Authors:
Hiroshi Arisawa
;
Takako Sato
and
Sinya Hata
Affiliation:
Graduate School of Yokohama Nat. Univ., Japan
Keyword(s):
Medical Imaging, PET-CT, Expert Systems.
Related
Ontology
Subjects/Areas/Topics:
Biomedical Engineering
;
Cardiovascular Technologies
;
Computing and Telecommunications in Cardiology
;
Decision Support Systems
;
Health Engineering and Technology Applications
;
Health Information Systems
;
Medical and Nursing Informatics
Abstract:
Computer Assisted Diagnosis (CAD) is one of the promising technologies for the future Medical Image Processing Systems. Among them, whole-body PET (Positron Emission Tomography) and X-ray CT (Computer Tomography) image based cancer detection has been playing an essential role in the modern medical world. Using PET-CT images the Radiologist can find a very small cancer or a malignant tumor. On the other hand, this diagnosis process is very stressful work, because such area is too small and localized but may appear at any place of patient bodies. This paper presents an automated diagnosis system in order to improve the above difficulties. The system consists of three parts, Diagnosing Algorithm, Algorithm Interpreter(Engine) and Image Viewer. The algorithm and the engine can reproduce doctor's methods faithfully as the rule-based inference system. Using this system, we made an retrospective studies for the actual group of patients and the results shows the usefulness of this approach.