Adam Szmul - Tony Cheung - CMIC seminar series

Adam Szmul - Tony Cheung - CMIC seminar series

Adam Szmul

Title: Supervoxel-based Image Registration and Analysis with Application to Lung Cancer

Abstract:

Lung diseases, including lung cancer, are amongst the largest burdens to healthcare systems worldwide. Improving the extraction of information from imaging data has the potential to provide more accurate diagnosis and more effective treatment. One meaningful regional representation is achieved by locally clustering similar image voxels together using the concept of supervoxels, which allows for bulk regional processing and analysis, whilst removing redundant information. We demonstrate how a supervoxel-based image representation can be combined with graph cuts as a discrete optimisation-based approach to provide efficient and accurate 3D deformable image registration. 

The resulting deformation fields might be further applied to estimate lung ventilation maps. We present a novel approach for estimating regional lung ventilation from dynamic lung CT imaging. Our method combines a supervoxel-based image representation with deformable image registration, performed between peak breathing phases, during which we track changes in intensity for a number of layers of previously extracted supervoxels. Such a region-based approach is expected to be physiologically more consistent with lung anatomy than methods relying on voxel-wise

relationships.

Our work also presents novel approaches for performing more accurate hyperpolarised Xenon129 MRI (XeMRI) analysis. We propose a multimodal lung image registration enhanced with personalised motion model approach derived from lung 4DCT. The approach with a prior motion model is particularly important for regions where there is not enough information to reliably drive the registration process, as in the case of XeMRI and proton density MRI (pMRI) to CT registration. Subsequently we introduce a framework for breathing motion correction of the dynamic

sequence of XeMRI using a lung atlas-based approach. Such a method for the breathing motion correction is expected to facilitate the analysis of temporal lung ventilation.

Tony Cheung

Title: Ultrasound elastography, contrast enhanced ultrasound and their applications to clinical and translational research studies

Abstract:

Ultrasound elastography and contrast enhanced ultrasound have been widely used in medical diagnosis. In this talk, I will first present an ultrasound elastography technique for detecting and quantifying inhomogeneous displacement in horse tendon. Then, I will talk about the development of several algorithms which correct attenuation artefacts and quantify microbubbles signals in contrast enhanced ultrasound images. These techniques have been applied to clinical studies including quantification of plaque neovascularisation in carotid arteries and vascular density in human lower limb.