scholarly journals Diagnostic utility of artificial intelligence for left ventricular scar identification using cardiac magnetic resonance imaging—A systematic review

2021 ◽  
Author(s):  
Nikesh Jathanna ◽  
Anna Podlasek ◽  
Albert Sokol ◽  
Dorothee Auer ◽  
Xin Chen ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Artificial Intelligence ◽  
Systematic Review ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Cardiac Magnetic Resonance Imaging ◽  
Left Ventricular ◽  
Diagnostic Utility ◽  
Resonance Imaging

Abstract MP31: Left Ventricular Hypertrophy by Electrocardiogram versus Cardiac Magnetic Resonance Imaging as a Predictor for Heart Failure: The MESA Study

Circulation ◽  
2016 ◽  
Vol 133 (suppl_1) ◽  
Author(s):  
Abdullahi O Oseni ◽  
Waqas T Qureshi ◽  
Mohammed F Almahmoud ◽  
Alain Bertoni ◽  
David A Bluemke ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Heart Failure ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Left Ventricular Hypertrophy ◽  
Cardiac Magnetic Resonance Imaging ◽  
Ventricular Hypertrophy ◽  
Predictive Ability ◽  
Left Ventricular ◽  
Resonance Imaging

Background: Left ventricular hypertrophy (LVH) is an established risk factor for heart failure (HF). However, it is unknown whether LVH detected by electrocardiogram (ECG-LVH) is equivalent to LVH ascertained by cardiac magnetic resonance imaging (MRI-LVH) in terms of prediction of incident HF using risk prediction models like the Framingham Heart Failure Risk Score (FHFRS). Methods: This analysis included 4745 (mean age 61+10 years, 53.5% women, 61.7% non-whites) from the Multi-Ethnic Study of Atherosclerosis who were free of cardiovascular disease at the time of enrollment. ECG-LVH was defined using Cornell’s criteria while MRI-LVH was derived from left ventricular (LV) mass measured by cardiac MRI. Cox proportional hazard regression was used to examine the association between ECG-LVH and MRI-LVH with incident HF. Harrell’s concordance C-index was used to estimate the predictive ability of the FHFRS when either ECG-LVH or MRI-LVH were included as one of its components. The added predictive ability of ECG-LVH and MRI-LVH were investigated using integrated discrimination improvement (IDI) index and relative IDI. Results: ECG-LVH was present in 291(6.1%) while MRI-LVH was present in 499 (10.5%) of the participants. Over a median follow up of 10.4 years, 140 participants developed HF. Both ECG-LVH [HR (95% CI): 2.25(1.38-3.69)] and MRI-LVH [HR (95% CI): 3.80(1.56-5.63)] were associated with an increased risk of HF in multivariable adjusted models (Table 1). The ability of FHFRS to predict HF was improved with MRI-LVH (C-index 0.871, 95% CI: 0.842-0.899) when compared with ECG-LVH (C-index 0.860, 95% CI: 0.833-0.888) (p < 0.0001). To assess the potential clinical utility of using LVH-MRI instead of ECG-LVH, we calculated several measures of reclassification (Table 1), which were consistent with the statistically significantly improved C-statistic with MRI-LVH. Conclusion: Both ECG-LVH and MRI-LVH are predictive of HF when used in the FHFRS. Substituting MRI-LVH for ECG-LVH improves the predictive ability of the FHFRS.

Abstract 9634: As Compared With Cardiac Magnetic Resonance Imaging, Transthoracic Echocardiography Has an Excellent Accuracy in Diagnosing a Left Ventricular Thrombus After an Anterior Myocardial Infarction Complicated With Left Ventricular Dysfunction: Results of a Prospective Multicentric Study

Circulation ◽  
2014 ◽  
Vol 130 (suppl_2) ◽  
Author(s):  
Philippe Meurin ◽  
Virginie Brandao Carreira ◽  
Raphaelle D Dumaine ◽  
Alain Shqueir ◽  
Olivier Milleron ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Myocardial Infarction ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Cardiac Magnetic Resonance Imaging ◽  
Antiplatelet Therapy ◽  
Transthoracic Echocardiography ◽  
Dual Antiplatelet Therapy ◽  
Left Ventricular ◽  
Resonance Imaging

Introduction: The generalization of reperfusion techniques to treat acute myocardial infarction (MI) has allowed for markedly reduced incidence in left ventricular (LV) thrombi because of the reduced myocardial damage. LV thrombi are estimated to complicate 5% to 10% of unselected anterior-wall MI (Ant-MI). However, the incidence and evolution of LV thrombi in high-risk patients with Ant-MI complicated by LV systolic dysfunction is not well known. Cardiac magnetic resonance imaging with contrast delayed enhancement (CMR-DE) is the gold standard in assessing LV thrombus, but comparisons of transthoracic echocardiography (TTE) and CMR-DE are scarce. Hypothesis: We assessed whether LV thrombi are still frequent after major Ant-MI, despite systematic dual antiplatelet therapy, and whether focused TTE has a good accuracy for detection as compared with CMR-DE. Methods: From 2011 to 2013, from 7 centers, we prospectively included patients with LV ejection fraction (LVEF) < 45% at a first TTE performed < 7 days after Ant-MI. A second evaluation including TTE and CMR-DE (analyzed by blinded examiners) was performed at 30 days. A third TTE and assessment of clinical status and adverse events were performed between months 6 and 12. Results: We included 100 consecutive patients (71% males; mean age 59.1 ± 12.1 years; LVEF 33.5 ± 6.0%) at a mean of 4.8 ± 1.9 days after Ant-MI; 88% had undergone primary coronary angioplasty. In total, 26 patients had LV thrombi detected at a mean of 23.2 ± 34.8 days after MI (6 during the first week after the MI, 16 from days 8 to 30, 4 after day 30). As compared with CMR-DE, TTE sensitivity and specificity were 94.7% and 98.5%, respectively. For 24 patients (92.3%), the LV thrombi disappeared with triple antithrombotic therapy including dual antiplatelet therapy and a vitamin K antagonist. One patient died from a recurrent subdural haematoma and another had a peripheral embolism. Conclusions: In this prospective multicenter study, LV thrombus occurred in 26% of patients after Ant-MI complicated by LV dysfunction. Focused TTE has a high accuracy for detection. CMR-DE should be performed only when the apex is not clearly seen.

Sign in / Sign up

Export Citation Format

Share Document