Component-based technique for determining the effects of acupuncture for fighting migraine using SPECT images

Abstract

In this work, SPECT brain images are analyzed automatically in order to determine the effects of acupuncture applied for fighting migraine. For this purpose, two different groups of patients are randomly collected and received verum and sham acupuncture, respectively. Changes in the brain perfusion patterns can be measured quantitatively by dealing with the images in a classification context. A classification scheme consisting of a component-based feature extraction technique in combination with Support Vector Machines allows us to accurately determine the regions of interest (ROIs) where acupuncture produced more intense effects, and whether these effects are correlated with a decrease or an increase of the brain activity. Effects produced by verum and sham acupuncture are studied, and the best method for intensity normalization is discussed. The result is a complete, objective system which can be used for general purposes in the visual assessment of perfusion images.