Multi-task learning Y-Net

Code for Sobotka, Daniel, et al. "Improving vessel segmentation with multi-task learning and auxiliary data available only during model training." Computerized Medical Imaging and Graphics 114 (2024): 102369. https://doi.org/10.1016/j.compmedimag.2024.102369

As baseline the 3D-UNet from https://github.com/wolny/pytorch-3dunet was used.