Key result
CNN and Bidirectional LSTM framework detects end-diastole and end-systole within ~1 frame.
Why the study?
Detecting end-diastole and end-systole frames in echocardiography is challenging, and the largest open-access dataset, EchoNet-Dynamic, shows poor phase detection likely due to noise.
Does a customized CNN-BLSTM architecture with preprocessing improve the accuracy of end-diastole and end-systole frame detection in 2D echocardiography?
Does a customized CNN-BLSTM architecture with preprocessing improve the accuracy of end-diastole and end-systole frame detection in 2D echocardiography?
Answers are grounded in this paper and cite their sources.
A customized CNN-BLSTM deep learning framework with specific preprocessing steps accurately detects end-diastole and end-systole frames in 2D echocardiograms, outperforming previous models.
The first and critical step of accurate heart function analysis (i.e., ventricles volume, Ejection Fraction, and stroke volume measurement) in echocardiography is the detection of end-diastole (ED) and end-systole (ES) frames. Detecting these frames is challenging due to variations in cardiac structure, heart rate associated with clinical conditions, and the low-resolution nature of echo sequences. Several deep learning techniques have recently emerged, primarily combining convolutional neural networks (CNN) and recurrent neural networks (RNN). These models were trained on a range of datasets, including open-access sources like CAMUS, PACS, and EchoNet-Dynamic, as well as private datasets. The largest open-access dataset, EchoNet Dynamic, has poor echo phase detection compared to other datasets. This suggests that the dataset has noise, which is removed in this study using three preprocessing steps: Noise reduction using heart rate formulation, Video frame synchronization, and Non-oscillating Mean absolute frame difference (MAFD). Additionally, this study formulates the echo phase detection problem as frame-level binary classification, distinguishing between diastole and systole phases. The proposed architecture takes the echo sequence as an input and a customized time-distributed CNN extracts spatial features from each frame. These frame features are fed into a bidirectional long short-term memory (BLSTM) network, capturing temporal information, followed by a classification layer. The model is trained on the pre-processed EchoNet-Dynamic dataset. Compared to true labels, an average absolute frame distance (aaFD) of 1.02 and 1.04 frames is achieved for ED and ES frames, respectively. Moreover, it operates with an inference time of less than 65 ms for an input sequence consisting of 32 frames on a graphics processing unit (GPU).
No takes yet. Share an insight, caveat, or question.
Singh et al. (2023) studied Cardiac Phase Detection in 2D Echocardiography (n=10,030). Time-Distributed CNN and Bidirectional LSTM framework vs. Other deep learning models was evaluated on Average absolute frame distance (aaFD) for end-diastole (ED) and end-systole (ES) frames. The proposed Time-Distributed CNN and Bidirectional LSTM framework achieved an average absolute frame distance of 1.02 and 1.04 frames for end-diastole and end-systole detection, respectively.
Synapse has enriched 10 related papers. Consider them for comparative context: