Early Detection of Left Atrial Energy Loss and Mechanics Abnormalities in Diabetic Patients with Normal Left Atrial Size: A Study Combining Vector Flow Mapping and Tissue Tracking Echocardiography.

BACKGROUND: Whether left atrial (LA) functional abnormalities already exist when the LA is of normal size is unknown. The aim of this study was to explore LA energy loss and mechanics changes using vector flow mapping (VFM) and two-dimensional tissue tracking (2DTT) echocardiography in patients with diabetes and normal LA size. MATERIAL/METHODS: This study included 47 normotensive patients with diabetes and 45 controls. The following indexes were measured: LA energy loss during systole (LAELs), early diastole (LAELed), and atrial contraction (LAELac); atrial longitudinal strain during systole (SLAs), early diastole (SLAed) and late diastole (SLAac); and peak LA strain rate during systole (SRLAs), early diastole (SRLAed), and atrial contraction (SRLAac). RESULTS: The LAELs and LAELed decreased in diabetic patients compared with controls (P=0.002, P<0.01, respectively), whereas the LAELac increased in diabetic patients (P<0.001). The SLAs, SLAed, SRLAs, and SRLAed (all P<0.01) were all lower in diabetic patients than in controls. However, there was no difference in the SLAac and SRLAac between the two groups. Multivariate regression analysis showed that the LAELs, LAELac, and SRLAs were independently associated with HbA(1c) in the whole study population. CONCLUSIONS: LA energy loss and deformation mechanics are already impaired in diabetic patients with normal LA size and the long-term parameter of glycemic control was correlated with them. VFM combined with 2DTT might be a promising tool for the early detection of LA dysfunction caused by impaired glucose metabolism.

A multi-task deep learning approach for real-time view classification and quality assessment of echocardiographic images.

High-quality standard views in two-dimensional echocardiography are essential for accurate cardiovascular disease diagnosis and treatment decisions. However, the quality of echocardiographic images is highly dependent on the practitioner's experience. Ensuring timely quality control of echocardiographic images in the clinical setting remains a significant challenge. In this study, we aimed to propose new quality assessment criteria and develop a multi-task deep learning model for real-time multi-view classification and image quality assessment (six standard views and "others"). A total of 170,311 echocardiographic images collected between 2015 and 2022 were utilized to develop and evaluate the model. On the test set, the model achieved an overall classification accuracy of 97.8% (95%CI 97.7-98.0) and a mean absolute error of 6.54 (95%CI 6.43-6.66). A single-frame inference time of 2.8 ms was achieved, meeting real-time requirements. We also analyzed pre-stored images from three distinct groups of echocardiographers (junior, senior, and expert) to evaluate the clinical feasibility of the model. Our multi-task model can provide objective, reproducible, and clinically significant view quality assessment results for echocardiographic images, potentially optimizing the clinical image acquisition process and improving AI-assisted diagnosis accuracy.

An Automated Heart Shunt Recognition Pipeline Using Deep Neural Networks.

Automated recognition of heart shunts using saline contrast transthoracic echocardiography (SC-TTE) has the potential to transform clinical practice, enabling non-experts to assess heart shunt lesions. This study aims to develop a fully automated and scalable analysis pipeline for distinguishing heart shunts, utilizing a deep neural network-based framework. The pipeline consists of three steps: (1) chamber segmentation, (2) ultrasound microbubble localization, and (3) disease classification model establishment. The study's normal control group included 91 patients with intracardiac shunts, 61 patients with extracardiac shunts, and 84 asymptomatic individuals. Participants' SC-TTE images were segmented using the U-Net model to obtain cardiac chambers. The segmentation results were combined with ultrasound microbubble localization to generate multivariate time series data on microbubble counts in each chamber. A classification model was then trained using this data to distinguish between intracardiac and extracardiac shunts. The proposed framework accurately segmented heart chambers (dice coefficient = 0.92 ± 0.1) and localized microbubbles. The disease classification model achieved high accuracy, sensitivity, specificity, F1 score, kappa value, and AUC value for both intracardiac and extracardiac shunts. For intracardiac shunts, accuracy was 0.875 ± 0.008, sensitivity was 0.891 ± 0.002, specificity was 0.865 ± 0.012, F1 score was 0.836 ± 0.011, kappa value was 0.735 ± 0.017, and AUC value was 0.942 ± 0.014. For extracardiac shunts, accuracy was 0.902 ± 0.007, sensitivity was 0.763 ± 0.014, specificity was 0.966 ± 0.008, F1 score was 0.830 ± 0.012, kappa value was 0.762 ± 0.017, and AUC value was 0.916 ± 0.006. The proposed framework utilizing deep neural networks offers a fast, convenient, and accurate method for identifying intracardiac and extracardiac shunts. It aids in shunt recognition and generates valuable quantitative indices, assisting clinicians in diagnosing these conditions.

A robust local pulse wave imaging method based on digital image processing techniques.

The original diameter velocity loop method (ln(D)U-loop) cannot accurately extract the blood vessel diameter waveform when the quality of ultrasound image data is not high (such as obesity, age, and the operation of the ultrasound doctor), so it is unable to measure the pulse wave velocity (PWV) of the ascending aorta. This study proposes a diameter waveform extraction method combining threshold, gradient filtering, and the center of gravity method. At the same time, the linear regression method of searching for the rising point of the systolic period is replaced by the optimal average of two linear regression methods. This method can also extract the diameter waveform with poor-quality images and obtain a more accurate PWV. In vivo experimental data from 17 (age 60.5 ± 9.2) elderly patients with cerebral infarction and 12 (age 32.5 ± 5.6) healthy young adults were used for processing, and the results showed that the mean PWV using the ln(D)U-loop method was 12.56 (SD = 3.47) ms-1 for patients with cerebral infarction and 6.81 (SD = 1.73) ms-1 for healthy young adults. The PWV results based on the Wilcoxon rank-sum test and calculated based on the improved ln(D)U-loop method were both statistically significant (p < 0.01). The agreement analysis (Bland-Altman analysis) between the QA-loop and ln(D)U-loop methods showed that the mean deviation of the measured PWV was 0.07 m/s and the standard deviation of the deviation was 1.18 m/s. The experimental results demonstrated the effectiveness of the improved ln(D)U-loop method proposed in this paper on poor-quality images. This study can improve the possibility of the ln(D)U-loop method being widely used in the clinical measurement of ascending aortic PWV.

Abnormal left ventricular systolic reserve function detected by treadmill exercise stress echocardiography in asymptomatic type 2 diabetes.

Aims: Subclinical left ventricular (LV) dysfunction may occur in T2DM patients at the early asymptomatic stage, and LV reserve function is a sensitive index to detect subtle LV dysfunction. The purpose of our study is (1) to assess the LV reserve function using treadmill exercise stress echocardiography (ESE) in asymptomatic type 2 diabetes mellitus (T2DM) patients; (2) to explore the link of serum biological parameters and LV reserve function. Methods: This study included 84 patients with asymptomatic T2DM from September 2021 to July 2022 and 41 sex- and age-matched healthy controls during the corresponding period. All subjects completed treadmill ESE, LV systolic function-related parameters such as global longitudinal strain (GLS) and systolic strain rate (SRs), as well as diastolic function-related parameters such as E wave (E), early diastolic velocity (e'), E/e' ratio, early diastolic SR (SRe), and late diastolic SR (SRa) were compared at rest and immediately after exercise. The difference between LV functional parameters after treadmill exercise and its corresponding resting value was used to compute LV reserve function. In addition, the associations of LV reserve function and serum biological parameters were analyzed. Results: Patients with T2DM did not significantly vary from the controls in terms of alterations in LV diastolic reserve measures, the changes of LVGLS and SRs (ΔGLS: 2.19 ± 2.72% vs. 4.13 ± 2.79%, P < 0.001 and ΔSRs:0.78 ± 0.33 s-1 vs. 1.02 ± 0.28 s-1, P < 0.001) in the T2DM group were both lower than those in the control group. Glycated hemoglobin (HbA1c), N-terminal pro-brain natriuretic peptide (NTproBNP), waist circumference, and high-sensitive C-reactive protein (hsCRP) were identified as independent predictors of LV systolic reserve by stepwise multiple linear regression analysis. Conclusion: LV systolic reserve function, as measured by pre- and post-exercise differences in GLS and SRs were significantly impaired in patients with asymptomatic T2DM, whereas diastolic reserve remained normal during exercise and was comparable to that of the control group. This was different from previous findings. High levels of HbA1c, NTproBNP, hsCRP, and increasing waist circumference were independent predictors of LV systolic reserve.

Echocardiographic Normal Reference of Left Ventricular Contractile Reserve During Treadmill Exercise Stress Echocardiography in Healthy Chinese Adults - New Non-Sex-Specific Parameter for Left Ventricular Contractile Reserve Evaluation.

PURPOSE: Left ventricular (LV) contractile reserve is commonly used for LV systolic function assessment, while data on normal LV contractile reserve to exercise and the effect of gender on it are contradictory and limited, especially in Chinese adults. The aims of the present study are to clarify echocardiographic normal reference of LV contractile reserve during treadmill exercise stress echocardiography in healthy Chinese adults and to evaluate the sex-specific impact on it. PATIENTS AND METHODS: The study population consisted of 157 healthy Chinese adults. All subjects underwent comprehensive echocardiographic assessment at rest and immediately after a symptom-limited treadmill stress test. The impact of gender on LV contractile reserve was analyzed. RESULTS: The study population consisted of 157 healthy Chinese adults. All subjects underwent comprehensive echocardiographic assessment at rest and immediately after a symptom-limited treadmill stress test. The impact of gender on LV contractile reserve was analyzed. CONCLUSION: Traditional LV contractile reserve of men was much higher than that of women in a healthy Chinese population. The difference might be because of higher BSA in men. ΔGLS was less influenced by METs and CI at rest compared to ΔEF. ΔGLS, and especially the ΔGLS index, might be considered as a more preferable contractile reserve parameter for clinical cardiac function evaluation.

Left Ventricular Energy Loss Assessed by Vector Flow Mapping in Patients with Prediabetes and Type 2 Diabetes Mellitus.

The aim of this study was to assess left ventricular (LV) energy loss (EL) using vector flow mapping in patients with prediabetes (pre-DM) and type 2 diabetes mellitus (DM). Thirty pre-DM patients, 51 DM patients, and 38 controls were studied by transthoracic echocardiography. EL-total, EL-base, EL-mid and EL-apex climaxed at different phases. Compared with controls, pre-DM and DM patients showed increased EL-total during slow ejection, isovolumic relaxation, rapid filling and slow filling (p < 0.05). Similarly, EL-base, EL-mid and EL-apex increased during certain phases. Stepwise multiple regression analysis revealed that the early transmitral valve blood flow velocity E, the late transmitral valve blood flow velocity A, the ratio of E/A, LV peak torsion, diastolic untwisting velocity, vortex circulation and area were independently associated with EL during different phases (all p < 0.05). Our study suggests that LV EL is increased during diastole and certain phases of systole in DM patients compared with controls. The changes in LV vortex and deformation mechanics were correlated with EL.