John Michael Knapman

Interactive image segmentation for radiation treatment planning

COVIRA (COmputer VIsion in RAdiology) is a project in the European Communitys Advanced Informatics in Medicine program. The goal is to improve the diagnosis and planning of treatment (radiotherapy) for patients with brain tumors and other diseases. The aim of radiotherapy is to provide a high dose of radiation to a tumor while sparing as much as possible of the surrounding healthy tissue. A necessary first step is defining the target volume and organs at risk by manually outlining the required contours on magnetic resonance or computed tomography scans. For a full three-dimensional plan this is time-consuming, as 40 or more scans are used. Computer image segmentation speeds up the process, and a method that combines information from edge and region detectors is described. Since this method is not able to completely meet the clinical requirements, an interactive image segmentation algorithm has been developed that enables the operator to employ clinical judgment. Probabilities are assigned to edges and regions and presented to the user as a hierarchy of segmentations. The approach is being subjected to extensive clinical evaluation, using pilot applications running on IBM RISC System/6000 workstations.

Modeling of Architectures with UML

A critical level of abstraction in the modeling of a large, complex system is its architecture. At an architectural level one models the principal system elements and their interaction. Architectural models are typically used to provide an intellectually tractable, birds-eye view of a system and to permit design-time reasoning about system-level concerns such as performance, reliability, portability, and conformance to external standards and architectural styles. In practice most architectural descriptions are informal documents. They are usually centered on box-and-line diagrams, with explanatory prose. Visual conventions are idiosyncratic, and usually project specific. As a result, architectural descriptions are only vaguely understood by developers, they cannot be analyzed for consistency or completeness, they are only hypothetically related to implementations, their properties cannot be enforced as a system evolves, and they cannot be supported by tools to help software architects with their tasks. There exist several architecture description languages, but we are interested in the use of UML. We aim to identify requirements on architectural modeling and how different modeling concepts of UML meet these requirements. This paper is not intended as a critique of the UML but as a discussion of approaches to modeling architectures that have been tried, more or less successfully.

Interactive Image Segmentation Applied to CT and MR Images

In radiation treatment planning, it is necessary to mark out selected parts of CT or MR images which are clinically relevant, before designing the treatment plan. This process is usually done manually with a mouse, and for a typical data set of some 40 slices it can be a time consuming and tedious process. Automatic image segmentation techniques have been investigated, but are of little use in this field since it is difficult to incorporate the relevant clinical knowledge. We describe here two methods: Interactive Volume Growing and Hierarchical Probabilistic Segmentation, which assist the clinical user in outlining relevant volumes whilst allowing him full control of the process. Both methods have been integrated into one of the pilot systems developed as part of the COVIRA project.

Dupin's cyclide and the cyclide patch

Abstract To retrieve models from a database for recognizing objects in stereo, a new formulation of Dupins cyclide patches provides a succinct representation of surface shape. The parameters can be extracted from the Weingarten map and its derivatives at a point where a contour meets an extremal boundary.

Business-oriented constraint language

The Business-oriented Constraint Language (BCL) is proposed as a means of annotating diagrams in UML. BCL is grounded in the Object Constraint Language (OCL) but is designed particularly to address the needs of people who are concerned with enterprise application integration (EAI), although it may be more widely applicable. EAI often requires a loosely coupled event-based architecture in which timing and statistical measures are important; these are described in another paper (1). BCL provides these features together with a syntax that is flexible and extensible. It is intended to be accessible to most practitioners, including those who do not have a mathematical background.

Segmenting Cranial MR Images using Edge-Region Integration

As part of the EEC funded AIM project COVIRA to improve visualisation of MR data for radiology, radiation therapy planning and neurosurgery, a data driven method for image segmentation has been developed. Its results have been compared to ideal data prepared manually by a group of radiologists. There was found to be an overlap between the manually and automatically produced results of 84% to 93% and an evaluation showed the usefulness of the overlapping edges as high.

Advances in 3D Model Identification from Stereo Data.

Volumetric (CAD) models of objects are converted to a wire frame representation which is compiled into a data base and represented in vectors and matrices that characterise both local geometrical relationships and the structure of the models. This characterisation is independent of position and orientation and supports variable size. Using a stereo vision system, instances of these objects are then identified from pairs of images containing single objects or more than one object.