Co-learning of appearance and shape for precise ejection fraction estimation from echocardiographic sequences

Wei, Hongrong; Ma, Junqiang; Zhou, Yongjin; Xue, Wufeng; Ni, Dong

doi:10.1016/j.media.2022.102686

Cited by 12 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wei et al introduced a novel multi-task semi-supervised framework in their recent study [20]. This framework integrates a collaborative learning mechanism and introduces two auxiliary tasks: view classification and EF regression.…”

Section: Multi-task Learning Methodsmentioning

confidence: 99%

An AI-based multitask framework for cardiac function assessment through echocardiograms

Guo,

Mo,

et al. 2024

Fifteenth International Conference on Graphics and Image Processing (ICGIP 2023)

View full text Add to dashboard Cite

The accurate assessment of cardiac function is crucial for preventing and controlling cardiovascular diseases and reducing global mortality rates. In recent years, the rapid advancement of machine learning and deep learning, particularly the utilization of artificial intelligence technologies such as convolutional neural networks and multi-task learning, has significantly improved the objectivity and precision of assessing echocardiogram images. However, existing methods lack a thorough exploration of the intrinsic relationship between ejection fraction (EF), end-diastolic volume (EDV), and end-systolic volume (ESV) calculations, which ultimately influence the accuracy of cardiac function assessment. Therefore, we propose an AI-Based Multi-task Framework for Cardiac Function Assessment through echocardiograms. The framework utilizes a 3DCNN network to concurrently extract spatial and temporal features from echocardiographic videos. It employs multi-task learning by assigning varying weights to sub-tasks, enhancing the prediction accuracy of ejection fraction through joint training. Experimental results on the publicly available Echonet-Dynamic dataset demonstrate that the proposed framework achieves promising performance in ejection fraction prediction, with mean absolute error, root mean square error, and R² scores of 3.89%, 5.13%, and 0.82, respectively, surpassing other comparative methods. This framework will further aid clinicians in more accurate cardiac function assessment, offering promising prospects for its practical application.

show abstract

Section: Multi-task Learning Methodsmentioning

confidence: 99%

An AI-based multitask framework for cardiac function assessment through echocardiograms

Guo,

Mo,

et al. 2024

Fifteenth International Conference on Graphics and Image Processing (ICGIP 2023)

View full text Add to dashboard Cite

show abstract

“…Attention mechanism [4], [17], [63]- [67], transformer [7], [16], multi-scale mechanism [17], [68]- [70] are hypothesized to reduce speckle noise effect. Boundary/residual correction and refinement [6], [17], [71] is popular in breast ultrasound segmentation, while motion-enhanced representation [72], [73] and multi-analysis task-aware learning are popular in echocardiography segmentation [66], [67], [72], [73]. Xue et al [73] 2D+t echocardiography motion information is not leveraged or is only utilized as an auxiliary task.…”

Section: A Deep Learning Modelmentioning

confidence: 99%

“…Holistic segmentation building blocks, including attention mechanism [4], [17], [63]- [67], [7], [16], multi-scale mechanism [17], [68] are the most popular strategies. Moreover, boundary correction and refinement [6], [17], [71] is popular in breast ultrasound segmentation, while motion enhanced representation [72], [73] and multi-task learning (chamber-view classification, quantification, uncertainty estimation) [66], [67], [72], [73] is popular in echocardiography segmentation.…”

Section: Summarizationmentioning

confidence: 99%

Is denoising necessary for ultrasound image segmentation deep learning: review and benchmark

Liu,

Dong,

Qin

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Additionally, based on these two datasets, El Rai et al ( 12 ) presented a new semi-supervised approach called GraphECV for the segmentation of the LV in echocardiography by using graph signal processing, respectively resulting in Dice coefficients of 0.936 and 0.940 with 1/2 labeled data for the left ventricular segmentation. Wei et al ( 13 ) used a co-learning mechanism to explore the mutual benefits of cardiac segmentation, therefore alleviating the noisy appearance. It was validated on the training set of CAMUS dataset using 10-fold cross-validation, achieving a Dice of 0.923, 0.948 and 0.895 for the segmentation of LV, myocardium and left atrium (LA).…”

Section: Introductionmentioning

confidence: 99%

An improved contrastive learning network for semi-supervised multi-structure segmentation in echocardiography

Guo,

Zhang,

Qiu

et al. 2023

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

Cardiac diseases have high mortality rates and are a significant threat to human health. Echocardiography is a commonly used imaging technique to diagnose cardiac diseases because of its portability, non-invasiveness and low cost. Precise segmentation of basic cardiac structures is crucial for cardiologists to efficiently diagnose cardiac diseases, but this task is challenging due to several reasons, such as: (1) low image contrast, (2) incomplete structures of cardiac, and (3) unclear border between the ventricle and the atrium in some echocardiographic images. In this paper, we applied contrastive learning strategy and proposed a semi-supervised method for echocardiographic images segmentation. This proposed method solved the above challenges effectively and made use of unlabeled data to achieve a great performance, which could help doctors improve the accuracy of CVD diagnosis and screening. We evaluated this method on a public dataset (CAMUS), achieving mean Dice Similarity Coefficient (DSC) of 0.898, 0.911, 0.916 with 1/4, 1/2 and full labeled data on two-chamber (2CH) echocardiography images, and of 0.903, 0.921, 0.928 with 1/4, 1/2 and full labeled data on four-chamber (4CH) echocardiography images. Compared with other existing methods, the proposed method had fewer parameters and better performance. The code and models are available at https://github.com/gpgzy/CL-Cardiac-segmentation.

show abstract

Co-learning of appearance and shape for precise ejection fraction estimation from echocardiographic sequences

Cited by 12 publications

References 27 publications

An AI-based multitask framework for cardiac function assessment through echocardiograms

An AI-based multitask framework for cardiac function assessment through echocardiograms

Is denoising necessary for ultrasound image segmentation deep learning: review and benchmark

An improved contrastive learning network for semi-supervised multi-structure segmentation in echocardiography

Contact Info

Product

Resources

About