Generative adversarial networks (GAN) are among the most exciting deep-learning breakthroughs of the past decade. GANs use adversarial discriminators to supervise the learning of generative neural networks, without explicitly modelling the pattern of the training samples. They have demonstrated superior performance, especially in medical data synthesis, and quickly evolved to become the state-of-the-art for data-generation tasks.
In cardiovascular medicine, GANs are increasingly adopted in a wide range of applications in cardiovascular imaging, electrocardiography signals, and patient characteristics. This topic focuses on novel GAN approaches and applications to various cardiovascular research, including but not limited to domain transfer, domain adaption, dose reduction, missing modality, data augmentation, image reconstruction, synthesis, segmentation, detection, and classification. It also aims to address the technical challenges of GANs in model training, validation, transparency, and robustness.
We welcome researchers with a background in machine learning, basic science, clinical cardiology, or cross-disciplinary field to contribute high-quality papers in methodology, original investigations, clinical applications, ethics, reviews, and mini-reviews.
Generative adversarial networks (GAN) are among the most exciting deep-learning breakthroughs of the past decade. GANs use adversarial discriminators to supervise the learning of generative neural networks, without explicitly modelling the pattern of the training samples. They have demonstrated superior performance, especially in medical data synthesis, and quickly evolved to become the state-of-the-art for data-generation tasks.
In cardiovascular medicine, GANs are increasingly adopted in a wide range of applications in cardiovascular imaging, electrocardiography signals, and patient characteristics. This topic focuses on novel GAN approaches and applications to various cardiovascular research, including but not limited to domain transfer, domain adaption, dose reduction, missing modality, data augmentation, image reconstruction, synthesis, segmentation, detection, and classification. It also aims to address the technical challenges of GANs in model training, validation, transparency, and robustness.
We welcome researchers with a background in machine learning, basic science, clinical cardiology, or cross-disciplinary field to contribute high-quality papers in methodology, original investigations, clinical applications, ethics, reviews, and mini-reviews.
