To investigate the relationship between metabolic dysfunction–associated steatotic liver disease (MASLD) and myocardial ischemia in patients with suspected or known coronary artery disease (CAD). This retrospective study enrolled 281 patients with suspected or known CAD who underwent single-photon emission computed tomography myocardial perfusion imaging (SPECT-MPI) at the Third Affiliated Hospital of Soochow University from January 1, 2022, to December 31, 2023. The mean CT values of the liver and spleen, coronary artery calcification score (CACS), and epicardial fat volume (EFV) were acquired through non-enhanced computed tomography (CT). Myocardial ischemia is defined by SDS ≥ 2 detected by MPI. Obstructive CAD is defined as the degree of coronary artery lumen narrowing ≥ 50

Background:The impact of epicardial adipose tissue (EAT) on the risk of non-obstructive coronary artery disease (CAD) remains unclear. This study aims to investigate the association between EAT and ischemia with non-obstructive coronary arteries (INOCA). Methods:This study enrolled 281 patients with angina or other symptoms suggestive of myocardial ischemia who underwent single-photon emission computed tomography myocardial perfusion imaging (SPECT-MPI). All patients had confirmed non-obstructive coronary artery disease (stenosis <50%) by either coronary angiography (CAG) or coronary CT angiography (CCTA) within 3 months before or after MPI. Based on MPI results, patients were categorized into ischemic and non-ischemic groups. Epicardial adipose tissue (EAT) density and volume were measured, and relevant clinical parameters were collected for analysis. Results:The results revealed that 37.72% of the patients had INOCA, and these patients exhibited significantly higher body mass index (BMI) and EAT density. No statistically significant difference in EAT volume was observed between groups. Both EAT density (OR = -1.846, 95% CI: 1.353-2.559, p < 0.05) and volume (OR = -1.703, 95% CI: 1.151-2.551, p < 0.05) were identified as independent risk factors for INOCA. Furthermore, EAT density demonstrated a linear relationship with disease risk. In statin users, the positive association between EAT density and INOCA was attenuated. (β = -0.039, p = 0.046). Conclusions:EAT density is an independent risk factor for INOCA, with its increase showing a linear association with INOCA risk. Further, statin use was associated with a reduction in this EAT density-related INOCA risk.

Background:Epicardial adipose tissue (EAT) is associated with coronary artery disease (CAD) and adverse cardiovascular outcomes; however, its prognostic relevance across different cardiovascular risk scores remains uncertain. This study evaluated the long-term prognostic significance of EAT, coronary anatomical findings, and functional ischemia in patients with suspected CAD stratified by Framingham risk score (FRS) categories. Methods:A consecutive retrospective cohort of 361 symptomatic patients who underwent both single-photon emission computed tomography/computed tomography myocardial perfusion imaging (SPECT/CT MPI) and coronary angiography (CAG) or coronary computed tomography angiography (CCTA) at baseline was analyzed. Epicardial fat volume (EFV) and coronary artery calcium score (CACS) were quantified from integrated computed tomography (CT) with SPECT/CT. Ischemia burden was defined as ≥5% ischemic myocardium on MPI, and obstructive CAD as ≥50% stenosis. Major adverse cardiovascular events (MACE) were recorded during follow-up. Results:Over a median follow-up of 4.6 years, 54 patients (15%) experienced MACE. In the multivariable analysis, EFV [adjusted hazard ratio (aHR) =2.09; 95% confidence interval (CI): 1.08-4.05; P=0.029], CACS (aHR =2.58; 95% CI: 1.39-4.80; P=0.003), and obstructive CAD (aHR =3.07; 95% CI: 1.59-5.90; P<0.001) were independently associated with MACE, while ischemia burden showed borderline significance (aHR =2.27; 95% CI: 0.94-5.49; P=0.069). EFV improved risk discrimination over FRS, CACS, obstructive CAD, and ischemia burden (all P<0.05). In the stratified analyses, CACS (aHR =2.88; P=0.017) and obstructive CAD (aHR =3.69; P=0.005) predicted MACE in low-to-intermediate FRS patients, whereas EFV (aHR =3.13; P=0.020) and ischemia burden (aHR =4.67; P=0.005) were only associated with MACE in high-risk FRS patients. EFV was significantly associated with MACE in patients with diabetes (aHR =6.61; P=0.010), but not in those without diabetes (aHR =1.45; P=0.359), with ischemia showing a similar pattern. Conclusions:EFV independently predicted long-term MACE and improved prognostic discrimination beyond clinical risk, anatomical imaging, and functional ischemia. Prognostic determinants varied by cardiovascular risk level, with anatomical markers being more informative in lower-risk patients, while EFV and functional ischemia were more informative in high-risk and diabetic patients. These findings support a risk-adapted strategy for the use of quantitative imaging biomarkers in patients with suspected CAD.

Background:Identifying obstructive coronary artery disease (OCAD) via non-invasive imaging modalities in patients with suspected unstable angina (UA) holds substantial clinical significance. This study aimed to develop and validate a diagnostic model for OCAD in patients with suspected UA, by leveraging resting 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) myocardial ischemia memory imaging combined with clinical indicators. Methods:We retrospectively analyzed 162 patients with a Global Registry of Acute Coronary Events (GRACE) score ≤140 who presented with chest pain or chest tightness and were clinically suspected of having UA. After collecting clinical indicators, predictive factors were screened using logistic regression. A diagnostic model was constructed using binary logistic regression based on the predictive factors, with internal validation via 1,000 bootstrap resamples. The discriminative ability, calibration, and clinical net benefit of the established model were evaluated by the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). Furthermore, the enhancement value of the final model over the basic model was quantified using net reclassification improvement (NRI) and integrated discrimination improvement (IDI). Results:Of the 162 enrolled patients with suspected UA, 89 (54.9%) were diagnosed with OCAD. Six predictors were incorporated into the diagnostic model, including hyperlipidemia, diabetes, typical angina pectoris, 18F-FDG PET results, serum creatinine, and left ventricular ejection fraction (LVEF). The area under the curve (AUC) of the model was 0.89 [95% confidence interval (CI): 0.84-0.94], with a sensitivity of 0.84 and a specificity of 0.81; the Brier score was 0.1319. The Hosmer-Lemeshow goodness-of-fit test revealed good model fit (χ2=6.15, P=0.63). After internal validation via the bootstrap method, the optimism-corrected AUC was 0.87 (95% CI: 0.82-0.93). Calibration curve and DCA demonstrated that the model exhibited satisfactory calibration and promising clinical utility. Conclusions:The OCAD diagnostic model for suspected UA patients, based on resting 18F-FDG PET and clinical indicators, demonstrated excellent diagnostic performance.

Epicardial adipose tissue (EAT) plays an important role in the pathogenesis of coronary artery disease (CAD). The association between EAT and obstructive CAD or myocardial ischemia has been established, but its relationship with CAD phenotypes based on anatomical and functional imaging remains unclear. A total of 495 suspected CAD patients who underwent both single-photon emission computed tomography/computed tomography myocardial perfusion imaging (SPECT/CT MPI) and coronary angiography (CAG/CTA) were enrolled in this retrospective study. Epicardial fat volume (EFV) and epicardial fat volume indexed to body surface (EFVi) were measured on non-contrast CT. CAD phenotypes were categorized into 4 groups based on the presence or absence of obstructive CAD (any epicardial coronary diameter stenosis ≥ 50

Objective Major adverse cardiovascular events (MACE) still occur in the normal left ventricular ejection fraction (LVEF) patients with coronary artery disease (CAD). Currently, there are no studies related to the prognostic value of left ventricular diastolic dyssynchrony (LVDD) in combination with perfusion, systolic dyssynchrony, and cardiovascular risk factors in patients with normal LVEF. Therefore, we aimed to investigate the incremental prognostic value of LVDD in patients with normal LVEF and to establish a model to predict MACE. Methods This study included 239 suspected or known CAD patients with a normal LVEF who underwent gated single-photon emission computerized tomography myocardial perfusion imaging. Clinical data such as age, sex, and cardiovascular risk factors were collected. Myocardial perfusion, and left ventricular dyssynchrony parameters were assessed using QPS and Emory Toolbox software, respectively. The least absolute shrinkage and selection operator and multivariable Cox regression were used to select the variables. Results The subjects were followed up for a total of 73.2±16.4 months and MACE occurred in 28 patients. In multivariate Cox regression, rest diastolic bandwidth (BW) was closely related to MACE [hazard ratio (95% confidence interval), 10.78 (1.65–70.35); P =0.013]. The C-index of the model was increased from 0.748 to 0.783 by increasing the rest diastolic BW on the basis of summed difference score (SDS), stress systolic SD, age, hypertension, and chest pain ( P <0.001). A final model for predicting MACE was constructed based on age, hypertension, chest pain, SDS, stress systolic SD, and rest diastolic BW. The C-index of the model was 0.783, and the area under the curves of the model predicting the occurrence of 3-year and 5-year MACE events were 0.766 and 0.827, respectively. The calibration curve showed a good calibration of the model. Conclusion LVDD is associated with MACE in patients with normal LVEF. In addition, based on SDS, stress systolic SD, age, hypertension, and chest pain, rest diastolic BW had an incremental predictive value for MACE.

Background:This study aimed to evaluate the correlation between the metabolic score for insulin resistance (METS-IR) and myocardial ischemia based on myocardial perfusion imaging (MPI) and further examine whether non-alcoholic fatty liver disease (NAFLD) has a potential role in mediating these associations. Methods:This retrospective study enrolled 1,242 patients with suspected coronary artery disease (CAD) who underwent single-photon emission computed tomography myocardial perfusion imaging (SPECT-MPI) at the Third Affiliated Hospital of Soochow University from 1 January 2022 to 31 December 2024. The association between METS-IR and myocardial ischemia was analyzed using the logistic regression model. The mediating effect of NAFLD was evaluated through mediation analysis to explore the potential mechanism underlying the association between METS-IR and myocardial ischemia. Results:The final group of participants included 335 patients; 179 (53.4%) patients had myocardial ischemia. Overall, the mean age was 61.45 ± 10.20 years, 188 (56.1%) were men, and the mean body mass index was 24.69 ± 3.21 kg/m2. Mean METS-IR was 2.40 ± 0.22 (range, 1.79-3.80). The results of the single-factor analysis showed that a per-SD increase in METS-IR was independently associated with myocardial ischemia (OR, 3.92; 95% CI: 1.38-11.08; P =0.010). After adjusting for all interfering factors, METS-IR had no associations with myocardial ischemia. The results of the mediation analysis show that NAFLD is a complete mediator variable between MERSIR and myocardial ischemia. Conclusion:This study provides evidence for the relationship between METS-IR and myocardial ischemia and highlights the important mediating role of NAFLD in this relationship.

Background: Atrial fibrillation (AF) has been identified to increase stroke risk, even after oral anticoagulants (OACs), and the recurrence rate is high after radiofrequency catheter ablation (RFCA). Inflammation is an essential factor in the occurrence and persistence of AF. 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) is an established molecular imaging modality to detect local inflammation. We aimed to investigate the relationship between atrial inflammatory activity and poor prognosis of AF based on 18F-FDG PET/CT.Methods: A total of 204 AF patients including 75 with paroxysmal AF (ParAF) and 129 with persistent AF (PerAF) who underwent PET/CT before treatment were enrolled in this prospective cohort study. Clinical data, electrocardiograph (ECG), echocardiography, and cardiac 18F-FDG uptake were collected. Follow-up information was obtained from patient clinical case notes or telephone reviews, with the starting point being the time of PET/CT scan. The follow-up deadline was either the date of AF recurrence after RFCA, new-onset stroke, or May 2023. Cox proportional hazards regression models were used to identify predictors of poor prognosis and hazard ratios (HRs) with 95% confidence intervals (CIs) was calculated.Results: Median follow-up time was 29 months [interquartile range (IQR), 22-36 months]. Poor prognosis occurred in 52 patients (25.5%), including 34 new-onset stroke patients and 18 recrudescence after RFCA. The poor prognosis group had higher congestive heart failure, hypertension, age >= 75 years (doubled), diabetes mellitus, prior stroke or transient ischemic attack (TIA) or thromboembolism (doubled), vascular disease, age 65-74 years, sex category (female) (CHA2DS2-VASc) score [3.0 (IQR, 1.0, 3.75) vs. 2.0 (IQR, 1.0, 3.0), P=0.01], right atrial (RA) wall maximum standardized uptake value (SUVmax) (4.13 +/- 1.82 vs. 3.74 +/- 1.58, P=0.04), higher percentage of PerAF [39 (75.0%) vs. 90 (59.2%), P=0.04], left atrial (LA) enlargement [45 (86.5%) vs. 104 (68.4%), P=0.01], and RA wall positive FDG uptake [40 (76.9%) vs. 79 (52.0%), P=0.002] compared with the non-poor prognosis group. Univariate and multivariate Cox proportional hazard regression analysis concluded that only CHA2DS2-VASc score (HR, 1.29; 95% CI: 1.06-1.57; P=0.01) and RA wall positive FDG uptake (HR, 2.68; 95% CI: 1.10-6.50; P=0.03) were significantly associated with poor prognosis.Conclusions: RA wall FDG positive uptake based on PET/CT is tightly related to AF recurrence after RFCA or new-onset stroke after antiarrhythmic and anticoagulation treatment.

Background:Under the Breast Imaging Reporting and Data System (BI-RADS), category 4 lesions have a high probability of malignancy. This study sought to investigate the efficacy of a model that combined the BI-RADS score with the enhancement score and clinical indicators in the diagnosis of BI-RADS 4 lesions based on contrast-enhanced spectral mammography (CESM) in breast cancer patients. Methods:The data of female patients with BI-RADS scores of 4 who underwent CESM at the Department of Medical Imaging of the Third Affiliated Hospital of Soochow University from January 2018 to July 2023 were retrospectively collected. In total, 170 patients were enrolled in the study. Based on their surgery or puncture pathology results, the patients were divided into malignant and benign groups. The clinical data, imaging characteristics, and enhancement degree of the patients in the two groups were compared. Model 3, which combined the BI-RADS score, enhancement score, and clinical indicators, was constructed using logistic regression. The predictive performance of Model 3 was evaluated and compared with Model 1 (BI-RADS score) and Model 2 (BI-RADS score + enhancement score). Results:Of the 170 patients, 69 had benign lesions and 101 had malignant lesions. There were significant differences between the malignant and benign groups in terms of age, menopause, a family history breast cancer, BI-RADS score, and enhancement score (all P<0.05). The areas under the curve (AUCs) of the receiver operator characteristic curves of Models 1, 2, and 3 were 0.830, 0.858, and 0.900, respectively. The best cut-off value for Model 3 was 0.766, with a sensitivity of 74.3% and a specificity of 94.2%. Based on the AUCs and decision curves, Model 3 performed better than Models 1 and 2. The calibration curve (intercept: 0.034; slope: 0.807) was plotted using bootstrap re-sampling (500 times), and showed good agreement between the predicted probability and the actual prevalence. Conclusions:In the suspected breast cancer patients with a BI-RADS score of 4, the combination of the enhancement score and clinical indicators based on the BI-RADS score improved the efficiency of CESM in diagnosing breast cancer.

BACKGROUND: Sex-specific differences in coronary phenotypes in response to stress have not been elucidated. This study investigated the sex-specific differences in the coronary computed tomography angiography-assessed coronary response to mental stress. METHODS: This retrospective study included patients with coronary artery disease and without cancer who underwent resting 18 F-fluorodexoyglucose positron emission tomography/computed tomography and coronary computed tomography angiography within 3 months. 18 F-flourodeoxyglucose resting amygdalar uptake, an imaging biomarker of stress-related neural activity, coronary inflammation (fat attenuation index), and high-risk plaque characteristics were assessed by coronary computed tomography angiography. Their correlation and prognostic values were assessed according to sex. RESULTS: A total of 364 participants (27.7% women and 72.3% men) were enrolled. Among those with heightened stress-related neural activity, women were more likely to have a higher fat attenuation index (43.0% versus 24.0%; P =0.004), while men had a higher frequency of high-risk plaques (53.7% versus 39.3%; P =0.036). High amygdalar 18 F-flourodeoxyglucose uptake (B-coefficient [SE], 3.62 [0.21]; P <0.001) was selected as the strongest predictor of fat attenuation index in a fully adjusted linear regression model in women, and the first-order interaction term consisting of sex and stress-related neural activity was significant ( P <0.001). Those with enhanced imaging biomarkers of stress-related neural activity showed increased risk of major adverse cardiovascular event both in women (24.5% versus 5.1%; adjusted hazard ratio, 3.62 [95% CI, 1.14–17.14]; P =0.039) and men (17.2% versus 6.9%; adjusted hazard ratio, 2.72 [95% CI, 1.10–6.69]; P =0.030). CONCLUSIONS: Imaging-assessed stress-related neural activity carried prognostic values irrespective of sex; however, a sex-specific mechanism linking psychological stress to coronary plaque phenotypes existed in the current hypothesis-generating study. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT05545618.

Abstract Background The metabolic tumour area (MTA) was found to be a promising predictor of prostate cancer. However, the role of MTA based on 18F-FDG PET/CT in diffuse large B-cell lymphoma (DLBCL) prognosis remains unclear. This study aimed to elucidate the prognostic significance of MTA and evaluate its incremental value to the National Comprehensive Cancer Network International Prognostic Index (NCCN-IPI) for DLBCL patients treated with first-line R-CHOP regimens. Methods A total of 280 consecutive patients with newly diagnosed DLBCL and baseline 18F-FDG PET/CT data were retrospectively evaluated. Lesions were delineated via a semiautomated segmentation method based on a 41% SUVmax threshold to estimate semiquantitative metabolic parameters such as total metabolic tumour volume (TMTV) and MTA. Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cut-off values. Progression-free survival (PFS) and overall survival (OS) were the endpoints that were used to evaluate the prognosis. PFS and OS were estimated via Kaplan‒Meier curves and compared via the log-rank test. Results Univariate analysis revealed that patients with high MTA, high TMTV and NCCN-IPI ≥ 4 were associated with inferior PFS and OS (P < 0.0001 for all). Multivariate analysis indicated that MTA remained an independent predictor of PFS and OS [hazard ratio (HR), 2.506; 95% confidence interval (CI), 1.337–4.696; P = 0.004; and HR, 1.823; 95% CI, 1.005–3.310; P = 0.048], whereas TMTV was not. Further analysis using the NCCN-IPI model as a covariate revealed that MTA and NCCN-IPI were still independent predictors of PFS (HR, 2.617; 95% CI, 1.494–4.586; P = 0.001; and HR, 2.633; 95% CI, 1.650–4.203; P < 0.0001) and OS (HR, 2.021; 95% CI, 1.201–3.401; P = 0.008; and HR, 3.869; 95% CI, 1.959–7.640; P < 0.0001; respectively). Furthermore, MTA was used to separate patients with high NCCN-IPI risk scores into two groups with significantly different outcomes. Conclusions Pre-treatment MTA based on 18F-FDG PET/CT and NCCN-IPI were independent predictor of PFS and OS in DLBCL patients treated with R-CHOP. MTA has additional predictive value for the prognosis of patients with DLBCL, especially in high-risk patients with NCCN-IPI ≥ 4. In addition, the combination of MTA and NCCN-IPI may be helpful in further improving risk stratification and guiding individualised treatment options. Trial registration This research was retrospectively registered with the Ethics Committee of the Third Affiliated Hospital of Soochow University, and the registration number was approval No. 155 (approved date: 31 May 2022).

Background: Recent studies have shown that resting amygdalar activity is associated with cardiovascular disease. Nevertheless, the underlying mechanisms that link resting amygdalar activity with persistent atrial fibrillation (PerAF) remain to be comprehensively delineated. We aimed to estimate the association between resting amygdalar activity, right atrium (RA) inflammatory activity, and PerAF. Methods: In the retrospective cohort study, 18F-fluorodeoxyglucose positron emission tomographycomputed tomography (18F-FDG-PET/CT) imaging was performed in 104 patients with AF between January 1, 2018, and December 31, 2022 in the third Affiliated Hospital of Soochow University. Clinical data, electrocardiograms, echocardiographic assessments, and cardiac 18F-FDG uptake measurements were systematically gathered. Validated methodologies were employed to assess resting amygdalar activity, right atrium target:background ratio (RATBR), and bone-marrow activity (BMA). Associations between resting amygdalar activity and PerAF were assessed via a logistic regression model and mediation (path) analyses. Results: Among the patients included, 60 (57.7%) had PerAF. Compared with patients with paroxysmal AF (ParAF), those with PerAF had higher resting amygdalar activity (1.11 +/- 0.16 vs. 1.47 +/- 0.27), BMA (1.66 +/- 0.48 vs. 2.12 +/- 0.65), and RATBR (1.55 +/- 0.36 vs. 2.31 +/- 0.73) (all P values <0.001). Resting amygdalar activity demonstrated a statistically significant correlation with BMA (r=0.38; P<0.001) and RATBR (r=0.44; P<0.001). BMA significantly mediated the associations between resting amygdalar activity and RATBR, accounting for 50.2% [95% confidence interval (CI): 35.8-76.7%] of this association. Resting amygdalar activity and RATBR emerged as the sole independent variable for PerAF (odds ratio =6.81; 95% CI: 2.34- significantly mediated the associations between resting amygdalar activity and PerAF, accounting for 33.2% (95% CI: 16.6-52.4%) of this association. Conclusions: Resting amygdalar activity and RATBR evaluated by PET/CT were significantly associated with PerAF. The association of testing amygdalar activity with PerAF was in part mediated by RA inflammatory activity, which could be a potential therapeutic target for PerAF. Further prospective research is needed to determine the generalizability of these findings.

BACKGROUND Stress-related neural activity (SNA) assessed by amygdalar activity can predict cardiovascular events. However, its mechanistic linkage with plaque vulnerability is not fully elucidated. OBJECTIVES The authors aimed to investigate the association of SNA with coronary plaque morphologic and inflam-matory features as well as their ability in predicting major adverse cardiovascular events (MACE).METHODS A total of 299 patients with coronary artery disease (CAD) and without cancer underwent 18F-fluorodex-oyglucose positron emission tomography/computed tomography (PET/CT) and available coronary computed tomographic angiography (CCTA) between January 1, 2013, and December 31, 2020. SNA and bone-marrow activity (BMA) were assessed with validated methods. Coronary inflammation (fat attenuation index [FAI]) and high-risk plaque (HRP) characteristics were assessed by CCTA. Relations between these features were analyzed. Relations between SNA and MACE were assessed with Cox models, log-rank tests, and mediation (path) analyses.RESULTS SNA was significant correlated with BMA (r = 0.39; P < 0.001) and FAI (r = 0.49; P < 0.001). Patients with heightened SNA are more likely to have HRP (40.7% vs 23.5%; P = 0.002) and increase risk of MACE (17.2% vs 5.1%, adjusted HR 3.22; 95% CI: 1.31-7.93; P = 0.011). Mediation analysis suggested that higher SNA associates with MACE via a serial mechanism involving BMA, FAI, and HRP.CONCLUSIONS SNA is significantly correlated with FAI and HRP in patients with CAD. Furthermore, such neural activity was associated with MACE, which was mediated in part by leukopoietic activity in the bone marrow, coronary inflammation, and plaque vulnerability. (J Am Coll Cardiol Img 2023;16:1404-1415) (c) 2023 by the American College of Cardiology Foundation.

Background The rest-only single photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) has low diagnostic performance for obstructive coronary artery disease (CAD). Coronary artery calcium score (CACS) is strongly associated with obstructive CAD. The aim of this study was to investigate the performance of rest-only gated SPECT MPI combined with CACS and cardiovascular risk factors in diagnosing obstructive CAD through machine learning (ML). Methods We enrolled 253 suspected CAD patients who underwent the 1-stop rest-only SPECT MPI and computed tomography (CT) scan due to stress test-related contraindications. Myocardial perfusion and wall motion were assessed using quantitative perfusion SPECT + quantitative gated SPECT (QPS + QGS) automated quantification software. The Agatston algorithm was used to calculate CACS. The clinical data of patients, including cardiovascular risk factors, were collected. Based on feature selection and clinical experience, 8 factors were identified as modeling variables. Subsequently, patients were divided randomly into 2 groups: the training (70%) and test (30%) groups. The performance of 8 supervised ML algorithms was evaluated in the training and test groups. Results Obstructive CAD was diagnosed by coronary angiography in 94 (37.2%, 94/253) patients. In the training group, the area under the receiver operator characteristic (ROC) curve (AUC) of the random forest was the highest, and the AUCs of Logistic, extreme gradient boosting (XGBoost), support vector machine (SVM), and adaptive boosting (AdaBoost) were all above 0.9. In the test group, the AUC of recursive partitioning and regression trees (Rpart) was the highest (0.911). Rpart and Naïve Bayes had the highest accuracy (0.840). Rpart had a sensitivity and specificity of 0.851 and 0.821, respectively; Naïve Bayes had a sensitivity and specificity of 0.809 and 0.893, respectively. Next was Logistic, with an accuracy of 0.827, a sensitivity of 0.872, and a specificity of 0.750. The random forest and XGBoost algorithms also had high accuracy, which was 0.813 for each algorithm. Conclusions Rest-only SPECT MPI combined with CACS and cardiovascular risk factors using an ML algorithm to detect obstructive CAD is feasible. Among the algorithms validated in the test group, Rpart, Naïve Bayes, XGBoost, Logistic, and random forest are all highly accurate for diagnosing obstructive CAD. The application of ML in resting MPI and CACS may be used for screening obstructive CAD.

Background: F-18-fluorodeoxyglucose positron emission tomography-computed tomography ( F-18-FDG PET-CT) is typically used to screen malignancy in patients with dermatomyositis (DM). The aim of this study was to investigate the value of using PET-CT in assessing the prognosis of patients with DM and without malignant tumors. Methods: A total of 62 patients with DM who underwent F-18-FDG PET-CT were enrolled in the retrospective cohort study. Clinical data and laboratory indicators were obtained. The muscle max standardized uptake value (SUVmax), splenic SUVmax, target-to-background ratio (TBR) of the aorta, pulmonary highest value (hv)/SUVmax, epicardial fat volume (EFV), and coronary artery calcium (CAC) were measured using F-18-FDG PET-CT. The follow-up was conducted until March 2021, and the endpoint was death from any cause. Univariable and multivariable Cox regression analyses were used to analyze prognostic factors. The survival curves were produced with the Kaplan-Meier method. Results: The median duration of follow-up was 36 [interquartile range (IQR), 14-53] months. The survival rates were 85.2% and 73.4% for 1 and 5 years, respectively. A total of 13 (21.0%) patients died during a median follow-up of 7 (IQR, 4-15.5) months. Compared with the survival group, the death group had significantly higher levels of C-reactive protein [CRP; median (IQR), 4.2 (3.0, 6.0) vs. 6.30 (3.7, 22.8)], hypertension [7 (14.3%) vs. 6 (46.2%)], interstitial lung disease [ILD; 26 (53.1%) vs. 12 (92.3%)], positive anti-Ro52 antibody [19 (38.8%) vs. 10 (76.9%)], pulmonary FDG uptake [median (IQR), 1.8 (1.5, 2.9) vs. 3.5 (2.0, 5.8)], CAC [1 (2.0%) vs. 4 (30.8%)], and EFV [median (IQR), 74.1 (44.8, 92.1) vs. 106.5 (75.0, 128.5)] (all P values < 0.001). Univariable and multivariable Cox analyses identified high pulmonary FDG uptake [hazard ratio (HR), 7.59; 95% confidence interval (CI), 2.08-27.76; P=0.002] and high EFV (HR, 5.86; 95% CI, 1.77-19.42; P=0.004) as independent risk factors for mortality. The survival rate was significantly lower in patients with the concurrent presence of high pulmonary FDG uptake and high EFV. Conclusions: Pulmonary FDG uptake and EFV detected with PET-CT were independent risk factors for death in patients with DM and without malignant tumors. Patients with the concurrent presence of high pulmonary FDG uptake and high EFV had a worse prognosis compared with patients with 1 or neither of these two risk factors. Early treatment should be applied in patients with concurrent presence of high pulmonary FDG uptake and high EFV to improve the survival rate.

Background Epicardial adipose tissue (EAT) is closely related to coronary artery disease (CAD). Hemodynamically significant CAD has a worse prognosis and is more likely to benefit from revascularization. However, the specific relationship between EAT and hemodynamically significant CAD remains unclear. Methods A total of 164 inpatients received single-photon emission computerized tomography-myocardial perfusion imaging (SPECT/MPI) and coronary angiography (CAG) between March 2018 and October 2019 at the Third Affiliated Hospital of Soochow University were enrolled in the retrospective cross-sectional study. Data on body mass index (BMI), hypertension, hyperlipidemia, diabetes mellitus (DM), active smoking, and symptoms were gathered. Epicardial fat volume (EFV) and coronary artery calcium (CAC) were quantified by noncontrast computed tomography (CT). Hemodynamically significant CAD was defined by coronary stenosis severity ≥50% with reversible perfusion defects in the corresponding areas of SPECT/MPI. Results A total of 37.8% of patients had hemodynamically significant CAD. Age and BMI increased with tertiles of EFV (P for trend =0.009 and P<0.001). The ratios of hemodynamically significant CAD in EFV from low to high were 16.4%, 37.0%, and 60.0%, respectively (P for the trend <0.001). In univariate regression analysis, EFV was associated with hemodynamically significant CAD [odds ratio (OR) per 10 cm3 =1.36; 95% confidence interval (CI): 1.20–1.55; P<0.001]. After correcting for traditional risk factors and CAC, EFV was firmly linked to hemodynamically significant CAD (OR per 10 cm3 =1.53; 95% CI: 1.25–1.88; P<0.001). With an increasing trend in EFV for the tripartite groups, the likelihood of hemodynamically significant CAD increased significantly (P for trend <0.001). There was a saturation effect between EFV and hemodynamically significant CAD according to the generalized additive model (GAM). When EFV <134.43 cm3, EFV was linearly correlated with hemodynamically significant CAD (OR per 10 cm3 =2.06; 95% CI: 1.45–2.94; P<0.001). When EFV ≥134.43 cm3, the hemodynamically significant CAD risk was steeper and gradually reached saturation. Hypertension affected the relationship between EFV and hemodynamically significant CAD (P for the interaction =0.02) with an interaction effect. Conclusions There was a robust relationship between EFV and hemodynamically significant CAD. After adjustment for confounders, we found that the risk of hemodynamically significant CAD onset increased nonlinearly for EFV above 134.4 cm3. This refined understanding of the relationship is helpful for the accurate clinical prediction of hemodynamically significant CAD.

We sought to establish an explainable machine learning (ML) model to screen for hemodynamically significant coronary artery disease (CAD) based on traditional risk factors, coronary artery calcium (CAC) and epicardial fat volume (EFV) measured from non-contrast CT scans. 184 symptomatic inpatients who underwent Single Photon Emission Computed Tomography/Myocardial Perfusion Imaging (SPECT/MPI) and Invasive Coronary Angiography (ICA) were enrolled. Clinical and imaging features (CAC and EFV) were collected. Hemodynamically significant CAD was defined when coronary stenosis severity ≥ 50% with a matched reversible perfusion defect in SPECT/MPI. Data was randomly split into a training cohort (70%) on which five-fold cross-validation was done and a test cohort (30%). The normalized training phase was preceded by the selection of features using recursive feature elimination (RFE). Three ML classifiers (LR, SVM, and XGBoost) were used to construct and choose the best predictive model for hemodynamically significant CAD. An explainable approach based on ML and the SHapley Additive exPlanations (SHAP) method was deployed to generate individual explanation of the model’s decision. In the training cohort, hemodynamically significant CAD patients had significantly higher age, BMI and EFV, higher proportions of hypertension and CAC comparing with controls ( P all < .05). In the test cohorts, hemodynamically significant CAD had significantly higher EFV and higher proportion of CAC. EFV, CAC, diabetes mellitus (DM), hypertension, and hyperlipidemia were the highest ranking features by RFE. XGBoost produced better performance (AUC of 0.88) compared with traditional LR model (AUC of 0.82) and SVM (AUC of 0.82) in the training cohort. Decision Curve Analysis (DCA) demonstrated that XGBoost model had the highest Net Benefit index. Validation of the model also yielded a favorable discriminatory ability with the AUC, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of 0.89, 68.0%, 96.8%, 94.4%, 79.0% and 83.9% in the XGBoost model. A XGBoost model based on EFV, CAC, hypertension, DM and hyperlipidemia to assess hemodynamically significant CAD was constructed and validated, which showed favorable predictive value. ML combined with SHAP can offer a transparent explanation of personalized risk prediction, enabling physicians to gain an intuitive understanding of the impact of key features in the model.

Aim Atrial fibrillation (AF) is a progressive disease from paroxysmal to persistent, and persistent AF (PerAF) had worse prognosis. AF has potential link with inflammation, but it is not clear whether PerAF or paroxysmal AF (ParAF) is more closely related to inflammation. On the basis of inhibiting myocardial physiological uptake, 18 F-fluorodeoxyglucosepositron emission tomography/computed tomography ( 18 F-FDG PET/CT) is an established imaging modality to detect cardiac inflammation. We aimed to decipher the association between AF and atrial inflammatory activity by 18 F-FDG PET/CT. Methods Thirty-five PerAF patients were compared to age and sex matched ParAF group with baseline 18 F-FDG PET/CT scans prior to radiofrequency catheter ablation (RFCA) in the prospective case-control study. High-fat and low-carbohydrate diet and prolonged fast (HFLC+Fast) was applied to all AF patients before PET/CT. Then 22 AF patients with positive right atrial (RA) wall FDG uptake (HFLC+Fast) were randomly selected and underwent HFLC+Fast+heparin the next day. The CHA2DS2-VASc score was calculated to evaluate the risk of stroke. Clinical data, ECG, echocardiography, and atrial 18 F-FDG uptake were compared. Results PerAF patients had significantly higher probability of RA wall positive FDG uptake and higher SUVmax than ParAF group [91.4% VS. 28.6%, P < 0.001; SUVmax: 4.10(3.20–4.90) VS. 2.60(2.40–3.10), P < 0.001]. Multivariate logistic regression analyses demonstrated that RA wall SUV max was the independent influencing factor of PerAF (OR = 1.80, 95%CI 1.02–3.18, P = 0.04). In 22 AF patients with RA wall positive FDG uptake (HFLC+Fast), the “HFLC+Fast+Heparin” method did not significantly change RA wall FDG uptake evaluated by either quantitative analysis or visual analysis. High CHA2DS2-VASc score group had higher RA wall 18 F-FDG uptake [3.35 (2.70, 4.50) vs, 2.8 (2.4, 3.1) P = 0.01]. Conclusions RA wall FDG positive uptake was present mainly in PerAF. A higher RA wall 18 F-FDG uptake was an independent influencing factor of PerAF. RA wall FDG uptake based on 18 F-FDG PET/CT may indicate pathological inflammation. Trial registration http://www.chictr.org.cn , ChiCTR2000038288.