Robert J.T. Sadleir

Colon centreline calculation for CT colonography using optimised 3D opological thinning

CT colonography is an emerging technique for colorectal cancer screening. This technique facilitates noninvasive imaging of the colon interior by generating virtual reality models of the colon lumen. Manual navigation through these models is a slow and tedious process. It is possible to automate navigation by calculating the centreline of the colon lumen. There are numerous well documented approaches for centreline calculation. Many of these techniques have been developed as alternatives to 3D topological thinning which has been discounted by others due to its computationally intensive nature. This paper describes a fully automated, optimised version of 3D topological thinning that has been specifically developed for calculating the centreline of the human colon.

Automated synthesis, insertion, and detection of polyps for CT colonography

CT Colonography (CTC) is a new non-invasive colon imaging technique which has the potential to replace conventional colonoscopy for colorectal cancer screening. A novel system which facilitates automated detection of colorectal polyps at CTC is introduced. As exhaustive testing of such a system using real patient data is not feasible, more complete testing is achieved through synthesis of artificial polyps and insertion into real datasets. The polyp insertion is semi-automatic: candidate points are manually selected using a custom GUI, suitable points are determined automatically from an analysis of the local neighborhood surrounding each of the candidate points. Local density and orientation information are used to generate polyps based on an elliptical model. Anomalies are identified from the modified dataset by analyzing the axial images. Detected anomalies are classified as potential polyps or natural features using 3D morphological techniques. The final results are flagged for review. The system was evaluated using 15 scenarios. The sensitivity of the system was found to be 65% with 34% false positive detections. Automated diagnosis at CTC is possible and thorough testing is facilitated by augmenting real patient data with computer generated polyps. Ultimately, automated diagnosis will enhance standard CTC and increase performance.

A Hybrid Lossless Compression Scheme for Efficient Delivery of Medical Image Data over the Internet

Medical imaging applications generate large volumes of medical data leading to challenges for transmission and storage. In this paper, a novel lossless 3D compression scheme for medical image delivery is proposed. It is based on Prediction by Partial Matching (PPM) in combination with 3D JPEG-4 compression. This scheme is intended for accelerating the delivery of data over the Internet and consequently we are concerned with both compression ratio and decompression time. Experimental results illustrate that the proposed scheme is efficient and feasible in terms of both compression ratio and decompression speed.

A visual programming environment for machine vision engineers

This paper details a free image analysis and software development environment for machine vision application development. The environment provides high‐level access to over 300 image manipulation, processing and analysis algorithms through a well‐defined and easy to use graphical interface. Users can extend the core library using the developers interface via a plug‐in which features automatic source code generation, compilation with full error feedback and dynamic algorithm updates. Also discusses key issues associated with the environment and outline the advantages in adopting such a system for machine vision application development.

A Segmentation Based Lossless Compression Scheme for Volumetric Medical Image Data

It is not acceptable to use loss less techniques when compressing medical image data, and as a result, it is difficult to achieve high compression ratios. We have developed a novel segmentation based compression scheme to overcome this problem and our experimental results indicate that this scheme is capable of achieving a high level of compression without sacrificing the quality of the patient data.

Distributed medical imaging applications using Java technology

Advances in Internet technologies have opened up new opportunities in relation to medical image interpretation. This task can be accomplished outside the hospital using distributed medical imaging applications. In this paper, a Java-based distributed medical imaging application is presented. This application is able to access to a remote medical image dataset via a network and provide the necessary interpretation functions, such as windowing and fly-through visualisations. Experimental results show that this Java-based medical imaging application has the ability to provide comprehensive functionality for radiology interpretation as well as a high level of user friendliness. Thus demonstrating Java is a suitable tool for developing distributed medical imaging applications.

A Web-Based Training System for Remote Access Mammography Screening

Breast cancer is a major cause of cancer related death worldwide. Screening for breast cancer is achieved using mammography. In this research, we propose a web-based training system for remote access mammography in order to train the radiologists for improving their skills and experiences in interpretation of mammograms, so as to reduce the misdiagnosis. At this stage, we have developed a software viewing tool, called Viewer that allows user to teach themselves by displaying the images on it and drawing the overlay on images.

Medical image compression using region-based prediction

This paper describes a novel technique that uses prior knowledge of anatomical information to improve the performance of medical image compression. This technique uses a series of predictors that have been optimised to deal with specific regions within medical image datasets. Instead of relying on a global prediction model, the proposed technique adaptively switches to an optimal predictor according to the characteristics of the region being compressed. Experimental results show that the proposed adaptive prediction method indeed achieves high prediction accuracy and when combined with an efficient entropy encoder, it provides a higher compression ratio than current general purpose state-of-the-art alternatives.

An edge-based prediction approach for medical image compression

The types of edges found in volumetric medical image data are typically smooth due to a phenomenon known as the Partial Volume Effect (PVE). These smooth edges are very different to the sharp edges typically found in real world images. Consequently, it is not appropriate to use conventional edge-based prediction schemes with medical image data. This paper proposes a novel edge-based prediction scheme that is optimised for use with the types of edges associated with the PVE. This technique exploits prior anatomical knowledge and the characteristics of the PVE regions to accurately predict unknown voxel data. Our experimental results show that the performance of this technique in edge regions is significantly better than previously documented edge-based prediction techniques. We also show that the inclusion of our proposed technique in a standard compression scheme improves the overall level of compression that can be achieved.