Explore the vast range of titles available.

As machine learning research in the field of cardiovascular imaging continues to grow, obtaining reliable model performance estimates is critical to develop reliable baselines and compare different algorithms. While the machine learning community has generally accepted methods such as k-fold stratified cross-validation (CV) to be more rigorous than single split validation, the standard research practice in medical fields is the use of single split validation techniques. This is especially concerning given the relatively small sample sizes of datasets used for cardiovascular imaging. We aim to examine how train-test split variation impacts the stability of machine learning (ML) model performance estimates in several validation techniques on two real-world cardiovascular imaging datasets: stratified split-sample validation (70/30 and 50/50 train-test splits), tenfold stratified CV, 10 × repeated tenfold stratified CV, bootstrapping (500 × repeated), and leave one out (LOO) validation. We demonstrate that split validation methods lead to the highest range in AUC and statistically significant differences in ROC curves, unlike the other aforementioned approaches. When building predictive models on relatively small data sets as is often the case in medical imaging, split-sample validation techniques can produce instability in performance estimates with variations in range over 0.15 in the AUC values, and thus any of the alternate validation methods are recommended.

ObjectivesWe aimed to investigate if lesion-specific ischaemia by invasive fractional flow reserve (FFR) can be predicted by an integrated machine learning (ML) ischaemia risk score from quantitative plaque measures from coronary computed tomography angiography (CTA).MethodsIn a multicentre trial of 254 patients, CTA and invasive coronary angiography were performed, with FFR in 484 vessels. CTA data sets were analysed by semi-automated software to quantify stenosis and non-calcified (NCP), low-density NCP (LD-NCP, < 30 HU), calcified and total plaque volumes, contrast density difference (CDD, maximum difference in luminal attenuation per unit area) and plaque length. ML integration included automated feature selection and model building from quantitative CTA with a boosted ensemble algorithm, and tenfold stratified cross-validation.ResultsEighty patients had ischaemia by FFR (FFR ≤ 0.80) in 100 vessels. Information gain for predicting ischaemia was highest for CDD (0.172), followed by LD-NCP (0.125), NCP (0.097), and total plaque volumes (0.092). ML exhibited higher area-under-the-curve (0.84) than individual CTA measures, including stenosis (0.76), LD-NCP volume (0.77), total plaque volume (0.74) and pre-test likelihood of coronary artery disease (CAD) (0.63); p < 0.006.ConclusionsIntegrated ML ischaemia risk score improved the prediction of lesion-specific ischaemia by invasive FFR, over stenosis, plaque measures and pre-test likelihood of CAD.Key Points• Integrated ischaemia risk score improved prediction of ischaemia over quantitative plaque measures• Integrated ischaemia risk score showed higher prediction of ischaemia than standard approach• Contrast density difference had the highest information gain to identify lesion-specific ischaemia

Aims/hypothesis Inflammation is a core component of residual cardiovascular risk in type 2 diabetes. With new anti-inflammatory therapeutics entering the field, accurate markers to evaluate their effectiveness in reducing cardiovascular disease are paramount. Gallium-68-labelled DOTATATE ( 68 Ga-DOTATATE) has recently been proposed as a more specific marker of arterial wall inflammation than 18 F-fluorodeoxyglucose ( 18 F-FDG). This study set out to investigate whether 68 Ga-DOTATATE uptake is amenable to therapeutic intervention in individuals with type 2 diabetes. Methods Individuals aged >50 years with type 2 diabetes underwent 68 Ga-DOTATATE positron emission tomography (PET)/computed tomography (CT) at baseline and after 3 months treatment with atorvastatin 40 mg once daily. Primary outcome was the difference in coronary 68 Ga-DOTATATE uptake, expressed as target-to-background ratio (TBR). The secondary outcome was difference in bone marrow and splenic uptake, expressed as the standardised uptake value (SUV). Results Twenty-two individuals with type 2 diabetes (mean age 63.2±6.4 years, 82% male, LDL-cholesterol 3.42±0.81 mmol/l, HbA 1c 55±12 mmol/mol [7.2%±3.2%]) completed both 68 Ga-DOTATATE PET/CT scans. The maximum TBR was −31% (95% CI −50, −12) lower in the coronary arteries, and bone marrow and splenic 68 Ga-DOTATATE uptake was also significantly lower post statin treatment, with a mean percentage reduction of −15% (95% CI −27, −4) and −17% (95% CI −32, −2), respectively. Conclusions/interpretation 68 Ga-DOTATATE uptake across the cardio–haematopoietic axis was lower after statin therapy in individuals with type 2 diabetes. Therefore, 68 Ga-DOTATATE is promising as a metric for vascular and haematopoietic inflammation in intervention studies using anti-inflammatory therapeutics in individuals with type 2 diabetes. Trial registration ClinicalTrials.gov NCT05730634 Graphical Abstract

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. Chest computed tomography (CT) is one of the most common diagnostic tests. Here, the authors combine two AI models to measure from CT coronary artery calcium, left ventricular mass index, and left and right atrial and ventricular volumes, and show their association with cardiovascular mortality.

18F-sodium fluoride (18F-NaF) positron emission tomography (PET) has emerged as a promising noninvasive imaging tool for the assessment of active calcification processes in coronary artery disease. 18F-NaF uptake colocalizes to high-risk and ruptured atherosclerotic plaques. Most recently, 18F-NaF coronary uptake was shown to be a robust and independent predictor of myocardial infarction in patients with advanced coronary artery disease. In this review, we provide an overview of the advances in coronary 18F-NaF imaging. In particular, we discuss the recently developed and validated motion correction techniques which address heart contractions, tidal breathing, and patient repositioning during the prolonged PET acquisitions. Additionally, we discuss a novel quantification approach—the coronary microcalcification activity (which has been inspired by the widely employed method in oncology total active tumor volume measurement). This new method provides a single number encompassing 18F-NaF activity within the entire coronary vasculature rather than just information regarding a single area of most intense tracer uptake.

BackgroundWe developed machine-learning (ML) models to estimate a patient’s risk of cardiac death based on adenosine myocardial perfusion SPECT (MPS) and associated clinical data, and compared their performance to baseline logistic regression (LR). We demonstrated an approach to visually convey the reasoning behind a patient’s risk to provide insight to clinicians beyond that of a “black box.”MethodsWe trained multiple models using 122 potential clinical predictors (features) for 8321 patients, including 551 cases of subsequent cardiac death. Accuracy was measured by area under the ROC curve (AUC), computed within a cross-validation framework. We developed a method to display the model’s rationale to facilitate clinical interpretation.ResultsThe baseline LR (AUC = 0.76; 14 features) was outperformed by all other methods. A least absolute shrinkage and selection operator (LASSO) model (AUC = 0.77; p = .045; 6 features) required the fewest features. A support vector machine (SVM) model (AUC = 0.83; p < .0001; 49 features) provided the highest accuracy.ConclusionsLASSO outperformed LR in both accuracy and simplicity (number of features), with SVM yielding best AUC for prediction of cardiac death in patients undergoing MPS. Combined with presenting the reasoning behind the risk scores, our results suggest that ML can be more effective than LR for this application.

Patients with cancer treated with immune checkpoint inhibitors (ICIs) are at increased risk of cardiovascular events. Preclinical studies suggest that this may result from inflammation-induced destabilization of atherosclerotic plaques. To evaluate changes in vessel wall inflammation assessed using [18F]FDG positron emission tomography/computed tomography (PET/CT) after ICI initiation. This was a single-center retrospective cohort study of patients with Merkel cell carcinoma who received at least one cycle of the programmed death ligand 1 (PD-L1) inhibitor avelumab and underwent [18F]FDG PET/CT before initiation of treatment and after 3 months. The primary outcome was the change in the target-to-background ratio (TBRmax) in the descending aorta between baseline and first follow-up scan. Secondary outcomes included the change in TBRmax in the carotid arteries, spleen, and bone marrow, and incidence of major adverse cardiovascular events. Fifty-three patients were included (66% male; median age 75 years). Most patients had established risk factors for cardiovascular disease (62%). The [18F]FDG TBRmax in the descending aorta increased from 1.52 (IQR, 1.39-1.70) at baseline to 1.64 (IQR, 1.41-1.97) after 3 months of treatment (change 7.8%. p = 0.022). No significant changes were observed in the carotid arteries, bone marrow, and spleen. Statin use was not associated with an attenuated change in TBRmax. During a median follow-up of 2.3 (IQR, 1.5-4.2) years, one nonfatal ischemic stroke occurred. Avelumab treatment was associated with an increase in [18F]FDG uptake in the descending aorta after 3 months of treatment, which may be a potential marker of inflammation-driven accelerated atherosclerosis in patients receiving ICIs.

Purpose We sought to evaluate inter-scan and inter-reader agreement of coronary calcium (CAC) scores obtained from dedicated, ECG-gated CAC scans (standard CAC scan) and ultra-low-dose, ungated computed tomography attenuation correction (CTAC) scans obtained routinely during cardiac PET/CT imaging. Methods From 2928 consecutive patients who underwent same-day 82 Rb cardiac PET/CT and gated CAC scan in the same hybrid PET/CT scanning session, we have randomly selected 200 cases with no history of revascularization. Standard CAC scans and ungated CTAC scans were scored by two readers using quantitative clinical software. We assessed the agreement between readers and between two scan protocols in 5 CAC categories (0, 1–10, 11–100, 101–400, and > 400) using Cohen’s Kappa and concordance. Results Median age of patients was 70 (inter-quartile range: 63–77), and 46% were male. The inter-scan concordance index and Cohen’s Kappa for readers 1 and 2 were 0.69; 0.75 (0.69, 0.81) and 0.72; 0.8 (0.75, 0.85) respectively. The inter-reader concordance index and Cohen’s Kappa (95% confidence interval [CI]) was higher for standard CAC scans: 0.9 and 0.92 (0.89, 0.96), respectively, vs. for CTAC scans: 0.83 and 0.85 (0.79, 0.9) for CTAC scans ( p  = 0.02 for difference in Kappa). Most discordant readings between two protocols occurred for scans with low extent of calcification (CAC score < 100). Conclusion CAC can be quantitatively assessed on PET CTAC maps with good agreement with standard scans, however with limited sensitivity for small lesions. CAC scoring of CTAC can be performed routinely without modification of PET protocol and added radiation dose.

Purpose18F-sodium fluoride (18F-NaF) has shown promise in assessing disease activity in coronary arteries, but currently used measures of activity – such as maximum target to background ratio (TBRmax) – are defined by single pixel count values. We aimed to develop a novel coronary-specific measure of 18F-NaF PET reflecting activity throughout the entire coronary vasculature (coronary microcalcification activity [CMA]).MethodsPatients with recent myocardial infarction and multi-vessel coronary artery disease underwent 18F-NaF PET and coronary CT angiography. We assessed the association between coronary 18F-NaF uptake (both TBRmax and CMA) and coronary artery calcium scores (CACS) as well as low attenuation plaque (LAP, attenuation < 30 Hounsfield units) volume.ResultsIn 50 patients (64% males, 63 ± 7 years), CMA and TBRmax were higher in vessels with LAP compared to those without LAP (1.09 [0.02, 2.34] versus 0.0 [0.0, 0.0], p < 0.001 and 1.23 [1.16, 1.37] versus 1.04 [0.93, 1.11], p < 0.001). Compared to a TBRmax threshold of 1.25, CMA > 0 had a higher diagnostic accuracy for detection of LAP: sensitivity of 93.1 (83.3–98.1)% versus 58.6 (44.9–71.4)% and a specificity of 95.7 (88.0–99.1)% versus 80.0 (68.7–88.6)% (both p < 0.001).18F-NaF uptake assessed by CMA correlated more closely with LAP (r = 0.86, p < 0.001) than the CT calcium score (r = 0.39, p < 0.001), with these associations outperforming those observed for TBRmax values (LAP r = 0.63, p < 0.001; CT calcium score r = 0.30, p < 0.001).ConclusionsAutomated assessment of disease activity across the entire coronary vasculature is feasible using 18F-NaF CMA, providing a single measurement that has closer agreement with CT markers of plaque vulnerability than more traditional measures of plaque activity.