Quantifying skull shape

This paper describes a technique for automatically quantifying the shape of the skull cavity seen on an axial slice in a CT brain scan. The development ofthis algontbm derives from the need to normalise CT scan data acqording to skull size and shape for the purpose of compar;lg new patient data with that from past cases. This algorithm uses image processing techniques to fmdthe h i e r edge of the bones of the skull on an axial slice, so that its shape can be represented by a radius function. A simple m e q e of shape asymmehy is dehned It is also shqwn that this shape can be quantified more precisely by harmonic analysis using the Fourier Transform of the radius function. This paper describes the design of the algorithm and its performance on axial slices taken from a database ofCT brain scans from 528 patients. Kev Words: human brain shaoe. CT head scanner. skull measurement, normalisation, shape analysis, Fourier series, . , morphomehy, grey level imaging (GLI).