Figure 1From: Assessment of significance of features acquired from thyroid ultrasonograms in Hashimoto's disease Average image power spectrum w (1) for different pictures. The values presented below images are values of the w(1) feature. A preliminary image analysis shows that the higher the homogeneity – lower frequency of changes, the lower the value of coefficient w(1). Higher, in terms of amplitude, changes, a higher frequency, a higher value of w(1). This feature is one of the 10 features that take part in creating a decision tree for classification of healthy subjects from patients. In this case, this feature is associated with Fourier decomposition and, to be more exact, with average image power spectrum.Back to article page