Code for the paper: https://arxiv.org/pdf/1912.03120.pdf
A Study into Echocardiography View Conversion, accepted to MedNeurIPS 2019.
To initiate training, run
python3 src/main.py --dataset_path=$DATASETS/CAMUS --config=configs/config_2CH_4CH.json
The environment variable $DATASET
is assumed to be set to
where the CAMUS dataset directory is stored.
-
Unified all modes of the codes. Now segmentation, patchGAN, and constrained_patchgan can be selected in the .json config file. Please refer to the samples in the "sample_configs" folder.
-
Bug for discriminator fixed. Now the input and target are not concatenated in the discriminator.
Abstract:
Transthoracic echo is one of the most common means of cardiac studies in the clinical routines. During the echo exam, the sonographer captures a set of standardcross sections (echo views) of the heart. Each 2D echo view cuts through the 3D cardiac geometry via a unique plane. Consequently, different views share some limited information. In this work, we investigate the feasibility of generating a 2D echo view using another view based on adversarial generative models. The objective optimized to train the view-conversion model is based on the ideas introduced by LSGAN, PatchGAN and Conditional GAN (cGAN). The size and length of the left ventricle in the generated target echo view is compared against that of the target ground-truth to assess the validity of the echo view conversion. Results show that there is a correlation of 0.50 between the LV areas and 0.49 between the LV lengths of the generated target frames and the real target frames.
Sample Results: