This model is trained on the CAMUS dataset: S. Leclerc, E. Smistad, J. Pedrosa, A. Ostvik, et al. "Deep Learning for Segmentation using an Open Large-Scale Dataset in 2D Echocardiography" in IEEE Transactions on Medical Imaging, vol. 38, no. 9, pp. 2198-2210, Sept. 2019. doi: 10.1109/TMI.2019.2900516
The model is the .onnx version of the U-Net1 model as described in the work mentioned above. It is trained on the first split of the CAMUS dataset.
The model can be directly used in the framework provided at https://github.com/gillesvntnu/GCN_UNET_agreement_demo.git Unlike the nnU-Net model, the U-Net 1 model requires the input to be scaled to [0,1]
Information on the splits of the CAMUS dataset can be found at: https://github.com/gillesvntnu/GCN_multistructure/tree/main/files/subgroups_CAMUS
For the corresponding version in .pb format, see https://huggingface.co/gillesvdv/unet1_camus_cv1