Title :
Automatic thermal image analysis for medical diagnosis
Author :
Varga, M.J. ; Hanka, R.
Author_Institution :
Defence Res. Agency, Malvern, UK
Abstract :
Thermographic images can be used to generate diagnostic images with classification of disease or diseases and indication of degree of severity. Statistical pattern recognition techniques are used to analyse the results of a temperature stress test. The hand is immersed in a cold water bath at 20°C for 1 minute. In order to record the bodies response to the temperature stress thermograms are taken at regular intervals for 2 minutes for 15 to 20 minutes. The series of thermograms for each patient or control are standardized and aligned using a dynamic stretching technique, so as to obtain a series of images that can be used in a fully automated diagnostic aid. The series of images are presented in a `dual representation´ for processing. The purpose of this aid is the classification of the response to the cold stress into different classes: Raynaud type conditions, inflammatory conditions (rheumatic and arthritic) or normal
Keywords :
computerised pattern recognition; computerised picture processing; medical diagnostic computing; Raynaud type conditions; SPR; diagnostic images; inflammatory conditions; medical diagnosis; pattern recognition techniques; temperature stress test; thermal image analysis; thermograms;
Conference_Titel :
Image Processing and its Applications, 1992., International Conference on
Conference_Location :
Maastricht
Print_ISBN :
0-85296-543-5