Tamaz Amiranashvili

TL;DR: The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations.

TL;DR: Casp et al. as mentioned in this paper proposed FlowSSM, a shape model that learns shape variability without requiring dense correspondence between training instances, based on a hierarchy of continuous deformation flows which are parametrized by a neural network.

TL;DR: In this paper , the authors proposed a method for high-resolution shape reconstruction from sparse measurements without relying on high resolution ground truth for training, which is based on neural implicit shape representations and learns a continuous shape prior only from highly anisotropic segmentations.

TL;DR: In this article , the authors proposed a method compressing complex traditional strategies for solving an inverse problem into a simple database query task, which can achieve one order speedup compared to existing approaches for solving the inverse problem.

TL;DR: In this article , contrastive self-supervised representation (CSR) is used as a metric to measure perceptual similarity between two images in the medical domain, and CSR can significantly improve the image quality in two image synthesis tasks.