatically quantifies coronary calcium, left atrial volume, left ventricular mass, and other cardiac chamber volumes in 29,687 patients from three cohorts. The model processes chamber volumes and coronary artery calcium with an end-to-end time of ~18 s, while failing to segment only 0.1% of cases. Coronary calcium, left atrial volume, and left ventricular mass index are independently associated with all-cause and cardiovascular mortality and significantly improve risk classification compared to identification of abnormalities by a radiologist. This automated approach can be integrated into clinical workflows to improve identification of abnormalities and risk stratification, allowing physicians to improve clinical decision-making.

Automated vessel-specific coronary artery calcification quantification with deep learning in a large multi-centre registry.

AIMS: Vessel-specific coronary artery calcification (CAC) is additive to global CAC for prognostic assessment. We assessed accuracy and prognostic implications of vessel-specific automated deep learning (DL) CAC analysis on electrocardiogram (ECG) gated and attenuation correction (AC) computed tomography (CT) in a large multi-centre registry. METHODS AND RESULTS: Vessel-specific CAC was assessed in the left main/left anterior descending (LM/LAD), left circumflex (LCX), and right coronary artery (RCA) using a DL model trained on 3000 gated CT and tested on 2094 gated CT and 5969 non-gated AC CT. Vessel-specific agreement was assessed with linear weighted Cohen's Kappa for CAC zero, 1-100, 101-400, and >400 Agatston units (AU). Risk of major adverse cardiovascular events (MACE) was assessed during 2.4 ± 1.4 years follow-up, with hazard ratios (HR) and 95% confidence intervals (CI). There was strong to excellent agreement between DL and expert ground truth for CAC in LM/LAD, LCX and RCA on gated CT [0.90 (95% CI 0.89 to 0.92); 0.70 (0.68 to 0.73); 0.79 (0.77 to 0.81)] and AC CT [0.78 (0.77 to 0.80); 0.60 (0.58 to 0.62); 0.70 (0.68 to 0.71)]. MACE occurred in 242 (12%) undergoing gated CT and 841(14%) of undergoing AC CT. LM/LAD CAC >400 AU was associated with the highest risk of MACE on gated (HR 12.0, 95% CI 7.96, 18.0, P < 0.001) and AC CT (HR 4.21, 95% CI 3.48, 5.08, P < 0.001). CONCLUSION: Vessel-specific CAC assessment with DL can be performed accurately and rapidly on gated CT and AC CT and provides important prognostic information.

Visually Estimated RESOLVE Score Based on Coronary Computed Tomography to Predict Side Branch Occlusion in Percutaneous Bifurcation Intervention.

PURPOSE: The quantitative RESOLVE (Risk prEdiction of Side branch OccLusion in coronary bifurcation interVEntion) score derived from coronary computed tomography angiography (coronary CTA) was developed as a noninvasive and accurate prediction tool for side branch (SB) occlusion in coronary bifurcation intervention. We aimed to determine the ability of a visually estimated CTA-derived RESOLVE score (V-RESOLVE score) to predict SB occlusion in coronary bifurcation intervention. MATERIALS AND METHODS: The present study included 363 patients with 400 bifurcation lesions. CTA-derived V-RESOLVE score was derived and compared with the quantitative CTA-derived RESOLVE score. The scoring systems were divided into quartiles, and classified as the high-risk and non-high-risk groups. SB occlusion was defined as any decrease in thrombolysis in myocardial infarction flow grade after main vessel stenting. RESULTS: In total, 28 SB occlusions (7%) occurred. The concordance between visual and quantitative CTA analysis showed poor to excellent agreement (weighted κ range: 0.099 to 0.867). The area under the receiver operating curve for the prediction of SB occlusion was significantly higher for the CTA-derived V-RESOLVE score than for quantitative CTA-derived RESOLVE score (0.792 vs. 0.709, P=0.049). The total net reclassification index was 42.7% (P=0.006), and CTA-derived V-RESOLVE score showed similar capability to discriminate between high-risk group (18.6% vs. 13.8%, P=0.384) and non-high-risk group (3.8% vs. 4.9%, P=0.510) as compared with quantitative CTA-derived RESOLVE score. CONCLUSIONS: Visually estimated CTA-derived V-RESOLVE score is an accurate and easy-to-use prediction tool for the stratification of SB occlusion in coronary bifurcation intervention.

Noncalcified plaque burden quantified from coronary computed tomography angiography improves prediction of side branch occlusion after main vessel stenting in bifurcation lesions: results from the CT-PRECISION registry.

OBJECTIVES: To assess the incremental value of quantitative plaque features measured from computed tomography angiography (CTA) for predicting side branch (SB) occlusion in coronary bifurcation intervention. METHODS: We included 340 patients with 377 bifurcation lesions in the post hoc analysis of the CT-PRECISION registry. Each bifurcation was divided into three segments: the proximal main vessel (MV), the distal MV, and the SB. Segments with evidence of coronary plaque were analyzed using semi-automated software allowing for quantitative analysis of coronary plaque morphology and stenosis. Coronary plaque measurements included calcified and noncalcified plaque volumes, and corresponding burdens (respective plaque volumes × 100%/vessel volume), remodeling index, and stenosis. RESULTS: SB occlusion occurred in 28 of 377 bifurcation lesions (7.5%). The presence of visually identified plaque in the SB segment, but not in the proximal and distal MV segments, was the only qualitative parameter that predicted SB occlusion with an area under the curve (AUC) of 0.792. Among quantitative plaque parameters calculated for the SB segment, the addition of noncalcified plaque burden (AUC 0.840, p = 0.003) and low-density plaque burden (AUC 0.836, p = 0.012) yielded significant improvements in predicting SB occlusion. Using receiver operating characteristic curve analysis, optimal cut-offs for noncalcified plaque burden and low-density plaque burden were > 33.6% (86% sensitivity and 78% specificity) and > 0.9% (89% sensitivity and 73% specificity), respectively. CONCLUSIONS: CTA-derived noncalcified plaque burden, when added to the visually identified SB plaque, significantly improves the prediction of SB occlusion in coronary bifurcation intervention. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03709836 registered on October 17, 2018.

Comparison of Computed Tomography Angiography Versus Invasive Angiography to Assess Medina Classification in Coronary Bifurcations.

The Medina classification is used to determine the presence of significant stenosis (≥50%) within each of the 3 arterial segments of coronary bifurcation in invasive coronary angiography (ICA). The utility of coronary computed tomography angiography (coronary CTA) for assessment of Medina classification is unknown. We aimed to compare the agreement and reproducibility of Medina classification between ICA and coronary CTA, and evaluate its ability to predict side branch (SB) occlusion following percutaneous coronary intervention (PCI). In total 363 patients with 400 bifurcations were included, and 28 (7%) SB occlusions among 26 patients were noted. Total agreement between CTA and ICA for assessment of Medina class was poor (kappa = 0.189), and discordance between both modalities was noted in 253 (63.3%) lesions. Larger diameter ratio between main vessel and SB in CTA, and larger bifurcation angle in ICA were independently associated with discordant Medina assessment. Whereas the interobserver agreement on Medina classification in CTA was moderate (kappa = 0.557), only fair agreement (kappa = 0.346) was observed for ICA. Finally, Medina class with any proximal involvement of main vessel and SB (1.X.1) on CTA or ICA was the most predictive of SB occlusion following PCI with no significant differences between both modalities (area under the curve 0.686 vs 0.663, p = 0.693, respectively). In conclusion, Medina classification was significantly affected by the imaging modality, and coronary CTA improved reproducibility of Medina classification compared with ICA. Both CTA and ICA-derived Medina class with any involvement of the proximal main vessel and SB was predictive of SB occlusion following PCI.

Accuracy of RESOLVE score derived from coronary computed tomography versus visual angiography to predict side branch occlusion in percutaneous bifurcation intervention.

INTRODUCTION: Visually estimated angiographic V-RESOLVE score was developed as a simple and accurate prediction tool for side branch (SB) occlusion in patients undergoing coronary bifurcation intervention. Data on the use of coronary computed tomography angiography (coronary CTA) for guiding percutaneous coronary intervention in bifurcation lesions is scarce. OBJECTIVES: We aimed to validate the ability of quantitative CTA-derived RESOLVE score for predicting SB occlusion in coronary bifurcation intervention and to compare its predictive value with that of the angiography-based V-RESOLVE score. METHODS: We included 363 patients with 400 bifurcation lesions. Angiographic V-RESOLVE score and CTA-derived RESOLVE score were calculated utilizing the weights from the QCA-based RESOLVE score. The scoring systems were divided into quartiles, and classified as the non-high-risk group and the high-risk group. Accuracy was assessed using areas under the receiver-operator characteristic curve (AUC). SB occlusion was defined as any decrease in Thrombolysis in Myocardial Infarction flow grade (including the absence of flow) in the SB after main vessel stenting. RESULTS: In total, 28 SB occlusions (7%) occurred. CTA-derived RESOLVE and V-RESOLVE scores achieved comparable predictive accuracy (0.709 vs. 0.752, respectively, p = 0.531) for predicting SB occlusion, and the analysis of AUC for each constituent element of the scores did not show any significant difference between CTA and visual angiography. The total net reclassification index was -18.6% (p = 0.194), and there were no significant differences in the rates of SB occlusion in the non-high-risk group (4.9% vs. 3.8%, p = 0.510) and the high-risk group (13.8% vs. 18.6%, p = 0.384) between CTA-derived RESOLVE and V-RESOLVE scores. CONCLUSIONS: The quantitative CTA-derived RESOLVE score is an accurate and reliable alternative to the visually estimated angiographic V-RESOLVE score for prediction of SB occlusion in coronary bifurcation intervention. CLINICAL TRIAL REGISTRATION: URL: https://www.clinicaltrials.gov. Unique identifier: NCT03709836.

Prevalence and Anatomic Characteristics of Single Coronary Artery Diagnosed by Computed Tomography Angiography.

Single coronary artery (SCA) is a rare congenital anomaly. We assessed the prevalence and anatomic characteristics of SCA diagnosed with coronary computed tomography angiography and compared the dimensions of the proximal SCA trunk with a reference group of 199 subjects with normal coronary arteries. We screened 30,230 patients who underwent coronary computed tomography angiography from 2008 to 2018 to identify 17 with SCA (age 55 ± 19.0 years, 8 men [47%]). The prevalence of SCA was 0.056%. SCA originated from the right sinus of Valsalva in 11 patients (65%) and from the left sinus of Valsalva in 6 subjects. According to Lipton's classification, the 17 SCAs were L1 (n = 5, 29%), L2-A (n = 1, 6%), R2-A (n = 2, 12%), R2-B (n = 6, 35%), R2-P (n = 2, 12%), and R3 (n = 1, 6%). (Lipton's classification consists of 3 groups and the division is based on the site of origin of SCA ["R" - right, "L" - left sinus of Valsalva] and its anatomical course relating to the ascending aorta and pulmonary trunk ["A" - anterior to the pulmonary trunk, "B" - between the aorta and pulmonary trunk, "P" - posterior to the aorta].) As compared with the reference group, SCA patients had shorter proximal trunks (5.0 ± 3.6 mm vs 8.6 ± 4.8 mm, p = 0.0012). The lumen area (LA) and lumen diameter of the proximal trunk in patients with SCA were larger than the LA and lumen diameter of the left main coronary artery from the reference group (49.5 ± 18.0 mm2 vs 21.3 ± 6.5 mm2, p <0.0001, and 7.8 ± 1.6 mm vs 5.1 ± 0.75 mm, p <0.0001, respectively). Moreover, the LA of the proximal SCA trunk was larger than the sum of respective measurement performed in left main coronary artery and proximal right coronary artery segments in the control group (49.5 ± 18.0 mm2 vs 34.0 ± 7.9mm2, p = 0.0001). In conclusion, the incidence of SCA is very low; but this condition is associated with significant enlargement of the proximal vessel segment.