RESUMO
INTRODUCTION: Manual Coronary Artery Calcium (CAC) scoring, crucial for assessing coronary artery disease risk, is time-consuming and variable. Deep learning, particularly through Convolutional Neural Networks (CNNs), promises to automate and enhance the accuracy of CAC scoring, which this study investigates. METHODS: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we conducted a comprehensive literature search across PubMed, Embase, Web of Science, and IEEE databases from their inception until November 1, 2023, and selected studies that employed deep learning for automated CAC scoring. We then evaluated the quality of these studies by using the Checklist for Artificial Intelligence in Medical Imaging and the Quality Assessment of Diagnostic Accuracy Studies 2. The main metric for evaluation was Cohen's kappa statistic, indicating an agreement between deep learning models and manual scoring methods. RESULTS: A total of 25 studies were included, with a pooled kappa statistic of 83 % (95 % CI of 79 %-87 %), indicating strong agreement between automated and manual CAC scoring. Subgroup analysis revealed performance variations based on imaging modalities and technical specifications. Sensitivity analysis confirmed the reliability of the results. CONCLUSIONS: Deep learning models, particularly CNNs, have great potential for use in automated CAC scoring applications, potentially enhancing the efficiency and accuracy of risk assessments for coronary artery disease. Further research and standardization are required to address the major heterogeneity and performance disparities between different imaging modalities. Overall, our findings underscore the evolving role of artificial intelligence in advancing cardiac imaging and patient care.
RESUMO
Background: Proximal protection devices, such as the Mo.Ma system provides better neurological outcomes than the distal filter system in the carotid artery stenting (CAS) procedure. This study first evaluated the safety and efficacy of the Mo.Ma system during CAS in a single tertiary referral hospital from Taiwan. The outcomes of distal vs. proximal embolic protection devices were also studied. Methods: A total of 294 patients with carotid artery stenosis who underwent the CAS procedure were retrospectively included and divided into two groups: 152 patients in the distal filter system group and 142 patients in the Mo.Ma system. The outcomes of interest were compared between the two groups. The factors contributing to occlusion intolerance (OI) in the Mo.Ma system were evaluated. Results: The procedure success rates were more than 98% in both groups. No major stroke occurred in this study. The minor stroke rates were 2.8% (4/142) and 4.6% (7/152) in the Mo.Ma system and filter system, respectively (p = 0.419). Patients with hypoalbuminemia significantly predicted the risk of stroke with an odds ratio of 0.08 [95% confidence interval (CI), 0.01-0.68, p = 0.020] per 1 g/day of serum albumin in the filter group. A total of 12 patients developed OI in the Mo.Ma system (12/142, 8%). Low occlusion pressure predicted the occurrence of OI in the Mo.Ma group with the hazard ratios of 0.88 (95% CI: 0.82-0.96) and 0.90 (95% CI: 0.84-0.98) per 1 mmHg of occlusion systolic pressure (OSP) and diastolic pressure (ODP), respectively. We further indicated that patients with an OSP of ≥60 mmHg or an ODP of ≥44 mmHg could tolerate the procedure of occlusion time up to 400 s, while patients with an OSP of <49 mmHg or an ODP of <34 mmHg should undergo the procedure of occlusion time less than 300 s to prevent the occurrence of OI. Conclusion: We have demonstrated the safety and effectiveness of the Mo.Ma system during CAS in an Asia population. By reducing the occlusion time, our study indicated a lower risk of OI in the Mo.Ma system and proposed the optimal occlusion time according to occlusion pressure to prevent OI during the CAS procedure. Further large-scale and prospective studies are needed to verify our results.
RESUMO
BACKGROUND: Literature describing recovery of left ventricular (LV) function post sacubitril/valsartan treatment and the optimal management of heart failure (HF) patients receiving sacubitril/valsartan remain sparse. METHODS: We recruited 437 consecutive chronic HF patients with baseline left ventricular ejection fraction (LVEF) less than 40%, who were treated with sacubitril/valsartan. All patients underwent routine echocardiographic measurement. RESULTS: During treatment period, recovery of LVEF to 50% or greater was observed in 77 (17.6%) patients. After multivariate analysis, recovery of LV dysfunction was associated with non-ischemic etiology of HF, smaller baseline LV end-diastolic diameter (LVEDD), and higher initial dosage of sacubitril/valsartan. Compared to those without recovery of LV dysfunction, death from cardiovascular causes or first unplanned hospitalization for HF (CVD/HFH) were significantly lower in patients with LVEF recovery [11.7% vs. 24.4%, hazard ratio (HR) 0.42, p = 0.014]. Among patients with recovery of LVEF, 51 patients continued to receive the same dosage of sacubitril/valsartan had higher LVEF and were less likely to have deterioration of LVEF than the other 26 patients who received either tapering dose of sacubitril/valsartan or switching from sacubitril/valsartan to renin-angiotensin-system blockers (LVEF 56.4 ± 5.3% vs. 45.0 ± 12.8%, p < 0.001; ΔLVEF 1.2 ± 5.1% vs. -9.3 ± 12.0%, p < 0.001). CVD/HFH occurred more frequently in the taper group than the maintenance group (23.1% vs. 5.9%, HR 0.22, p = 0.035). CONCLUSIONS: Non-ischemic etiology of HF, smaller baseline LVEDD, and higher initial dosage of sacubitril/valsartan could predict better recovery of LV function. Among patients with functional recovery, tapering sacubitril/valsartan dose was associated with deterioration of recovered heart function and had less favorable prognosis during follow-up.
