Ai. Penn et Mh. Loew, ESTIMATING FRACTAL DIMENSION WITH FRACTAL INTERPOLATION FUNCTION MODELS, IEEE transactions on medical imaging, 16(6), 1997, pp. 930-937
Fractal dimension (fd) is a feature which is widely used to characteri
ze medical images, Previously, researchers have shown that fd separate
s important classes of images and provides distinctive information abo
ut texture, We analyze limitations of two principal methods of estimat
ing fd: box-counting (BC) and power spectrum (PS), BC is ineffective w
hen applied to data-limited, low-resolution images; PS is based on a f
ractional Brownian motion (fBm) model-a model which is not universally
applicable, We also present background information on the use of frac
tal interpolation function (FIF) models to estimate fd of data which c
an be represented in the form of a function. We present a new method o
f estimating fd in which multiple FIF models are constructed, The mean
of the fd's of the FIF models is taken as the estimate of the fd of t
he original data, The standard deviation of the fd's of the FIF models
is used as a confidence measure of the estimate, We demonstrate how t
he new method can be used to characterize fractal texture of medical i
mages, In a pilot study, we generated plots of curvature values around
the perimeters of images of red blood cells from normal and sickle ce
ll subjects, The new method showed improved separation of the image cl
asses when compared to BC and PS methods.