Quantification of Aortic Valve Fibrotic and Calcific Tissue from CTA: Prospective Comparison with Histology.

Background Quantifying the fibrotic and calcific composition of the aortic valve at CT angiography (CTA) can be useful for assessing disease severity and outcomes of patients with aortic stenosis (AS); however, it has not yet been validated against quantitative histologic findings. Purpose To compare quantification of aortic valve fibrotic and calcific tissue composition at CTA versus histologic examination. Materials and Methods This prospective study included patients who underwent CTA before either surgical aortic valve replacement for AS or orthotopic heart transplant (controls) at two centers between January 2022 and April 2023. At CTA, fibrotic and calcific tissue composition were quantified using automated Gaussian mixture modeling applied to the density of aortic valve tissue components, calculated as [(volume/total tissue volume) × 100]. For histologic evaluation, explanted valve cusps were stained with Movat pentachrome as well as hematoxylin and eosin. For each cusp, three 5-µm slices were obtained. Fibrotic and calcific tissue composition were quantified using a validated artificial intelligence tool and averaged across the aortic valve. Correlations were assessed using the Spearman rank correlation coefficient. Intermodality and interobserver variability were measured using the intraclass correlation coefficient (ICC) and Bland-Altman plots. Results Twenty-nine participants (mean age, 63 years ± 10 [SD]; 23 male) were evaluated: 19 with severe AS, five with moderate AS, and five controls. Fibrocalcific tissue composition strongly correlated with histologic findings (r = 0.92; P < .001). The agreement between CTA and histologic findings for fibrocalcific tissue quantification was excellent (ICC, 0.94; P = .001), with underestimation of fibrotic composition at CTA (bias, -4.9%; 95% limits of agreement [LoA]: -18.5%, 8.7%). Finally, there was excellent interobserver repeatability for fibrotic (ICC, 0.99) and calcific (ICC, 0.99) aortic valve tissue volume measurements, with no evidence of a difference in measurements between readers (bias, -0.04 cm3 [95% LoA: -0.27 cm3, 0.19 cm3] and 0.02 cm3 [95% LoA: -0.14 cm3, 0.19 cm3], respectively). Conclusion In a direct comparison, standardized quantitative aortic valve tissue characterization at CTA showed excellent concordance with histologic findings and demonstrated interobserver reproducibility. Clinical trial registration no. NCT06136689 Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Almeida in this issue.

Mitigating bias in deep learning for diagnosis of coronary artery disease from myocardial perfusion SPECT images.

PURPOSE: Artificial intelligence (AI) has high diagnostic accuracy for coronary artery disease (CAD) from myocardial perfusion imaging (MPI). However, when trained using high-risk populations (such as patients with correlating invasive testing), the disease probability can be overestimated due to selection bias. We evaluated different strategies for training AI models to improve the calibration (accurate estimate of disease probability), using external testing. METHODS: Deep learning was trained using 828 patients from 3 sites, with MPI and invasive angiography within 6 months. Perfusion was assessed using upright (U-TPD) and supine total perfusion deficit (S-TPD). AI training without data augmentation (model 1) was compared to training with augmentation (increased sampling) of patients without obstructive CAD (model 2), and patients without CAD and TPD < 2% (model 3). All models were tested in an external population of patients with invasive angiography within 6 months (n = 332) or low likelihood of CAD (n = 179). RESULTS: Model 3 achieved the best calibration (Brier score 0.104 vs 0.121, p < 0.01). Improvement in calibration was particularly evident in women (Brier score 0.084 vs 0.124, p < 0.01). In external testing (n = 511), the area under the receiver operating characteristic curve (AUC) was higher for model 3 (0.930), compared to U-TPD (AUC 0.897) and S-TPD (AUC 0.900, p < 0.01 for both). CONCLUSION: Training AI models with augmentation of low-risk patients can improve calibration of AI models developed to identify patients with CAD, allowing more accurate assignment of disease probability. This is particularly important in lower-risk populations and in women, where overestimation of disease probability could significantly influence down-stream patient management.

Diagnostic safety of a machine learning-based automatic patient selection algorithm for stress-only myocardial perfusion SPECT.

BACKGROUND: Stress-only myocardial perfusion imaging (MPI) markedly reduces radiation dose, scanning time, and cost. We developed an automated clinical algorithm to safely cancel unnecessary rest imaging with high sensitivity for obstructive coronary artery disease (CAD). METHODS AND RESULTS: Patients without known CAD undergoing both MPI and invasive coronary angiography from REFINE SPECT were studied. A machine learning score (MLS) for prediction of obstructive CAD was generated using stress-only MPI and pre-test clinical variables. An MLS threshold with a pre-defined sensitivity of 95% was applied to the automated patient selection algorithm. Obstructive CAD was present in 1309/2079 (63%) patients. MLS had higher area under the receiver operator characteristic curve (AUC) for prediction of CAD than reader diagnosis and TPD (0.84 vs 0.70 vs 0.78, P < .01). An MLS threshold of 0.29 had superior sensitivity than reader diagnosis and TPD for obstructive CAD (95% vs 87% vs 87%, P < .01) and high-risk CAD, defined as stenosis of the left main, proximal left anterior descending, or triple-vessel CAD (sensitivity 96% vs 89% vs 90%, P < .01). CONCLUSIONS: The MLS is highly sensitive for prediction of both obstructive and high-risk CAD from stress-only MPI and can be applied to a stress-first protocol for automatic cancellation of unnecessary rest imaging.

Metabolic syndrome, fatty liver, and artificial intelligence-based epicardial adipose tissue measures predict long-term risk of cardiac events: a prospective study.

BACKGROUND: We sought to evaluate the association of metabolic syndrome (MetS) and computed tomography (CT)-derived cardiometabolic biomarkers (non-alcoholic fatty liver disease [NAFLD] and epicardial adipose tissue [EAT] measures) with long-term risk of major adverse cardiovascular events (MACE) in asymptomatic individuals. METHODS: This was a post-hoc analysis of the prospective EISNER (Early-Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) study of participants who underwent baseline coronary artery calcium (CAC) scoring CT and 14-year follow-up for MACE (myocardial infarction, late revascularization, or cardiac death). EAT volume (cm3) and attenuation (Hounsfield units [HU]) were quantified from CT using fully automated deep learning software (< 30 s per case). NAFLD was defined as liver-to-spleen attenuation ratio < 1.0 and/or average liver attenuation < 40 HU. RESULTS: In the final population of 2068 participants (59% males, 56 ± 9 years), those with MetS (n = 280;13.5%) had a greater prevalence of NAFLD (26.0% vs. 9.9%), higher EAT volume (114.1 cm3 vs. 73.7 cm3), and lower EAT attenuation (-76.9 HU vs. -73.4 HU; all p < 0.001) compared to those without MetS. At 14 ± 3 years, MACE occurred in 223 (10.8%) participants. In multivariable Cox regression, MetS was associated with increased risk of MACE (HR 1.58 [95% CI 1.10-2.27], p = 0.01) independently of CAC score; however, not after adjustment for EAT measures (p = 0.27). In a separate Cox analysis, NAFLD predicted MACE (HR 1.78 [95% CI 1.21-2.61], p = 0.003) independently of MetS, CAC score, and EAT measures. Addition of EAT volume to current risk assessment tools resulted in significant net reclassification improvement for MACE (22% over ASCVD risk score; 17% over ASCVD risk score plus CAC score). CONCLUSIONS: MetS, NAFLD, and artificial intelligence-based EAT measures predict long-term MACE risk in asymptomatic individuals. Imaging biomarkers of cardiometabolic disease have the potential for integration into routine reporting of CAC scoring CT to enhance cardiovascular risk stratification. Trial registration NCT00927693.

Prediction of revascularization by coronary CT angiography using a machine learning ischemia risk score.

OBJECTIVES: The machine learning ischemia risk score (ML-IRS) is a machine learning-based algorithm designed to identify hemodynamically significant coronary disease using quantitative coronary computed tomography angiography (CCTA). The purpose of this study was to examine whether the ML-IRS can predict revascularization in patients referred for invasive coronary angiography (ICA) after CCTA. METHODS: This study was a post hoc analysis of a prospective dual-center registry of sequential patients undergoing CCTA followed by ICA within 3 months, referred from inpatient, outpatient, and emergency department settings (n = 352, age 63 ± 10 years, 68% male). The primary outcome was revascularization by either percutaneous coronary revascularization or coronary artery bypass grafting. Blinded readers performed semi-automated quantitative coronary plaque analysis. The ML-IRS was automatically computed. Relationships between clinical risk factors, coronary plaque features, and ML-IRS with revascularization were examined. RESULTS: The study cohort consisted of 352 subjects with 1056 analyzable vessels. The ML-IRS ranged between 0 and 81% with a median of 18.7% (6.4-34.8). Revascularization was performed in 26% of vessels. Vessels receiving revascularization had higher ML-IRS (33.6% (21.1-55.0) versus 13.0% (4.5-29.1), p < 0.0001), as well as higher contrast density difference, and total, non-calcified, calcified, and low-density plaque burden. ML-IRS, when added to a traditional risk model based on clinical data and stenosis to predict revascularization, resulted in increased area under the curve from 0.69 (95% CI: 0.65-0.72) to 0.78 (95% CI: 0.75-0.81) (p < 0.0001), with an overall continuous net reclassification improvement of 0.636 (95% CI: 0.503-0.769; p < 0.0001). CONCLUSIONS: ML-IRS from quantitative coronary CT angiography improved the prediction of future revascularization and can potentially identify patients likely to receive revascularization if referred to cardiac catheterization. KEY POINTS: • Machine learning ischemia risk from quantitative coronary CT angiography was significantly higher in patients who received revascularization versus those who did not receive revascularization. • The machine learning ischemia risk score was significantly higher in patients with invasive fractional flow ≤ 0.8 versus those with > 0.8. • The machine learning ischemia risk score improved the prediction of future revascularization significantly when added to a standard prediction model including stenosis.

Rationale and design of the REgistry of Fast Myocardial Perfusion Imaging with NExt generation SPECT (REFINE SPECT).

BACKGROUND: We aim to establish a multicenter registry collecting clinical, imaging, and follow-up data for patients who undergo myocardial perfusion imaging (MPI) with the latest generation SPECT scanners. METHODS: REFINE SPECT (REgistry of Fast Myocardial Perfusion Imaging with NExt generation SPECT) uses a collaborative design with multicenter contribution of clinical data and images into a comprehensive clinical-imaging database. All images are processed by quantitative software. Over 290 individual imaging variables are automatically extracted from each image dataset and merged with clinical variables. In the prognostic cohort, patient follow-up is performed for major adverse cardiac events. In the diagnostic cohort (patients with correlating invasive angiography), angiography and revascularization results within 6 months are obtained. RESULTS: To date, collected prognostic data include scans from 20,418 patients in 5 centers (57% male, 64.0 ± 12.1 years) who underwent exercise (48%) or pharmacologic stress (52%). Diagnostic data include 2079 patients in 9 centers (67% male, 64.7 ± 11.2 years) who underwent exercise (39%) or pharmacologic stress (61%). CONCLUSION: The REFINE SPECT registry will provide a resource for collaborative projects related to the latest generation SPECT-MPI. It will aid in the development of new artificial intelligence tools for automated diagnosis and prediction of prognostic outcomes.

Upper reference limits of transient ischemic dilation ratio for different protocols on new-generation cadmium zinc telluride cameras: A report from REFINE SPECT registry.

BACKGROUND: Upper reference limits for transient ischemic dilation (TID) have not been rigorously established for cadmium-zinc-telluride (CZT) camera systems. We aimed to derive TID limits for common myocardial perfusion imaging protocols utilizing a large, multicenter registry (REFINE SPECT). METHODS: One thousand six hundred and seventy-two patients with low likelihood of coronary artery disease with normal perfusion findings were identified. Images were processed with Quantitative Perfusion SPECT software (Cedars-Sinai Medical Center, Los Angeles, CA). Non-attenuation-corrected, camera-, radiotracer-, and stress protocol-specific TID limits in supine position were derived from 97.5th percentile and mean + 2 standard deviations (SD). Reference limits were compared for different solid-state cameras (D-SPECT vs. Discovery), radiotracers (technetium-99m-sestamibi vs. tetrofosmin), different types of stress (exercise vs. four different vasodilator-based protocols), and different vasodilator-based protocols. RESULTS: TID measurements did not follow Gaussian distribution in six out of eight subgroups. TID limits ranged from 1.18 to 1.52 (97.5th percentile) and 1.18 to 1.39 (mean + 2SD). No difference was noted between D-SPECT and Discovery cameras (P = 0.71) while differences between exercise and vasodilator-based protocols (adenosine, regadenoson, or regadenoson-walk) were noted (all P < 0.05). CONCLUSIONS: We used a multicenter registry to establish camera-, radiotracer-, and protocol-specific upper reference limits of TID for supine position on CZT camera systems. Reference limits did not differ between D-SPECT and Discovery camera.

Standardized volumetric plaque quantification and characterization from coronary CT angiography: a head-to-head comparison with invasive intravascular ultrasound.

OBJECTIVES: We sought to evaluate the accuracy of standardized total plaque volume (TPV) measurement and low-density non-calcified plaque (LDNCP) assessment from coronary CT angiography (CTA) in comparison with intravascular ultrasound (IVUS). METHODS: We analyzed 118 plaques without extensive calcifications from 77 consecutive patients who underwent CTA prior to IVUS. CTA TPV was measured with semi-automated software comparing both scan-specific (automatically derived from scan) and fixed attenuation thresholds. From CTA, %LDNCP was calculated voxels below multiple LDNCP thresholds (30, 45, 60, 75, and 90 Hounsfield units [HU]) within the plaque. On IVUS, the lipid-rich component was identified by echo attenuation, and its size was measured using attenuation score (summed score ∕ analysis length) based on attenuation arc (1 = < 90°; 2 = 90-180°; 3 = 180-270°; 4 = 270-360°) every 1 mm. RESULTS: TPV was highly correlated between CTA using scan-specific thresholds and IVUS (r = 0.943, p < 0.001), with no significant difference (2.6 mm3, p = 0.270). These relationships persisted for calcification patterns (maximal IVUS calcium arc of 0°, < 90°, or ≥ 90°). The fixed thresholds underestimated TPV (- 22.0 mm3, p < 0.001) and had an inferior correlation with IVUS (p < 0.001) compared with scan-specific thresholds. A 45-HU cutoff yielded the best diagnostic performance for identification of lipid-rich component, with an area under the curve of 0.878 vs. 0.840 for < 30 HU (p = 0.023), and corresponding %LDNCP resulted in the strongest correlation with the lipid-rich component size (r = 0.691, p < 0.001). CONCLUSIONS: Standardized noninvasive plaque quantification from CTA using scan-specific thresholds correlates highly with IVUS. Use of a < 45-HU threshold for LDNCP quantification improves lipid-rich plaque assessment from CTA. KEY POINTS: • Standardized scan-specific threshold-based plaque quantification from coronary CT angiography provides an accurate total plaque volume measurement compared with intravascular ultrasound. • Attenuation histogram-based low-density non-calcified plaque quantification can improve lipid-rich plaque assessment from coronary CT angiography.

Post-pericardiotomy Syndrome.

Post-pericardiotomy syndrome (PPS) occurs in a subgroup of patients who have undergone cardiothoracic surgery and is characterized by fever, pleuritic pain, pleural effusion, and pericardial effusion. It is associated with significant morbidity, and the leading complications include tamponade and constrictive pericarditis. Epidemiologic studies have found that PPS often occurs among younger patients; however, there is a lack of comprehensive risk stratification. It is therefore important to be able to identify patients who are at high risk for developing this disease. The diagnosis is made if patients present with 2 out of the following 5 criteria; fever, pericardial or pleuritic chest pain, pericardial or pleural friction rub, pericardial effusion, and pleural effusion with elevated C-reactive protein (CRP). Pericardial effusion associated with PPS is detected by echocardiography, and cardiac MRI is used for evaluation of pericardial thickening as well as inflammation associated with PPS. These imaging modalities have been invaluable for monitoring the efficacy of treatment in PPS. Aspirin, nonsteroidal anti-inflammatory agents (NSAID), and colchicine are the mainstay of the current treatment for PPS. Although steroids are used for refractory cases of PPS, they are associated with significant side effects when used for long-term treatment of this disease. It is important for future research to focus on identification of clinical, serologic, and genetic markers that may predispose patients to PPS. There is also a need for clinical trials to address the use of targeted immunomodulatory treatment for this disease.

Infectious Myocarditis on FDG-PET Imaging Mimicking Sarcoidosis.

Cardiac positron emission tomography with fluorine-18 fluorodeoxyglucose (FDG-PET) is often used for the diagnosis of cardiac involvement in sarcoidosis. Areas of segmental perfusion defects coupled with FDG uptake are considered to represent active inflammation. However, these findings may be associated with other inflammatory myocardial diseases. We describe a case of tuberculous myocarditis with imaging findings mimicking those found in cardiac sarcoidosis.

Atherosclerotic plaque uptake of a novel integrin tracer ¹8F-Flotegatide in a mouse model of atherosclerosis.

INTRODUCTION: Rupture of unstable atherosclerotic plaque that leads to stroke and myocardial infarction may be induced by macrophage infiltration and neovessel formation. A tracer that selectively binds to integrin αvß3 a protein expressed by macrophages and neovascular endothelium may identify rupture prone plaque. METHODS: (18)F-labeled "R-G-D" containing tripeptide (Flotegatide), a click chemistry derived radiotracer that binds to integrin αvß3 was injected in ApoE knockout mice fed a high fat diet. Uptake of Flotegatide by atherosclerotic plaque was visualized by micro-PET, autoradiography, and correlated to histologic markers of inflammation and angiogenesis. RESULTS: We found that Flotegatide preferentially binds to aortic plaque in an ApoE knockout mouse model of atherosclerosis. The tracer's uptake is strongly associated with presence of histologic markers for macrophage infiltration and integrin expression. There is a weaker but detectable association between Flotegatide uptake and presence of an immunohistochemical marker for neovascularization. DISCUSSION: We hypothesize that Flotegatide may be a useful tracer for visualization of inflamed plaque in clinical subjects with atherosclerosis and may have potential for detecting vulnerable plaque.

Chemotherapy Related Cardiotoxicity Evaluation-A Contemporary Review with a Focus on Cardiac Imaging.

The long-term survivorship of patients diagnosed with cancer has improved due to accelerated detection and rapidly evolving cancer treatment strategies. As such, the evaluation and management of cancer therapy related complications has become increasingly important, including cardiovascular complications. These have been captured under the umbrella term "cardiotoxicity" and include left ventricular dysfunction and heart failure, acute coronary syndromes, valvular abnormalities, pericardial disease, arrhythmia, myocarditis, and vascular complications. These complications add to the burden of cardiovascular disease (CVD) or are risk factors patients with cancer treatment are presenting with. Of note, both pre- and newly developing CVD is of prognostic significance, not only from a cardiovascular perspective but also overall, potentially impacting the level of cancer therapy that is possible. Currently, there are varying recommendations and practices regarding CVD risk assessment and mitigating strategies throughout the cancer continuum. This article provides an overview on this topic, in particular, the role of cardiac imaging in the care of the patient with cancer. Furthermore, it summarizes the current evidence on the spectrum, prevention, and management of chemotherapy-related adverse cardiac effects.

Variation of computed tomography-derived extracellular volume fraction and the impact of protocol parameters: A systematic review and meta-analysis.

BACKGROUND: Cardiac computed tomography quantification of extracellular volume fraction (CT-ECV) is an emerging biomarker of myocardial fibrosis which has demonstrated high reproducibility, diagnostic and prognostic utility. However, there has been wide variation in the CT-ECV protocol in the literature and useful disease cut-offs are yet to be established. The objectives of this meta-analysis were to describe mean CT-ECV estimates and to estimate the effect of CT-ECV protocol parameters on between-study variation. METHODS: We conducted a meta-analysis of studies assessing CT-ECV in healthy and diseased participants. We used meta-analytic methods to pool estimates of CT-ECV and performed meta-regression to identify the contribution of protocol parameters to CT-ECV heterogeneity. RESULTS: Thirteen studies had a total of 248 healthy participants who underwent CT-ECV assessment. Studies of healthy participants had high variation in CT-ECV protocol parameters. The pooled estimate of CT-ECV in healthy participants was 27.6% (95%CI 25.7%-29.4%) with significant heterogeneity (I2 â€‹= â€‹93%) compared to 50.2% (95%CI 46.2%-54.2%) in amyloidosis, 31.2% (28.5%-33.8%) in severe aortic stenosis and 36.9% (31.6%-42.3%) in non-ischaemic dilated cardiomyopathies. Meta-regression revealed that CT protocol parameters account for approximately 25% of the heterogeneity in CT-ECV estimates. CONCLUSION: CT-ECV estimates for healthy individuals vary widely in the literature and there is significant overlap with estimates in cardiac disease. One quarter of this heterogeneity is explained by differences in CT-ECV protocol parameters. Standardization of CT-ECV protocols is necessary for widespread implementation of CT-ECV assessment for diagnosis and prognosis.

Soft tissue displacement for detection of left ventricle apical dyskinesis with transthoracic echocardiography.

We tested the hypothesis that the use of outward displacement of the soft tissue between the apex and the chest wall as seen in TTE, is a sign of apical displacement and would allow for more accurate diagnosis of apical dyskinesis. This is a retrospective study of 123 patients who underwent TTE and cardiac magnetic resonance imaging (MRI) within a time frame of 6 months between 2008 and 2019. 110 subjects were deemed to have good quality studies and included in the final analysis. An observer blinded to the study objectives evaluated the echocardiograms and recorded the presence or absence of apical dyskinesis. Two independent observers evaluated the echocardiograms based on the presence or absence of outward displacement of the overlying tissue at the LV apex. Cardiac MRI was used to validate the presence of apical dyskinesis. The proportion of studies which were identified as having apical dyskinesis with conventional criteria defined as outward movement of the left ventricular apex during systole were compared to those deemed to have dyskinesis based on tissue displacement. By cardiac MRI, 90 patients had apical dyskinesis. Using conventional criteria on TTE interpretation, 21 were diagnosed with apical dyskinesis (23.3%). However, when soft tissue displacement was used as the diagnostic marker of dyskinesis, 78 patients (86.7%) were diagnosed with dyskinesis, p < 0.01. Detection of displacement of soft tissue overlying the LV apex facilitates better recognition of LV apical dyskinesis.

Cardiac Computed Tomography for Quantification of Myocardial Extracellular Volume Fraction: A Systematic Review and Meta-Analysis.

BACKGROUND: Extracellular volume (ECV) is a quantitative measure of extracellular compartment expansion, and an increase in ECV is a marker of myocardial fibrosis. Although cardiac magnetic resonance (CMR) is considered the standard imaging tool for ECV quantification, cardiac computed tomography (CT) has also been used for ECV assessment. OBJECTIVES: The aim of this meta-analysis was to evaluate the correlation and agreement in the quantification of myocardial ECV by CT and CMR. METHODS: PubMed and Web of Science were searched for relevant publications reporting on the use of CT for ECV quantification compared with CMR as the reference standard. The authors employed a meta-analysis using the restricted maximum-likelihood estimator with a random-effects method to estimate summary correlation and mean difference. A subgroup analysis was performed to compare the correlation and mean differences between single-energy CT (SECT) and dual-energy CT (DECT) techniques for the ECV quantification. RESULTS: Of 435 papers, 13 studies comprising 383 patients were identified. The mean age range was 57.3 to 82 years, and 65% of patients were male. Overall, there was an excellent correlation between CT-derived ECV and CMR-derived ECV (mean: 0.90 [95% CI: 0.86-0.95]). The pooled mean difference between CT and CMR was 0.96% (95% CI: 0.14%-1.78%). Seven studies reported correlation values using SECT, and 4 studies reported those using DECT. The pooled correlation from studies utilizing DECT for ECV quantification was significantly higher compared with those with SECT (mean: 0.94 [95% CI: 0.91-0.98] vs 0.87 [95% CI: 0.80-0.94], respectively; P = 0.01). There was no significant difference in pooled mean differences between SECT vs DECT (P = 0.85). CONCLUSIONS: CT-derived ECV showed an excellent correlation and mean difference of <1% with CMR-derived ECV. However, the overall quality of the included studies was low, and larger, prospective studies are needed to examine the accuracy and diagnostic and prognostic utility of CT-derived ECV.

Elevated Coronary Artery Calcium Quantified by a Validated Deep Learning Model From Lung Cancer Radiotherapy Planning Scans Predicts Mortality.

PURPOSE: Coronary artery calcium (CAC) quantified on computed tomography (CT) scans is a robust predictor of atherosclerotic coronary disease; however, the feasibility and relevance of quantitating CAC from lung cancer radiotherapy planning CT scans is unknown. We used a previously validated deep learning (DL) model to assess whether CAC is a predictor of all-cause mortality and major adverse cardiac events (MACEs). METHODS: Retrospective analysis of non-contrast-enhanced radiotherapy planning CT scans from 428 patients with locally advanced lung cancer is performed. The DL-CAC algorithm was previously trained on 1,636 cardiac-gated CT scans and tested on four clinical trial cohorts. Plaques ≥ 1 cubic millimeter were measured to generate an Agatston-like DL-CAC score and grouped as DL-CAC = 0 (very low risk) and DL-CAC ≥ 1 (elevated risk). Cox and Fine and Gray regressions were adjusted for lung cancer and cardiovascular factors. RESULTS: The median follow-up was 18.1 months. The majority (61.4%) had a DL-CAC ≥ 1. There was an increased risk of all-cause mortality with DL-CAC ≥ 1 versus DL-CAC = 0 (adjusted hazard ratio, 1.51; 95% CI, 1.01 to 2.26; P = .04), with 2-year estimates of 56.2% versus 45.4%, respectively. There was a trend toward increased risk of major adverse cardiac events with DL-CAC ≥ 1 versus DL-CAC = 0 (hazard ratio, 1.80; 95% CI, 0.87 to 3.74; P = .11), with 2-year estimates of 7.3% versus 1.2%, respectively. CONCLUSION: In this proof-of-concept study, CAC was effectively measured from routinely acquired radiotherapy planning CT scans using an automated model. Elevated CAC, as predicted by the DL model, was associated with an increased risk of mortality, suggesting a potential benefit for automated cardiac risk screening before cancer therapy begins.

Explainable Deep Learning Improves Physician Interpretation of Myocardial Perfusion Imaging.

Artificial intelligence may improve accuracy of myocardial perfusion imaging (MPI) but will likely be implemented as an aid to physician interpretation rather than an autonomous tool. Deep learning (DL) has high standalone diagnostic accuracy for obstructive coronary artery disease (CAD), but its influence on physician interpretation is unknown. We assessed whether access to explainable DL predictions improves physician interpretation of MPI. Methods: We selected a representative cohort of patients who underwent MPI with reference invasive coronary angiography. Obstructive CAD, defined as stenosis ≥50% in the left main artery or ≥70% in other coronary segments, was present in half of the patients. We used an explainable DL model (CAD-DL), which was previously developed in a separate population from different sites. Three physicians interpreted studies first with clinical history, stress, and quantitative perfusion, then with all the data plus the DL results. Diagnostic accuracy was assessed using area under the receiver-operating-characteristic curve (AUC). Results: In total, 240 patients with a median age of 65 y (interquartile range 58-73) were included. The diagnostic accuracy of physician interpretation with CAD-DL (AUC 0.779) was significantly higher than that of physician interpretation without CAD-DL (AUC 0.747, P = 0.003) and stress total perfusion deficit (AUC 0.718, P < 0.001). With matched specificity, CAD-DL had higher sensitivity when operating autonomously compared with readers without DL results (P < 0.001), but not compared with readers interpreting with DL results (P = 0.122). All readers had numerically higher accuracy with CAD-DL, with AUC improvement 0.02-0.05, and interpretation with DL resulted in overall net reclassification improvement of 17.2% (95% CI 9.2%-24.4%, P < 0.001). Conclusion: Explainable DL predictions lead to meaningful improvements in physician interpretation; however, the improvement varied across the readers, reflecting the acceptance of this new technology. This technique could be implemented as an aid to physician diagnosis, improving the diagnostic accuracy of MPI.

Differences in Prognostic Value of Myocardial Perfusion Single-Photon Emission Computed Tomography Using High-Efficiency Solid-State Detector Between Men and Women in a Large International Multicenter Study.

BACKGROUND: Semiquantitative assessment of stress myocardial perfusion defect has been shown to have greater prognostic value for prediction of major adverse cardiac events (MACE) in women compared with men in single-center studies with conventional single-photon emission computed tomography (SPECT) cameras. We evaluated sex-specific difference in the prognostic value of automated quantification of ischemic total perfusion defect (ITPD) and the interaction between sex and ITPD using high-efficiency SPECT cameras with solid-state detectors in an international multicenter imaging registry (REFINE SPECT [Registry of Fast Myocardial Perfusion Imaging With Next-Generation SPECT]). METHODS: Rest and exercise or pharmacological stress SPECT myocardial perfusion imaging were performed in 17 833 patients from 5 centers. MACE was defined as the first occurrence of death or myocardial infarction. Total perfusion defect (TPD) at rest, stress, and ejection fraction were quantified automatically by software. ITPD was given by stressTPD-restTPD. Cox proportional hazards model was used to evaluate the association between ITPD versus MACE-free survival and expressed as a hazard ratio. RESULTS: In 10614 men and 7219 women, with a median follow-up of 4.75 years (interquartile range, 3.7-6.1), there were 1709 MACE. In a multivariable Cox model, after adjusting for revascularization and other confounding variables, ITPD was associated with MACE (hazard ratio, 1.08 [95% CI, 1.05-1.1]; P<0.001). There was an interaction between ITPD and sex (P<0.001); predicted survival for ITPD<5% was worse among men compared to women, whereas survival among women was worse than men for ITPD≥5%, P<0.001. CONCLUSIONS: In the international, multicenter REFINE SPECT registry, moderate and severe ischemia as quantified by ITPD from high-efficiency SPECT is associated with a worse prognosis in women compared with men.

Clinical Deployment of Explainable Artificial Intelligence of SPECT for Diagnosis of Coronary Artery Disease.

BACKGROUND: Explainable artificial intelligence (AI) can be integrated within standard clinical software to facilitate the acceptance of the diagnostic findings during clinical interpretation. OBJECTIVES: This study sought to develop and evaluate a novel, general purpose, explainable deep learning model (coronary artery disease-deep learning [CAD-DL]) for the detection of obstructive CAD following single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI). METHODS: A total of 3,578 patients with suspected CAD undergoing SPECT MPI and invasive coronary angiography within a 6-month interval from 9 centers were studied. CAD-DL computes the probability of obstructive CAD from stress myocardial perfusion, wall motion, and wall thickening maps, as well as left ventricular volumes, age, and sex. Myocardial regions contributing to the CAD-DL prediction are highlighted to explain the findings to the physician. A clinical prototype was integrated using a standard clinical workstation. Diagnostic performance by CAD-DL was compared to automated quantitative total perfusion deficit (TPD) and reader diagnosis. RESULTS: In total, 2,247 patients (63%) had obstructive CAD. In 10-fold repeated testing, the area under the receiver-operating characteristic curve (AUC) (95% CI) was higher according to CAD-DL (AUC: 0.83 [95% CI: 0.82-0.85]) than stress TPD (AUC: 0.78 [95% CI: 0.77-0.80]) or reader diagnosis (AUC: 0.71 [95% CI: 0.69-0.72]; P < 0.0001 for both). In external testing, the AUC in 555 patients was higher according to CAD-DL (AUC: 0.80 [95% CI: 0.76-0.84]) than stress TPD (AUC: 0.73 [95% CI: 0.69-0.77]) or reader diagnosis (AUC: 0.65 [95% CI: 0.61-0.69]; P < 0.001 for all). The present model can be integrated within standard clinical software and generates results rapidly (<12 seconds on a standard clinical workstation) and therefore could readily be incorporated into a typical clinical workflow. CONCLUSIONS: The deep-learning model significantly surpasses the diagnostic accuracy of standard quantitative analysis and clinical visual reading for MPI. Explainable artificial intelligence can be integrated within standard clinical software to facilitate acceptance of artificial intelligence diagnosis of CAD following MPI.

Deep learning-enabled coronary CT angiography for plaque and stenosis quantification and cardiac risk prediction: an international multicentre study.

BACKGROUND: Atherosclerotic plaque quantification from coronary CT angiography (CCTA) enables accurate assessment of coronary artery disease burden and prognosis. We sought to develop and validate a deep learning system for CCTA-derived measures of plaque volume and stenosis severity. METHODS: This international, multicentre study included nine cohorts of patients undergoing CCTA at 11 sites, who were assigned into training and test sets. Data were retrospectively collected on patients with a wide range of clinical presentations of coronary artery disease who underwent CCTA between Nov 18, 2010, and Jan 25, 2019. A novel deep learning convolutional neural network was trained to segment coronary plaque in 921 patients (5045 lesions). The deep learning network was then applied to an independent test set, which included an external validation cohort of 175 patients (1081 lesions) and 50 patients (84 lesions) assessed by intravascular ultrasound within 1 month of CCTA. We evaluated the prognostic value of deep learning-based plaque measurements for fatal or non-fatal myocardial infarction (our primary outcome) in 1611 patients from the prospective SCOT-HEART trial, assessed as dichotomous variables using multivariable Cox regression analysis, with adjustment for the ASSIGN clinical risk score. FINDINGS: In the overall test set, there was excellent or good agreement, respectively, between deep learning and expert reader measurements of total plaque volume (intraclass correlation coefficient [ICC] 0·964) and percent diameter stenosis (ICC 0·879; both p<0·0001). When compared with intravascular ultrasound, there was excellent agreement for deep learning total plaque volume (ICC 0·949) and minimal luminal area (ICC 0·904). The mean per-patient deep learning plaque analysis time was 5·65 s (SD 1·87) versus 25·66 min (6·79) taken by experts. Over a median follow-up of 4·7 years (IQR 4·0-5·7), myocardial infarction occurred in 41 (2·5%) of 1611 patients from the SCOT-HEART trial. A deep learning-based total plaque volume of 238·5 mm3 or higher was associated with an increased risk of myocardial infarction (hazard ratio [HR] 5·36, 95% CI 1·70-16·86; p=0·0042) after adjustment for the presence of deep learning-based obstructive stenosis (HR 2·49, 1·07-5·50; p=0·0089) and the ASSIGN clinical risk score (HR 1·01, 0·99-1·04; p=0·35). INTERPRETATION: Our novel, externally validated deep learning system provides rapid measurements of plaque volume and stenosis severity from CCTA that agree closely with expert readers and intravascular ultrasound, and could have prognostic value for future myocardial infarction. FUNDING: National Heart, Lung, and Blood Institute and the Miriam & Sheldon G Adelson Medical Research Foundation.