MEDIMAN - AN OBJECT-ORIENTED PROGRAMMING APPROACH FOR MEDICAL IMAGE-ANALYSIS

Mediman is a new image analysis package which has been developed to an alyze quantitatively Positron Emission Tomography (PET) data. It is ob ject-oriented, written in C++ and its user interface is based on Inter Views on top of which new classes have been added. Mediman accesses da ta using external data representation or import/export mechanism which avoids data duplication. Multimodality studies are organized in a sim ple database which includes images, headers, color tables, lists and o bjects of interest (OOI's) and history files. Stored color table param eters allow to focus directly on the interesting portion of the dynami c range. Lists allow to organize the study according to modality, acqu isition protocol, time and spatial properties. OOI's (points, lines an d regions) are stored in absolute 3-D coordinates allowing correlation with other co-registered imaging modalities such as MRI or SPECT. OOI 's have visualization properties and are organized into groups. Quanti tative ROI analysis of anatomic images consists of position, distance, volume calculation on selected OOI's. An image calculator is connecte d to mediman. Quantitation of metabolic images is performed via profil es, sectorization, time activity curves and kinetic modelling. Mediman is menu and mouse driven, macro-commands can be registered and replay ed. Its interface is customizable through a configuration file. The be nefit of the object-oriented approach are discussed from a development point of view.