An Object-Oriented Client-Server System for Interactive Segmentation of Medical Images Using the Generalised Active Contours Model Alan L. Scheinine, Marco Donizelli, Marco Pescosolido, Piero Pili, Andrea Giachetti, Massimiliano Tuveri and Gianluigi Zanetti {scheinin, donza, marco, piero, giach, mtuveri, zag}@crs4.it CRS4 (Center for Advanced Studies, Research and Development in Sardinia) via Nazario Sauro 10, 09123 Cagliari (Italy) Abstract. In this paper we describe the first prototype of a distributed medical imaging system suitable for the visualisation and processing of medical images. The prototype is an object- oriented client-server system that provides a complete framework for the interactive segmentation of blood vessel contours from X-Ray Computed Tomography (CT) or Magnetic Resonance Imaging (MRI) scans, using the Generalised Active Contours Model. The system has been implemented exploiting the benefits of recent software developments, such as the Java programming language and the CORBA distributed object technology, which simplify the building, the maintenance and the portability of this kind of distributed applications. 1 Introduction Beyond the immediate diagnostic value of medical images from X-Ray Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), image data can be used for physical analysis and simulation. For example, the reconstruction as a geometric data set of a section of an artery could be used for medical procedural planning and for conducting research in vascular physiophathology through fluid dynamic simulations of blood flow [1]. This key role played by medical imaging has led to an increasing demand for electronic medical imaging systems (EMISs, [2]) that allow visualisation and processing of medical images. Recently, the growing demand for EMISs has been coupled with a need to access medical images and other diagnostic information remotely across networks, and to integrate and analyse data from various sources. These requirements, together with the development of data interchange standard formats such as DICOM [3], has also led to an increasing demand for distributed EMISs [4]. These systems go far beyond the original previsions of the instruments as stand-alone diagnostic workstations, and must have the capabilities to access images across networks, provide powerful image processing tools as well as image retrieval and storage mechanisms, manage various data from various sources, and integrate different software packages. Recent advances both in high-speed networks and in software developments, such as the Java programming language [5] and the CORBA distributed object technology [6], meet the needs underlying these distributed EMISs, improving their efficiency and simplifying their development, maintenance and portability. This paper describes the design and implementation of the first prototype of a distributed EMIS suitable for the visualisation and processing of medical images from different modalities. The system is characterised by an object-oriented client-server architecture that provides a complete framework for the interactive segmentation of CT or MRI scans of carotid artery sections using the Generalised Active Contour Model (GACM) [7]. The client side of the system is represented by a graphical user interface (GUI) written in Java, while a GACM-based algorithm written in C++ represents the server side. The client and the server are connected through a CORBA communication infrastructure, i.e., through a distributed object computing middleware. 2 The system requirements In the design of the proposed system, the following requirements have been considered: 1. The system GUI must be complete and usable, with all the tools useful to: • retrieve medical images from different sources (raw data or graphic files, DICOM archives, URLs); • visualise the images as in conventional film-based radiology; • manipulate the images in order to improve their appearance, quality, or to make easier the identification of the anatomical objects under interest; • interact with the GACM algorithm (due to the poor definition of medical images and the complexity of anatomical structures a user interaction is essential to guide the segmentation process [8]);