At the University of Joensuu in Finland, the researchers Tuomo Kauranne and Alexander Kolesnikov have developed a tool for interfacing a Digital Imaging and Communications in Medicine (DICOM) image database with the World Wide Web (WWW). As one of the four telemedicine platforms, the University of Joensuu is a major partner in the Euromed project with regard to remote medical analysis applications. The research team has chosen to work with a server-side HTML-embedded script language, called PHP/FI, which allows items in the database to be created, retrieved, updated and deleted. To use the WWW-DICOM system, the end-user runs his WWW-browser requesting the PHP/FI scripts via a query form to start consultation or manipulation of the required medical images.
The DICOM image archiving system, which is used for the project, actually consists of a web based relational Mini Software Query Language (SQL) database filled with records referring to stored DICOM files. The database table structure is based upon a four level hierarchy. The Patient level forms the top level, containing multiple Studies. A Study in turn, comprises one or several Series, but always belonging to a definite Study. The third level Series consists of one or multiple Images, but an Image always forms part of a definite Series. The lowest level is the Image level where Images originating from a single imaging modality, belong to the same Series.
In order to get access to the database, the user activates his WWW-browser to specify the URL of a DICOM Picture Archival Computer System (PACS) workstation. The thus invoked PHP/FI script generates an HTML file, displaying a query form with two fields: Patient name and Patient ID. The user fills out the form and submits his request via the PHP/FI script which enters the search into the relational database to determine the patients corresponding to the input criteria. The user selects a patient and subsequently obtains a list display of the patient's personal studies, series and images.
Once the images are displayed, the user can perform a range of image processing procedures by means of special feature detection tools. It is for instance possible to convert images from DICOM to JPEG format or to transform them in a piecewise-linear manner; to sharpen or equalise them or to execute isolines, equidensities and edge detection on them. These processing programmes are written in C and can be loaded from a PHP/FI script. In fact, the PHP/FI script in syntax is similar to C one and constitutes a Common Gateway Interface (CGI) wrapper language. It is fairly easy to add other image processing and medical analysis facilities without having to recompile the whole application.
Obviously, the DICOM archiving system needs updating, maintenance and support. The Joensuu team therefore has created a set of web based tools to upload client DICOM files to the DICOM database WWW-server; to edit items in the image database as well as to add DICOM files to it. Thus, the authorised remote client at all times is able to add new data to the database and to update the old ones. The scientists Kauranne and Kolesnikov are presenting a wide range of illustrative links at the web site of the University of Joensuu. Please, do have a look at them.