Improved Robustness for Deep Learning-based Segmentation of Multi-Center Myocardial Perfusion MRI Datasets Using Data Adaptive Uncertainty-guided Space-time Analysis.

BACKGROUND: Fully automatic analysis of myocardial perfusion MRI datasets enables rapid and objective reporting of stress/rest studies in patients with suspected ischemic heart disease. Developing deep learning techniques that can analyze multi-center datasets despite limited training data and variations in software (pulse sequence) and hardware (scanner vendor) is an ongoing challenge. METHODS: Datasets from 3 medical centers acquired at 3T (n = 150 subjects; 21,150 first-pass images) were included: an internal dataset (inD; n = 95) and two external datasets (exDs; n = 55) used for evaluating the robustness of the trained deep neural network (DNN) models against differences in pulse sequence (exD-1) and scanner vendor (exD-2). A subset of inD (n = 85) was used for training/validation of a pool of DNNs for segmentation, all using the same spatiotemporal U-Net architecture and hyperparameters but with different parameter initializations. We employed a space-time sliding-patch analysis approach that automatically yields a pixel-wise "uncertainty map" as a byproduct of the segmentation process. In our approach, dubbed Data Adaptive Uncertainty-Guided Space-time (DAUGS) analysis, a given test case is segmented by all members of the DNN pool and the resulting uncertainty maps are leveraged to automatically select the "best" one among the pool of solutions. For comparison, we also trained a DNN using the established approach with the same settings (hyperparameters, data augmentation, etc.). RESULTS: The proposed DAUGS analysis approach performed similarly to the established approach on the internal dataset (Dice score for the testing subset of inD: 0.896 ± 0.050 vs. 0.890 ± 0.049; p = n.s.) whereas it significantly outperformed on the external datasets (Dice for exD-1: 0.885 ± 0.040 vs. 0.849 ± 0.065, p < 0.005; Dice for exD-2: 0.811 ± 0.070 vs. 0.728 ± 0.149, p < 0.005). Moreover, the number of image series with "failed" segmentation (defined as having myocardial contours that include bloodpool or are noncontiguous in ≥1 segment) was significantly lower for the proposed vs. the established approach (4.3% vs. 17.1%, p < 0.0005). CONCLUSIONS: The proposed DAUGS analysis approach has the potential to improve the robustness of deep learning methods for segmentation of multi-center stress perfusion datasets with variations in the choice of pulse sequence, site location or scanner vendor.

Improved Robustness for Deep Learning-based Segmentation of Multi-Center Myocardial Perfusion MRI Datasets Using Data Adaptive Uncertainty-guided Space-time Analysis.

Background: Fully automatic analysis of myocardial perfusion MRI datasets enables rapid and objective reporting of stress/rest studies in patients with suspected ischemic heart disease. Developing deep learning techniques that can analyze multi-center datasets despite limited training data and variations in software (pulse sequence) and hardware (scanner vendor) is an ongoing challenge. Methods: Datasets from 3 medical centers acquired at 3T (n = 150 subjects; 21,150 first-pass images) were included: an internal dataset (inD; n = 95) and two external datasets (exDs; n = 55) used for evaluating the robustness of the trained deep neural network (DNN) models against differences in pulse sequence (exD-1) and scanner vendor (exD-2). A subset of inD (n = 85) was used for training/validation of a pool of DNNs for segmentation, all using the same spatiotemporal U-Net architecture and hyperparameters but with different parameter initializations. We employed a space-time sliding-patch analysis approach that automatically yields a pixel-wise "uncertainty map" as a byproduct of the segmentation process. In our approach, dubbed Data Adaptive Uncertainty-Guided Space-time (DAUGS) analysis, a given test case is segmented by all members of the DNN pool and the resulting uncertainty maps are leveraged to automatically select the "best" one among the pool of solutions. For comparison, we also trained a DNN using the established approach with the same settings (hyperparameters, data augmentation, etc.). Results: The proposed DAUGS analysis approach performed similarly to the established approach on the internal dataset (Dice score for the testing subset of inD: 0.896 ± 0.050 vs. 0.890 ± 0.049; p = n.s.) whereas it significantly outperformed on the external datasets (Dice for exD-1: 0.885 ± 0.040 vs. 0.849 ± 0.065, p < 0.005; Dice for exD-2: 0.811 ± 0.070 vs. 0.728 ± 0.149, p < 0.005). Moreover, the number of image series with "failed" segmentation (defined as having myocardial contours that include bloodpool or are noncontiguous in ≥1 segment) was significantly lower for the proposed vs. the established approach (4.3% vs. 17.1%, p < 0.0005). Conclusions: The proposed DAUGS analysis approach has the potential to improve the robustness of deep learning methods for segmentation of multi-center stress perfusion datasets with variations in the choice of pulse sequence, site location or scanner vendor.

Association of Adverse Clinical Outcomes With Peri-Infarct Ischemia Detected by Stress Cardiac Magnetic Imaging.

BACKGROUND: Early invasive revascularization guided by moderate to severe ischemia did not improve outcomes over medical therapy alone, underlying the need to identify high-risk patients for a more effective invasive referral. CMR could determine the myocardial extent and matching locations of ischemia and infarction. OBJECTIVES: This study sought to investigate if CMR peri-infarct ischemia is associated with adverse events incremental to known risk markers. METHODS: Consecutive patients were included in an expanded cohort of the multicenter SPINS (Stress CMR Perfusion Imaging in the United States) study. Peri-infarct ischemia was defined by the presence of any ischemic segment neighboring an infarcted segment by late gadolinium enhancement imaging. Primary outcome events included acute myocardial infarction and cardiovascular death, whereas secondary events included any primary events, hospitalization for unstable angina, heart failure hospitalization, and late coronary artery bypass surgery. RESULTS: Among 3,915 patients (age: 61.0 ± 12.9 years; 54.7% male), ischemia, infarct, and peri-infarct ischemia were present in 752 (19.2%), 1,123 (28.8%), and 382 (9.8%) patients, respectively. At 5.3 years (Q1-Q3: 3.9-7.2 years) of median follow-up, primary and secondary events occurred in 406 (10.4%) and 745 (19.0%) patients, respectively. Peri-infarct ischemia was the strongest multivariable predictor for primary and secondary events (HRadjusted: 1.72 [95% CI: 1.23-2.41] and 1.71 [95% CI: 1.32-2.20], respectively; both P < 0.001), adjusted for clinical risk factors, left ventricular function, ischemia extent, and infarct size. The presence of peri-infarct ischemia portended to a >6-fold increased annualized primary event rate compared to those with no infarct and ischemia (6.5% vs 0.9%). CONCLUSIONS: Peri-infarct ischemia is a novel and robust prognostic marker of adverse cardiovascular events.

The role of artificial intelligence in cardiovascular magnetic resonance imaging.

Cardiovascular magnetic resonance (CMR) imaging is the gold standard test for myocardial tissue characterization and chamber volumetric and functional evaluation. However, manual CMR analysis can be time-consuming and is subject to intra- and inter-observer variability. Artificial intelligence (AI) is a field that permits automated task performance through the identification of high-level and complex data relationships. In this review, we review the rapidly growing role of AI in CMR, including image acquisition, sequence prescription, artifact detection, reconstruction, segmentation, and data reporting and analysis including quantification of volumes, function, myocardial infarction (MI) and scar detection, and prediction of outcomes. We conclude with a discussion of the emerging challenges to widespread adoption and solutions that will allow for successful, broader uptake of this powerful technology.

Cardiovascular Imaging, Climate Change, and Environmental Sustainability.

Environmental exposures including poor air quality and extreme temperatures are exacerbated by climate change and are associated with adverse cardiovascular outcomes. Concomitantly, the delivery of health care generates substantial atmospheric greenhouse gas (GHG) emissions contributing to the climate crisis. Therefore, cardiac imaging teams must be aware not only of the adverse cardiovascular health effects of climate change, but also the downstream environmental ramifications of cardiovascular imaging. The purpose of this review is to highlight the impact of climate change on cardiovascular health, discuss the environmental impact of cardiovascular imaging, and describe opportunities to improve environmental sustainability of cardiac MRI, cardiac CT, echocardiography, cardiac nuclear imaging, and invasive cardiovascular imaging. Overarching strategies to improve environmental sustainability in cardiovascular imaging include prioritizing imaging tests with lower GHG emissions when more than one test is appropriate, reducing low-value imaging, and turning equipment off when not in use. Modality-specific opportunities include focused MRI protocols and low-field-strength applications, iodine contrast media recycling programs in cardiac CT, judicious use of US-enhancing agents in echocardiography, improved radiopharmaceutical procurement and waste management in nuclear cardiology, and use of reusable supplies in interventional suites. Finally, future directions and research are highlighted, including life cycle assessments over the lifespan of cardiac imaging equipment and the impact of artificial intelligence tools. Keywords: Heart, Safety, Sustainability, Cardiovascular Imaging Supplemental material is available for this article. © RSNA, 2024.

The Renin-Angiotensin-Aldosterone System Regulates Sarcoidosis Granulomatous Inflammation.

Rationale: Sarcoidosis is a granulomatous disorder of unclear cause notable for abnormal elevation of blood and tissue ACE1 (angiotensin converting enzyme 1) levels and activity. ACE1 regulates the renin-angiotensin-aldosterone system (RAAS), the terminal product of which is aldosterone, which selectively engages mineralocorticoid receptors to promote inflammation. Objectives: We sought to determine whether the RAAS promotes sarcoidosis granuloma formation and related inflammatory responses. Methods: Using an established ex vivo model, we first determined whether aldosterone was produced by sarcoidosis granulomas and verified the presence of CYP11B2, the enzyme required for its production. We then evaluated the effects of selective inhibitors of ACE1 (captopril), angiotensin type 1 receptor (losartan), and mineralocorticoid receptors (spironolactone, eplerenone) on granuloma formation, reflected by computer image analysis-generated granuloma area, and selected cytokines incriminated in sarcoidosis pathogenesis. Measurements and Main Results: Aldosterone was spontaneously produced by sarcoidosis peripheral blood mononuclear cells, and both intra- and extracellular levels steadily increased during granuloma formation. In parallel, peripheral blood mononuclear cells were shown to express more CYP11B2 during granuloma formation. Significant inhibition of sarcoidosis granulomas and related cytokines (TNFα, IL-1ß, IFNγ, IL-10) was observed in response to pretreatments with captopril, losartan, spironolactone, or eplerenone, comparable to that of prednisone. Conclusions: The RAAS is intact in sarcoidosis granulomas and contributes significantly to early granuloma formation and to related inflammatory mediator responses, with important implications for clinical management.

Microvascular Obstruction and Intramyocardial Hemorrhage in Reperfused Myocardial Infarctions: Pathophysiology and Clinical Insights From Imaging.

Microvascular injury immediately following reperfusion therapy in acute myocardial infarction (MI) has emerged as a driving force behind major adverse cardiovascular events in the postinfarction period. Although postmortem investigations and animal models have aided in developing early understanding of microvascular injury following reperfusion, imaging, particularly serial noninvasive imaging, has played a central role in cultivating critical knowledge of progressive damage to the myocardium from the onset of microvascular injury to months and years after in acute MI patients. This review summarizes the pathophysiological features of microvascular injury and downstream consequences, and the contributions noninvasive imaging has imparted in the development of this understanding. It also highlights the interventional trials that aim to mitigate the adverse consequences of microvascular injury based on imaging, identifies potential future directions of investigations to enable improved detection of disease, and demonstrates how imaging stands to play a major role in the development of novel therapies for improved management of acute MI patients.

Stress Perfusion Cardiac Magnetic Resonance vs SPECT Imaging for Detection of Coronary Artery Disease.

BACKGROUND: GadaCAD2 was 1 of 2 international, multicenter, prospective, Phase 3 clinical trials that led to U.S. Food and Drug Administration approval of gadobutrol to assess myocardial perfusion and late gadolinium enhancement (LGE) in adults with known or suspected coronary artery disease (CAD). OBJECTIVES: A prespecified secondary objective was to determine if stress perfusion cardiovascular magnetic resonance (CMR) was noninferior to single-photon emission computed tomography (SPECT) for detecting significant CAD and for excluding significant CAD. METHODS: Participants with known or suspected CAD underwent a research rest and stress perfusion CMR that was compared with a gated SPECT performed using standard clinical protocols. For CMR, adenosine or regadenoson served as vasodilators. The total dose of gadobutrol was 0.1 mmol/kg body weight. The standard of reference was a 70% stenosis defined by quantitative coronary angiography (QCA). A negative coronary computed tomography angiography could exclude CAD. Analysis was per patient. CMR, SPECT, and QCA were evaluated by independent central core lab readers blinded to clinical information. RESULTS: Participants were predominantly male (61.4% male; mean age 58.9 ± 10.2 years) and were recruited from the United States (75.0%), Australia (14.7%), Singapore (5.7%), and Canada (4.6%). The prevalence of significant CAD was 24.5% (n = 72 of 294). Stress perfusion CMR was statistically superior to gated SPECT for specificity (P = 0.002), area under the receiver operating characteristic curve (P < 0.001), accuracy (P = 0.003), positive predictive value (P < 0.001), and negative predictive value (P = 0.041). The sensitivity of CMR for a 70% QCA stenosis was noninferior and nonsuperior to gated SPECT. CONCLUSIONS: Vasodilator stress perfusion CMR, as performed with gadobutrol 0.1 mmol/kg body weight, had superior diagnostic accuracy for diagnosis and exclusion of significant CAD vs gated SPECT.

Temporal Uncertainty Localization to Enable Human-in-the-loop Analysis of Dynamic Contrast-enhanced Cardiac MRI Datasets.

Dynamic contrast-enhanced (DCE) cardiac magnetic resonance imaging (CMRI) is a widely used modality for diagnosing myocardial blood flow (perfusion) abnormalities. During a typical free-breathing DCE-CMRI scan, close to 300 time-resolved images of myocardial perfusion are acquired at various contrast "wash in/out" phases. Manual segmentation of myocardial contours in each time-frame of a DCE image series can be tedious and time-consuming, particularly when non-rigid motion correction has failed or is unavailable. While deep neural networks (DNNs) have shown promise for analyzing DCE-CMRI datasets, a "dynamic quality control" (dQC) technique for reliably detecting failed segmentations is lacking. Here we propose a new space-time uncertainty metric as a dQC tool for DNN-based segmentation of free-breathing DCE-CMRI datasets by validating the proposed metric on an external dataset and establishing a human-in-the-loop framework to improve the segmentation results. In the proposed approach, we referred the top 10% most uncertain segmentations as detected by our dQC tool to the human expert for refinement. This approach resulted in a significant increase in the Dice score (p<0.001) and a notable decrease in the number of images with failed segmentation (16.2% to 11.3%) whereas the alternative approach of randomly selecting the same number of segmentations for human referral did not achieve any significant improvement. Our results suggest that the proposed dQC framework has the potential to accurately identify poor-quality segmentations and may enable efficient DNN-based analysis of DCE-CMRI in a human-in-the-loop pipeline for clinical interpretation and reporting of dynamic CMRI datasets.

Worldwide variation in cardiovascular magnetic resonance practice models.

INTRODUCTION: The use of cardiovascular magnetic resonance (CMR) for diagnosis and management of a broad range of cardiac and vascular conditions has quickly expanded worldwide. It is essential to understand how CMR is utilized in different regions around the world and the potential practice differences between high-volume and low-volume centers. METHODS: CMR practitioners and developers from around the world were electronically surveyed by the Society for Cardiovascular Magnetic Resonance (SCMR) twice, requesting data from 2017. Both surveys were carefully merged, and the data were curated professionally by a data expert using cross-references in key questions and the specific media access control IP address. According to the United Nations classification, responses were analyzed by region and country and interpreted in the context of practice volumes and demography. RESULTS: From 70 countries and regions, 1092 individual responses were included. CMR was performed more often in academic (695/1014, 69%) and hospital settings (522/606, 86%), with adult cardiologists being the primary referring providers (680/818, 83%). Evaluation of cardiomyopathy was the top indication in high-volume and low-volume centers (p = 0.06). High-volume centers were significantly more likely to list evaluation of ischemic heart disease (e.g., stress CMR) as a primary indicator compared to low-volume centers (p < 0.001), while viability assessment was more commonly listed as a primary referral reason in low-volume centers (p = 0.001). Both developed and developing countries noted cost and competing technologies as top barriers to CMR growth. Access to scanners was listed as the most common barrier in developed countries (30% of responders), while lack of training (22% of responders) was the most common barrier in developing countries. CONCLUSION: This is the most extensive global assessment of CMR practice to date and provides insights from different regions worldwide. We identified CMR as heavily hospital-based, with referral volumes driven primarily by adult cardiology. Indications for CMR utilization varied by center volume. Efforts to improve the adoption and utilization of CMR should include growth beyond the traditional academic, hospital-based location and an emphasis on cardiomyopathy and viability assessment in community centers.

Cardiac Magnetic Resonance Imaging Versus Invasive-Based Strategies in Patients With Chest Pain and Detectable to Mildly Elevated Serum Troponin: A Randomized Clinical Trial.

BACKGROUND: The optimal diagnostic strategy for patients with chest pain and detectable to mildly elevated serum troponin is not known. The objective was to compare clinical outcomes among an early decision for a noninvasive versus an invasive-based care pathway. METHODS: The CMR-IMPACT trial (Cardiac Magnetic Resonance Imaging Strategy for the Management of Patients with Acute Chest Pain and Detectable to Elevated Troponin) was conducted at 4 United States tertiary care hospitals from September 2013 to July 2018. A convenience sample of 312 participants with acute chest pain symptoms and a contemporary troponin between detectable and 1.0 ng/mL were randomized early in their care to 1 of 2 care pathways: invasive-based (n=156) or cardiac magnetic resonance (CMR)-based (n=156) with modification allowed as the patient condition evolved. The primary outcome was a composite including death, myocardial infarction, and cardiac-related hospital readmission or emergency visits. RESULTS: Participants (N=312, mean age, 60.6 years, SD 11.3; 125 women [59.9%]), were followed over a median of 2.6 years (95% CI, 2.4-2.9). Early assigned testing was initiated in 102 out of 156 (65.3%) CMR-based and 110 out of 156 (70.5%) invasive-based participants. The primary outcome (CMR-based versus invasive-based) occurred in 59% versus 52% (hazard ratio, 1.17 [95% CI, 0.86-1.57]), acute coronary syndrome after discharge 23% versus 22% (hazard ratio, 1.07 [95% CI, 0.67-1.71]), and invasive angiography at any time 52% versus 74% (hazard ratio, 0.66 [95% CI, 0.49-0.87]). Among patients completing CMR imaging, 55 out of 95 (58%) were safely identified for discharge based on a negative CMR and did not have angiography or revascularization within 90 days. Therapeutic yield of angiography was higher in the CMR-based arm (52 interventions in 81 angiographies [64.2%] versus 46 interventions in 115 angiographies [40.0%] in the invasive-based arm [P=0.001]). CONCLUSIONS: Initial management with CMR or invasive-based care pathways resulted in no detectable difference in clinical and safety event rates. The CMR-based pathway facilitated safe discharge, enriched the therapeutic yield of angiography, and reduced invasive angiography utilization over long-term follow-up. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT01931852.

The role of coronary computed tomography angiography in low- and intermediate-risk patients with chest pain: a paradigm shift in modern cardiology.

Chest pain is a common complaint among patients presenting to primary care physicians. The management of chest pain secondary to coronary artery disease is rapidly changing as new evidence increase our knowledge of this complex clinical problem. The 2021 multisociety guidelines developed by the American College of Cardiology and the American Heart Association along with other organizations and imaging societies represent the first international guidelines for the evaluation and diagnosis of patients with acute or stable chest pain. This review will discuss in details the evaluation of low- and intermediate risk subjects presenting with acute and stable chest pain both in the emergency and office settings, providing a practical approach, supported by contemporary evidence, for the management of this important clinical problem leveraging on the central role played by coronary computed tomography angiography as documented by current clinical guidelines and available scientific literature.

Cardiovascular Magnetic Resonance Imaging in Patients With Ibrutinib-Associated Cardiotoxicity.

Importance: Ibrutinib has been associated with serious cardiotoxic arrhythmias. In preclinical models, these events are paralleled or proceeded by diffuse myocardial injury (inflammation and fibrosis). Yet whether this is seen in patients or has implications for future cardiotoxic risk is unknown. Objective: To assess the incidence and outcomes of myocardial injury among patients with ibrutinib-related cardiotoxicity. Design, Setting, and Participants: This cohort study included consecutive patients treated with ibrutinib from 2012 to 2019, phenotyped using cardiovascular magnetic resonance (CMR) from a large US Comprehensive Cancer Center registry. Exposures: Ibrutinib treatment for cancer control. Main Outcomes and Measures: The primary outcome was the presence of late gadolinium enhancement (LGE) fibrosis. The secondary outcome was the occurrence of major adverse cardiac events (MACE), defined as atrial fibrillation, heart failure, symptomatic ventricular arrhythmias, and sudden death of probable or definite ibrutinib association after CMR. We also assessed parametric-mapping subclinical fibrosis (native-T1, extracellular volume fraction) and inflammation/edema (max-T2) measures. Cardiovascular magnetic resonance measures were compared with those obtained in similar consecutive patients with cancer without ibrutinib treatment (pretreatment controls). Observed measures were also compared with similar-aged broad population rates (general-population controls) and a broader pool of cardiovascular disease (CVD) risk-matched cancer controls. Multivariable regression was used to assess the association between CMR measures and MACE. Results: Overall, 49 patients treated with ibrutinib were identified, including 33 imaged after treatment initiation (mean [SD] age, 65 [10] years, 9 [27%] with hypertension, and 23 [69.7%] with index-arrhythmias); median duration of ibrutinib-use was 14 months. The mean (SD) pretreatment native T1 was 977.0 (73.0) ms, max-T2 56.5 (4.0) ms, and 4 (13.3%) had LGE. Posttreatment initiation, mean (SD) native T1 was 1033.7 (48.2) ms, max-T2 61.5 (4.8) ms, and 17 (54.8%) had LGE (P < .001, P = .01, and P < .001, respectively, pre- vs post-ibrutinib treatment). Native T12SDs was elevated in 9 (28.6%), and max-T22SDs in 21 (63.0%), respectively. Cardiovascular magnetic resonance measures were highest in those with suspected toxic effects (P = .01 and P = .01, respectively). There was no association between traditional CVD-risk or cancer-treatment status and abnormal CMR measures. Among those without traditional CVD, 16 (58.6%) had LGE vs 38 (13.3%) in matched-controls (relative-risk, 4.8; P < .001). Over a median follow-up of 19 months, 13 (39.4%) experienced MACE. In multivariable models inclusive of traditional CVD risk factors, LGE (hazard ratio [HR], 4.9; P = .04), and native-T12SDs (HR, 3.3; P = .05) associated with higher risks of MACE. Conclusions and Relevance: In this cohort study, myocardial injury was common in ibrutinib users, and its presence was associated with higher cardiotoxic risk.

Sex-Specific Stress Perfusion Cardiac Magnetic Resonance Imaging in Suspected Ischemic Heart Disease: Insights From SPINS Retrospective Registry.

BACKGROUND: Cardiovascular disease (CVD) remains the leading cause of mortality in women, but current noninvasive cardiac imaging techniques have sex-specific limitations. OBJECTIVES: In this study, the authors sought to investigate the effect of sex on the prognostic utility and downstream invasive revascularization and costs of stress perfusion cardiac magnetic resonance (CMR) for suspected CVD. METHODS: Sex-specific prognostic performance was evaluated in a 2,349-patient multicenter SPINS (Stress CMR Perfusion Imaging in the United States [SPINS] Study) Registry. The primary outcome measure was a composite of cardiovascular death and nonfatal myocardial infarction; secondary outcomes were hospitalization for unstable angina or heart failure, and late unplanned coronary artery bypass grafting. RESULTS: SPINS included 1,104 women (47% of cohort); women had higher prevalence of chest pain (62% vs 50%; P < 0.0001) but lower use of medical therapies. At the 5.4-year median follow-up, women with normal stress CMR had a low annualized rate of primary composite outcome similar to men (0.54%/y vs 0.75%/y, respectively; P = NS). In contrast, women with abnormal CMR were at higher risk for both primary (3.74%/y vs 0.54%/y; P < 0.0001) and secondary (9.8%/y vs 1.6%/y; P < 0.0001) outcomes compared with women with normal CMR. Abnormal stress CMR was an independent predictor for the primary (HR: 2.64 [95% CI: 1.20-5.90]; P = 0.02) and secondary (HR: 2.09 [95% CI: 1.43-3.08]; P < 0.0001) outcome measures. There was no effect modification for sex. Women had lower rates of invasive coronary angiography (3.6% vs 7.3%; P = 0.0001) and downstream costs ($114 vs $171; P = 0.001) at 90 days following CMR. There was no effect of sex on diagnostic image quality. CONCLUSIONS: Stress CMR demonstrated excellent prognostic performance with lower rates of invasive coronary angiography referral in women. Stress CMR should be considered as a first-line noninvasive imaging tool for the evaluation of women. (Stress CMR Perfusion Imaging in the United States [SPINS] Study [SPINS]; NCT03192891).

Integrating Coronary Atherosclerosis Burden and Progression with Coronary Artery Disease Risk Factors to Guide Therapeutic Decision Making.

IMPORTANCE: Although atherosclerosis represents the primary driver of coronary artery disease, evaluation and treatment approaches have historically relied upon indirect markers of atherosclerosis that include surrogates (cholesterol), signs (angina), and sequelae (ischemia) of atherosclerosis. Direct quantification and characterization of atherosclerosis may encourage a precision heart care paradigm that improves diagnosis, risk stratification, therapeutic decision-making, and longitudinal disease tracking in a personalized fashion. OBSERVATIONS: The American College of Cardiology Innovations in Prevention Working Group introduce the Atherosclerosis Treatment Algorithms that personalize medical interventions based upon atherosclerosis findings from coronary computed tomography angiography (CTA) and cardiovascular risk factors. Through integration of coronary CTA-based atherosclerosis evaluation, clinical practice guidelines, and contemporary randomized controlled trial evidence, the Atherosclerosis Treatment Algorithms leverage patient-specific atherosclerosis burden and progression as primary targets for therapeutic intervention. After defining stages of atherosclerosis severity by coronary CTA, Atherosclerosis Treatment Algorithms are described for worsening stages of atherosclerosis for patients with lipid disorders, diabetes, hypertension, obesity, and tobacco use. The authors anticipate a rapid pace of research in the field, and conclude by providing perspectives on future needs that may improve efforts to optimize precision prevention of coronary artery disease. Importantly, the Atherosclerosis Treatment Algorithms are not endorsed by the American College of Cardiology, and should not be interpreted as a statement of American College of Cardiology policy. CONCLUSIONS AND RELEVANCE: We describe a precision heart care approach that emphasizes atherosclerosis as the primary disease target for evaluation and treatment. To our knowledge, this is the first proposal to use coronary atherosclerosis burden and progression to personalize therapy selection and therapy changes, respectively. DISCLOSURE: The American College of Cardiology Foundation has made an investment in Cleerly, Inc., makers of a software solution that utilizes coronary CT angiography findings to evaluate coronary artery disease.

Temporal Uncertainty Localization to Enable Human-in-the-Loop Analysis of Dynamic Contrast-Enhanced Cardiac MRI Datasets.

Dynamic contrast-enhanced (DCE) cardiac magnetic resonance imaging (CMRI) is a widely used modality for diagnosing myocardial blood flow (perfusion) abnormalities. During a typical free-breathing DCE-CMRI scan, close to 300 time-resolved images of myocardial perfusion are acquired at various contrast "wash in/out" phases. Manual segmentation of myocardial contours in each time-frame of a DCE image series can be tedious and time-consuming, particularly when non-rigid motion correction has failed or is unavailable. While deep neural networks (DNNs) have shown promise for analyzing DCE-CMRI datasets, a "dynamic quality control" (dQC) technique for reliably detecting failed segmentations is lacking. Here we propose a new space-time uncertainty metric as a dQC tool for DNN-based segmentation of free-breathing DCE-CMRI datasets by validating the proposed metric on an external dataset and establishing a human-in-the-loop framework to improve the segmentation results. In the proposed approach, we referred the top 10% most uncertain segmentations as detected by our dQC tool to the human expert for refinement. This approach resulted in a significant increase in the Dice score (p < 0.001) and a notable decrease in the number of images with failed segmentation (16.2% to 11.3%) whereas the alternative approach of randomly selecting the same number of segmentations for human referral did not achieve any significant improvement. Our results suggest that the proposed dQC framework has the potential to accurately identify poor-quality segmentations and may enable efficient DNN-based analysis of DCE-CMRI in a human-in-the-loop pipeline for clinical interpretation and reporting of dynamic CMRI datasets.

Society for cardiovascular magnetic resonance recommendations for training and competency of CMR technologists.

The Society for Cardiovascular Magnetic Resonance (SCMR) recommendations for training and competency of cardiovascular magnetic resonance (CMR) technologists document will define the knowledge, experiences and skills required for a technologist to be competent in CMR imaging. By providing a framework for CMR training and competency the overarching goal is to promote the performance of high-quality CMR and to foster the increased adoption of CMR into clinical care.