City University of Hong Kong

Jing Liao

Sam Tak Wu Kwong

Jialu Huang

Hao Du

Qihua Dong

A weakly-supervised framework for segmenting a three-dimensional (3D) medical image incorporates information of geometric prior and contrastive similarity in a loss-based fashion for enhancing distinguishability of an organ in presence of complex organ shape and image artifacts (e.g., low-contrast tissues) in the image. Training images with bounding-box annotations are used to train the framework. The geometric prior is advantageously built on point-cloud representations, which are much more fine-grained and flexible than volume representations, thus serving as better supervision than an inherent property of the bounding-box annotations (i.e. height and width). Furthermore, using contrastive similarity encourages organ pixels to gather around in a contrastive embedding space. It helps more-effectively distinguish low-contrast tissues from the background than using the conventionally-used gray space.

Weakly-supervised 3D medical image segmentation using geometric prior and contrastive similarity

A method of providing an image to image translation using a style based generator architecture for generative adversarial networks. The method includes the step of applying a feature alignment loss function defined by:Loss=Σ1(ϕ(),ϕ())wherein Gand Gare respective generators of a source model and a target model, ϕ(·)is the extraction of the ith layer of features and wrepresents the weight of Lloss at the ith layer, the Lloss weight wbeing reduced as layers decrease such that Lloss at higher layers has a higher weight than the Lloss at lower layers.

Growing community of inventors

Hong Kong, China

Jing Liao