Adversarial Pseudo Healthy Synthesis Needs Pathology Factorization
This work addresses the problem of generating subject-specific healthy images for tasks like anomaly detection and data augmentation in medical imaging, but it appears incremental as it builds on existing adversarial methods.
The paper tackled pseudo healthy synthesis from pathological images by treating it as a factor decomposition problem to separate healthy and disease components, and showed in experiments on ISLES and BraTS datasets that their method outperforms conditional GAN and CycleGAN.
Pseudo healthy synthesis, i.e. the creation of a subject-specific `healthy' image from a pathological one, could be helpful in tasks such as anomaly detection, understanding changes induced by pathology and disease or even as data augmentation. We treat this task as a factor decomposition problem: we aim to separate what appears to be healthy and where disease is (as a map). The two factors are then recombined (by a network) to reconstruct the input disease image. We train our models in an adversarial way using either paired or unpaired settings, where we pair disease images and maps (as segmentation masks) when available. We quantitatively evaluate the quality of pseudo healthy images. We show in a series of experiments, performed in ISLES and BraTS datasets, that our method is better than conditional GAN and CycleGAN, highlighting challenges in using adversarial methods in the image translation task of pseudo healthy image generation.