GAN

Improving realism in abdominal ultrasound simulation combining a segmentation-guided loss and polar coordinates training
We propose a GAN that improves the realism of US simulations by combining a segmentation-guided loss function with polar coordinates training, leading to more realistic organ boundaries and reduced artifacts.
Santiago Vitale, Alejandro Díaz, José Ignacio Orlando, Emmanuel Iarussi, Ignacio Larrabide
PDF Cite Project DOI
Improving realism in abdominal ultrasound simulation combining a segmentation-guided loss and polar coordinates training