Supervised learning-based segmentation methods typically require a large number of
annotated training data to generalize well at test time. In medical applications, curating such …

Supervised deep learning methods for segmentation require large amounts of labelled
training data, without which they are prone to overfitting, not generalizing well to unseen …

Recently, it has been shown that CycleGANs are masters of steganography. They cannot
only learn reliable map**s between two distributions without calling for paired training …

Pseudo-healthy synthesis is the task of creating a subject-specific 'healthy'image from a
pathological one. Such images can be helpful in tasks such as anomaly detection and …

Optical coherence tomography angiography (OCTA) is an imaging modality that can be
used for analyzing retinal vasculature. Quantitative assessment of en face OCTA images …

Current unsupervised anomaly localization approaches rely on generative models to learn
the distribution of normal images, which is later used to identify potential anomalous regions …

The success of machine learning methods for computer vision tasks has driven a surge in
computer assisted prediction for medicine and biology. Based on a data-driven relationship …

Anomaly detection and localization in medical images is a challenging task, especially when
the anomaly exhibits a change of existing structures, eg, brain atrophy or changes in the …

Supervised learning for dermatology requires a large volume of annotated images, but
collecting clinical data is costly, and it is virtually impossible to cover all clinical cases …

Most image-to-image translation models postulate that a unique correspondence exists
between the semantic classes of the source and target domains. However this assumption …