Automated Interpretation of Diagnosti Images: The AIDI-Heart Proje t Mattias Ohlsson, PhD1, Andreas J arund2, Lars Edenbrandt, MD, PhD2 1 Department of Theoreti al Physi s, Lund University, Sweden, mattias thep.lu.se 2 Department of Clini al Physiology, Lund University, Sweden http://ism.thep.lu.se/ Abstra t. This paper presents AIDI-Heart, a omputer-based de ision support tool for automated interpretation of diagnosti heart images, whi h is made available via the web. The tool is based on image pro essing te hniques, arti ial neural networks, and large and well validated medi al databases. The performan e of the tool has been evaluated in several re ent papers and the results show the high potential for the tool as a lini al de ision support system. The tool has now been integrated into a WWW environment for easy a ess and operation using your favourite web-browser. This rst version of AIDI-Heart is evaluated by three hospitals, two in Denmark and one in Sweden. 1 Introdu tion
[1]
L Edenbrandt,et al.
Automated interpretation of myocardial SPECT perfusion images using artificial neural networks.
,
1997,
Journal of nuclear medicine : official publication, Society of Nuclear Medicine.
[2]
G. Dorffner,et al.
Automated interpretation of planar thallium-201-dipyridamole stress-redistribution scintigrams using artificial neural networks.
,
1994,
Journal of nuclear medicine : official publication, Society of Nuclear Medicine.
[3]
L Edenbrandt,et al.
Improved classifications of myocardial bull's-eye scintigrams with computer-based decision support system.
,
1999,
Journal of nuclear medicine : official publication, Society of Nuclear Medicine.