Automatic cardiac evaluations using a deep video object segmentation network

Abstract

Background Accurate cardiac volume and function assessment have valuable and significant diagnostic implications for patients suffering from ventricular dysfunction and cardiovascular disease. This study has focused on finding a reliable assistant to help physicians have more reliable and accurate cardiac measurements using a deep neural network. EchoRCNN is a semi-automated neural network for echocardiography sequence segmentation using a combination of mask region-based convolutional neural network image segmentation structure with reference-guided mask propagation video object segmentation network. Results The proposed method accurately segments the left and right ventricle regions in four-chamber view echocardiography series with a dice similarity coefficient of 94.03% and 94.97%, respectively. Further post-processing procedures on the segmented left and right ventricle regions resulted in a mean absolute error of 3.13% and 2.03% for ejection fraction and fractional area change parameters, respectively. Conclusion This study has achieved excellent performance on the left and right ventricle segmentation, leading to more accurate estimations of vital cardiac parameters such as ejection fraction and fractional area change parameters in the left and right ventricle functionalities, respectively. The results represent that our method can predict an assured, accurate, and reliable cardiac function diagnosis in clinical screenings.

Semi-supervised Learning for Real-time Segmentation of Ultrasound Video Objects: A Review

“…As different methods have varying optimization approaches and hardware, and different models focus on different datasets, this information is provided for reference only. In terms of real-time performance, several models [48,50,57,60,62,64,71] have achieved this. Notably, Wang et al [57] and El Rai et al [62] do not mention specific real-time metrics in their papers.…”

“…Sirjani et al [64] developed EchoRCNN, a video object segmentation network for echocardiograms, aimed at extracting cardiac features from echocardiogram sequences. The network is built upon the robust image segmentation architectures of Mask R-CNN, RetinaNet, and the RGMP video object segmentation network.…”

Predicting Cardiac Issues From Echocardiograms: A Literature Review Using Deep Learning and Machine Learning Techniques

“…To segment echocardiogram sequences, Sirjani et al [31] created EchoRCNN, a semi-automated neural network that combines a CNN cardiac image segmentation structure based on mask regions with a reference-guided mask propagation video object segmentation network. The network learns to distinguish between ventricles from ultrasound cardiac data.…”

Ejection Fraction Estimation from Echocardiograms Using Optimal Left Ventricle Feature Extraction Based on Clinical Methods

“…These techniques are computationally efficient, but they have a high signal-to-noise ratio and fail to produce acceptable results when there are unclear borders and non-uniform regional intensities. In ML approaches for image segmentation, an image is split into distinct regions by giving each pixel a label of its associated class, as in [ 5 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ]. Temporal coherence, which is found in heart motion between the frames, is also incorporated along with CNNs by [ 23 ] for segmentation.…”