Explainable deep learning for medical image segmentation

This work deals with explainability and trust in deep learning for medical image segmentation, specifically targeting the CAMUS dataset. The provided code yields excellent segmentation results, demonstrating the effectiveness of our models. Currently, the code is restricted for access and provided here solely for citation purposes. 

Requirements:

CAMUS Dataset: You need to download the "database_nifti" folder from the CAMUS dataset and place it in the same directory as the code repository.

NIfTI Toolbox: To read the .nii.gz files, you must download the NIfTI_20140122 toolbox and place it in the same repository.

These resources are essential to properly run and process the segmentation tasks.

T. Berghout, "Compressed Sensing and Certainty-Enabled Deep Learning in Cardiac Ultrasound Segmentation," 2025 22nd International Conference on Electrical Engineering, Computing Science and Automatic Control (CCE), Mexico City, Mexico, 2025, pp. 1-6, doi: 10.1109/CCE67728.2025.11271926.

T. Berghout, "Trustworthiness and Interpretability in Deep Learning for Cardiac Ultrasound," 2025 22nd International Conference on Electrical Engineering, Computing Science and Automatic Control (CCE), Mexico City, Mexico, 2025, pp. 1-6, doi: 10.1109/CCE67728.2025.11271922.