GAN-enhanced Conditional Echocardiogram Generation
This work addresses label scarcity in medical imaging for echocardiography, enabling semi-supervised training of models, but it is incremental as it builds on existing GAN and conditional generation techniques.
The paper tackled the problem of label scarcity in echocardiogram analysis by proposing a GAN-enhanced method to generate echocardiogram frames conditioned on segmentation masks, resulting in high-quality frames that match given masks for structures like the left ventricle.
Echocardiography (echo) is a common means of evaluating cardiac conditions. Due to the label scarcity, semi-supervised paradigms in automated echo analysis are getting traction. One of the most sought-after problems in echo is the segmentation of cardiac structures (e.g. chambers). Accordingly, we propose an echocardiogram generation approach using generative adversarial networks with a conditional patch-based discriminator. In this work, we validate the feasibility of GAN-enhanced echo generation with different conditions (segmentation masks), namely, the left ventricle, ventricular myocardium, and atrium. Results show that the proposed adversarial algorithm can generate high-quality echo frames whose cardiac structures match the given segmentation masks. This method is expected to facilitate the training of other machine learning models in a semi-supervised fashion as suggested in similar researches.