AQUATICS reconstruction software: the design of a diagnostic tool based on computer vision algorithms Andrea Giachetti 1 and Gianluigi Zanetti 1 CRS4 - Parco Scientifico e Tecnologico POLARIS, Edificio 1, Loc. Pixina Manna, 09010 Pula (CA), Italy {giach,zag}@crs4.it Abstract. Computer vision methods can be applied to a variety of medical and surgical applications, and many techniques and algorithms are available that can be used to recover 3D shapes and information from images range and volume data. Complex practical applica- tions, however, are rarely approachable with a single technique, and require detailed analysis on how they can be subdivided in subtasks that are computationally treatable and that, at the same time, allow for the appropriate level of user-interaction. In this paper we show an example of a complex application where, following criteria of efficiency, reliability and user friendliness, several computer vision techniques have been selected and customized to build a system able to support diagnosis and endovascular treatment of Abdominal Aortic Aneurysms. The system reconstructs the geometrical representation of four different struc- tures related to the aorta (vessel lumen, thrombus, calcifications and skeleton) from CT angiography data. In this way it supports the three dimensional measurements required for a careful geometrical evaluation of the vessel, that is fundamental to decide if the treat- ment is necessary and to perform, in this case, its planning. The system has been realized within the European trial AQUATICS (IST-1999-20226 EUTIST-M WP 12), and it has been widely tested on clinical data. 1 Introduction Computer vision (like computer graphics) provides many techniques and algorithms that can be used to recover 3D shapes and information from images range and volume data: pixel (voxel) classification techniques, isosurface extraction methods, etc. Some of them have been validated, even if only by using particular data under particular conditions, others are already used in industrial applications, while many others have only been proposed without the support of a large amount of experimental results. To choose the algorithm class that is best suited to a specific problem is indeed an important step in developing medical (as well as industrial) applications. Furthermore, a typical medical application may involve different sub-tasks, each with its own peculiarity, and often there is not a single reconstruction technique powerful enough to deal with all of them. The application designer needs, therefore, to find specific solutions for each subtask and to combine these solutions in a, possibly, user friendly and effective software tool. In the following sections we will show how we combined four different state-of-the art computer vision tools in a system that helps vascular surgeons and interventional radiologists in the pre– operative evaluation of the abdominal aorta. The system is capable of completely recover, with a fast and mostly automatic method, the geometrical structure of abdominal aorta from CT scans of the abdominal region, and to present clinicians with an interactive, measurable, 3D model of the vessel and ancillary structures. The reconstruction of the aorta is very important for the evaluation of Abdominal Aortic Aneurysms (AAA): the precise measurement and evaluation of their geometrical parameters is fundamental to estimate rupture risk and to plan surgical or endovascular interventions, see [2,5, 20, 24, 25]. Following the requirements coming from experts in vascular surgery, we developed specialized methods adapted to the different structures to be recovered, that is vessel lumen, vessel skeleton, plaques, thrombus. The result of our work are new flexible computer vision algorithms, and a user friendly software tool that includes all these algorithms and it can be used to obtain, in a fast and interactive way, a full vessel reconstruction.