Greek imaging system provides MITIS touch in ob/gyn

Digital imaging services have now reached into ob/gyn corridors, thanks to a system being tested by researchers in Athens. MITIS uses the Web to manage and process obstetric, gynecologic, and radiologic data.

The imaging system records all necessary medical information in terms of patient data, examinations, and operations. It provides the user with advanced image processing tools for manipulation, processing, and storage of ultrasound and mammographic images, according to George K. Matsopoulos, Ph.D., of the Institute of Communication and Computer Systems, National Technical University of Athens.

The institute has a track record of developing algorithms for image registration, including CT/MRI/SPECT and portal images.

"We developed a system for estimating patient setup error during radiotherapy treatment, where the errors calculated are less then 1 mm," Matsopoulos said.

That system, described in the July issue of Physics in Medicine and Biology, is now in clinical use after a year of testing at the Hygeia Hospital, a private hospital in Athens.

The MITIS system, described in the October issue of Computer Methods and Programs in Biomedicine, can be installed in a hospital's local area network, where it can access any available PACS servers or any other server within the radiology department. It has been installed at the Alexandra National Hospital in Athens and Attikon University General Hospital for test purposes. Demonstration projects are also under way in a number of local gynecology clinics.

Because it is Web-based, the system permits usage from within the hospital or from physician offices outside the hospital.

"In the near future, MITIS can even be offered as an ASP for private physicians as a patient management system, although the current installation works only as an in-hospital, intranet-based information system," Matsopoulos said.

MITIS offers diagnostic support as well as educational services to physicians and medical students using advanced image processing algorithms for segmentation and classification of mammographic images, according to Matsopoulos.

"The system currently can classify only normal and pathological cases," he said.

MITIS complies with DICOM 3.0 for image archiving and retrieval. It is composed of a set of independent Web modules and a Win32 application for mammography image processing and evaluation.