The Expectation Maximization (EM) algorithm is an alternative reconstruction method to the Filtered Back Projection method, providing many advantages including decreased sensitivity to noise. However the algorithm requires a large number of iterations to reach adequate convergence. Due to this, research has been carried out into accelerating the convergence rate of the EM algorithm. In this paper we present an analysis of an EM implementation which uses both OSEM and MGEM, comparing results on a per time basis with both acceleration techniques alone as well as a combination of the two methods. We provide an alternative stopping criterion based on the RMS error of the projections of the current reconstruction and compare the result with an existing variance based approach.
Welcome to Mr. Robert Palmer who started his PhD in Medical Image Analysis under the supervision of Dr. Xianghua Xie. Robert obtained his BSc in Physics (Aber) and MSc in Medical Physics (Swansea).
Welcome to Dr. Rhodri Bevan who joined the group as a post-doc Research Assistant under the supervision of Dr. Xainghua Xie. Rhodri’s expertise is in computational modelling and he will be working on coronary disease modelling. This project is funded by the Welsh Office of Research and Development.
Xianghua Xie et al.,Image Gradient Based Level Set Methods in 2D and 3D, In Deformation Models, Edited by G. Hidalgo et al., Springer, September, 2012.
The 16th Conference on Medical Image Understanding and Analysis (MIUA) was held at Swansea University this year. Swansea University is set in rolling parkland overlooking the majestic sweep of Swansea Bay. The campus is a stone’s throw from the old fishing village of Mumbles and a short distance to the Maritime Quarter. The University enjoys a prime position overlooking Swansea Bay, the start of the famously dramatic Gower coastline. MIUA is the principal UK forum for communicating research progress within the community interested in image analysis applied to medicine and related biological science.
A new fully automatic object tracking and segmen- tation framework is proposed. The framework consists of a motion-based bootstrapping algorithm concurrent to a shape-based active contour. The shape-based active contour uses finite shape memory that is automatically and continuously built from both the bootstrap process and the active-contour object tracker. A scheme is proposed to ensure that the finite shape memory is continuously updated but forgets unnecessary information. Two new ways of automatically extracting shape information from image data given a region of interest are also proposed. Results demonstrate that the bootstrapping stage provides important motion and shape infor- mation to the object tracker. This information is found to be es- sential for good (fully automatic) initialization of the active contour. Further results also demonstrate convergence properties of the content of the finite shape memory and similar object tracking performance in comparison with an object tracker with unlim- ited shape memory. Tests with an active contour using a fixed- shape prior also demonstrate superior performance for the pro- posed bootstrapped finite-shape-memory framework and similar performance when compared with a recently proposed active con- tour that uses an alternative online learning model.
J. Chiverton, X. Xie, and M. Mirmehdi, Automatic Bootstrapping and Tracking of Object Contours, IEEE Transactions on Image Processing (T-IP), volume 21, issue 3, pages 1231 – 1245, March 2012.
Dr. Xianghua Xie was appointed as an Associated Editor of the IET Computer Vision journal. The vision of the journal is to publish the highest quality research work that is relevant and topical to the field, but not forgetting those works that aim to introduce new horizons and set the agenda for future avenues of research in Computer Vision.
Warm welcome to Dr. Huaizhong Zhang who joined the group as a RIVIC Research Assistant under the supervision of Dr. Xianghua Xie. Previously, Huaizhong was a post-doc researcher at Dublin.