Key result
A deep learning framework combining a residual convolutional neural network with Bi-CLSTM achieved an overall diagnostic accuracy of 94% for automated cardiac disease classification on the ACDC testing dataset.
Why the study?
Most current deep learning segmentation methods are not fully applicable to CMR due to loss of spatial structure information or large slice gaps.
A novel deep learning architecture combining residual CNNs and Bi-CLSTM improves cardiac MRI segmentation and automated disease diagnosis.
Hypothesis-generating for automated CMR diagnosis; prospective validation required before clinical adoption.
Deep learning (DL) has been widely used in biomedical image segmentation and automatic disease diagnosis, leading to state-of-the-art performance. However, automated cardiac disease diagnosis heavily relies on cardiac segmentation maps from cardiac magnetic resonance (CMR), most current DL segmentation methods, such as 2D convolution on planes, 3D convolution, are not fully applicable to CMR due to loss of spatial structure information or large gap between slices. To make better exploit spatial aspects of the CMR data to improve cardiac segmentation accuracy, we propose a new DL segmentation structure, which consists of a residual convolution neural network for compressing the intra-slice information, and a bidirectional-convolutional long short term memory (Bi-CLSTM) for leveraging the inter-slice contexts. Moreover, automatic disease diagnosis has been conducted using the segmentation maps. Experimental results of the automatic cardiac diagnosis challenge (ACDC) show that our cardiac segmentation structure and disease diagnosis methods have achieved promising results and it can be widely extended to computer-aided diagnosis.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2020) studied Cardiac disease (Myocardial Infarction, Dilated Cardiomyopathy, Hypertrophic Cardiomyopathy, Abnormal Right Ventricle) (n=150). Residual Convolutional Neural Network with Bi-CLSTM vs. Other deep learning segmentation methods (e.g., U-Net) was evaluated on Overall accuracy of automated cardiac disease diagnosis. A deep learning framework combining a residual convolutional neural network with Bi-CLSTM achieved an overall diagnostic accuracy of 94% for automated cardiac disease classification on the ACDC testing dataset.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: