Search

Aims: Left ventricular ejection fraction (LVEF) calculation by echocardiography is pivotal in evaluating cancer patients' cardiac function. Artificial intelligence (AI) can facilitate the acquisition of optimal images and automated LVEF (autoEF) calculation. We sought to evaluate the feasibility and accuracy of LVEF calculation by oncology staff using an AI-enabled handheld ultrasound device (HUD)., Methods and Results: We studied 115 patients referred for echocardiographic LVEF estimation. All patients were scanned by a cardiologist using standard echocardiography (SE), and biplane Simpson's LVEF was the reference standard. Hands-on training using the Kosmos HUD was provided to the oncology staff before the study. Each patient was scanned by a cardiologist, a senior oncologist, an oncology resident, and a nurse using the TRIO AI and KOSMOS EF deep learning algorithms to obtain autoEF. The correlation between autoEF and SE-ejection fraction (EF) was excellent for the cardiologist ( r = 0.90), the junior oncologist ( r = 0.82), and the nurse ( r = 0.84), and good for the senior oncologist ( r = 0.79). The Bland-Altman analysis showed a small underestimation by autoEF compared with SE-EF. Detection of impaired LVEF < 50% was feasible with a sensitivity of 95% and specificity of 94% for the cardiologist; sensitivity of 86% and specificity of 93% for the senior oncologist; sensitivity of 95% and specificity of 91% for the junior oncologist; and sensitivity of 94% and specificity of 87% for the nurse., Conclusion: Automated LVEF calculation by oncology staff was feasible using AI-enabled HUD in a selected patient population. Detection of LVEF < 50% was possible with good accuracy. These findings show the potential to expedite the clinical workflow of cancer patients and speed up a referral when necessary., Competing Interests: Conflict of interest: This study was performed with a loan of handheld ultrasound devices and technical support by Echonous Inc., (© The Author(s) 2024. Published by Oxford University Press on behalf of the European Society of Cardiology.)

Aims: We sought to evaluate the reliability and diagnostic accuracy of a novel handheld ultrasound device (HUD) with artificial intelligence (AI) assisted algorithm to automatically calculate ejection fraction (autoEF) in a real-world patient population., Methods and Results: We studied 100 consecutive patients (57 ± 15 years old, 61% male), including 38 with abnormal left ventricular (LV) function [LV ejection fraction (LVEF) < 50%]. The autoEF results acquired using the HUD were independently compared with manually traced biplane Simpson's rule measurements on cart-based systems to assess method agreement using intra-class correlation coefficient (ICC), linear regression analysis, and Bland-Altman analysis. The diagnostic accuracy for the detection of LVEF <50% was also calculated. Test-retest reliability of measured EF by the HUD was assessed by calculating the ICC and the minimal detectable change (MDC). The ICC, linear regression analysis, and Bland-Altman analysis revealed good agreement between autoEF and reference manual EF (ICC = 0.85; r  = 0.87, P  < 0.001; mean bias -1.42% with limits of agreement 14.5%, respectively). Detection of abnormal LV function (EF < 50%) by autoEF algorithm was feasible with sensitivity 90% (95% CI 75-97%), specificity 87% (95% CI 76-94%), PPV 81% (95% CI 66-91%), NPV 93% (95% CI 83-98%), and a total diagnostic accuracy of 88%. Test-retest reliability was excellent (ICC = 0.91, P  < 0.001; r  = 0.91, P  < 0.001; mean difference ± SD: 0.54% ± 5.27%, P  = 0.308) and MDC for LVEF measurement by autoEF was calculated at 4.38%., Conclusion: Use of a novel HUD with AI-enabled capabilities provided similar LVEF results with those derived by manual biplane Simpson's method on cart-based systems and shows clinical potential., (© The Author(s) 2022. Published by Oxford University Press on behalf of the European Society of Cardiology.)

Aim: To investigate the efficacy of low-density lipoprotein (LDL) transport simulation in reconstructed arteries derived from computed tomography coronary angiography (CTCA) to predict coronary segments that are prone to progress., Methods and Results: Thirty-two patients admitted with an acute coronary event who underwent 64-slice CTCA after percutaneous coronary intervention and at 3-year follow-up were included in the analysis. The CTCA data were used to reconstruct the coronary anatomy of the untreated vessels at baseline and follow-up, and LDL transport simulation was performed in the baseline models. The computed endothelial shear stress (ESS), LDL concentration, and CTCA-derived plaque characteristics were used to identify predictors of substantial disease progression (defined as an increase in the plaque burden at follow-up higher than two standard deviations of the intra-observer variability of the expert who performed the analysis). Fifty-eight vessels were analysed. High LDL concentration [odds ratio (OR): 2.16; 95% confidence interval (CI): 1.64-2.84; P = 0.0054], plaque burden (OR: 1.40; 95% CI: 1.13-1.72; P = 0.0017), and plaque area (OR: 3.46; 95% CI: 2.20-5.44; P≤ 0.0001) were independent predictors of a substantial disease progression at follow-up. The ESS appears as a predictor of disease progression in univariate analysis but was not an independent predictor when the LDL concentration was entered into the multivariate model. The accuracy of the model that included the LDL concentration was higher than the accuracy of the model that included the ESS (65.1 vs. 62.5%)., Conclusions: LDL transport modelling appears a better predictor of atherosclerotic disease progression than the ESS, and combined with the atheroma characteristics provided by CTCA is able to detect with a moderate accuracy segments that will exhibit a significant plaque burden increase at mid-term follow-up., (Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2016. For permissions please email: journals.permissions@oup.com.)

Aims: The maintenance of sinus rhythm is crucial for the functional capacity of patients with hypertrophic cardiomyopathy (HCM). Using a multimodality approach, we attempted to identify potential predictors of paroxysmal atrial fibrillation (PAF) in HCM patients., Methods and Results: Thirty HCM patients (17 males, mean age 57.9 ± 13.6) with at least one documented PAF episode and 32 age- and sex-matched HCM control patients as well as 25 healthy volunteers were studied in sinus rhythm. Study subjects underwent 2D echocardiography including a colour Doppler myocardial imaging evaluation of the left atrium (LA). Additionally, an orthogonal electrocardiogram was acquired; P-wave duration, maximum, and mean energies were calculated for each subject at each orthogonal lead and the composite vector axis using the Morlet wavelet analysis. Compared with HCM controls, in HCM-PAF patients, LA antero-posterior diameter was significantly enlarged (LADAP: 46.1 ± 5.9 vs. 40.0 ± 4.7 mm, P < 0.001), peak strain rate of the LA lateral wall in the reservoir phase was significantly decreased (LAT peak SR-S: 1.93 ± 0.51 vs. 2.55 ± 0.83 s(-1), P < 0.01), and P-wave duration in the Z-lead was significantly prolonged (P-durZ: 106.9 ± 24.6 vs. 86.2 ± 14.3 ms, P < 0.001). Cut-off values and areas under the curve (AUCs) for individual parameters were 42.0 mm, 2.32 s(-1), and 98.8 ms and 0.81, 0.74, and 0.78, respectively. A multivariable model combining LADAP, LAT peak SR-S and P-durZ had an AUC of 0.90, a sensitivity of 0.87, and a specificity of 0.91 for identifying PAF patients., Conclusion: P-wave duration combined with LA antero-posterior diameter and myocardial deformation indices resulted in a higher power for discriminating HCM-PAF patients, when compared with individual parameters derived from either wavelet analysis or 2D echocardiography.

Aims: We evaluated the distribution and composition of atherosclerotic plaques at bifurcations with intravascular ultrasound-virtual histology (IVUS-VH) and multidetector computed tomography (MDCT) in relation to the bifurcation angle (BA)., Methods and Results: In 33 patients (age 63±11 years, 79% male) imaged with IVUS-VH and MDCT, 33 bifurcations were matched and studied. The analysed main vessel was divided into a 5 mm proximal segment, the in-bifurcation segment, and a 5 mm distal segment. Plaque contours were manually traced on MDCT and IVUS-VH. Plaques with >10% confluent necrotic core and <10% dense calcium on IVUS-VH were considered high risk, whereas plaque composition by MDCT was graded as non-calcified, calcified, or mixed. The maximum BA between the main vessel and the side branch was measured on diastolic MDCT data sets. Overall the mean plaque area decreased from the proximal to the distal segment [8.5±2.8 vs. 6.0±3.0 mm2 (P<0.001) by IVUS-VH and 9.0±2.6 vs. 6.5±2.5 mm2 (P<0.001) by MDCT]. Similarly, the necrotic core area was higher in the proximal compared with the distal segment (1.12±0.7 vs. 0.71±0.7 mm2, P=0.001). The proximal segment had the higher percentage of high-risk plaques (13/25, 52%), followed by the in-bifurcation (6/25, 24%), and the distal segment (6/25, 24%); these plaques were characterized by MDCT as non-calcified (72%) or mixed (28%). The presence of high-risk and non-calcified plaques in the proximal segment was associated with higher BA values (71±19° vs. 55±19°, P=0.028 and 74±20° vs. 50±14°, P=0.001, respectively)., Conclusion: The proximal segment of bifurcations is more likely to contain high-risk plaques, especially when the branching angle is wide.