An application of genetic algorithms to geometric model-guided interpretation of brain anatomy

• Published in: Pattern recognition Vol. 30; no. 2; pp. 217 - 227
• Main Authors: Undrill, P.E, Delibasis, K, Cameron, G.G
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.02.1997; Elsevier Science
• Subjects: Algorithmics. Computability. Computer arithmetics; Applied sciences; Artificial intelligence; Biological and medical sciences; Brain anatomy; Computer science; control theory; systems; Computerized, statistical medical data processing and models in biomedicine; Exact sciences and technology; Fourier descriptors; Genetic algorithms; Geometrical models; Image interpretation; Image segmentation; Magnetic resonance imaging; Medical sciences; Models and simulation; Object recognition; Optimization; Pattern recognition. Digital image processing. Computational geometry; Theoretical computing; Three dimensional; Volumetric quantitation; Geometrical models; Fourier descriptors; Image interpretation; Volumetric quantitation; Brain anatomy; Genetic algorithms; Volumetric analysis; Segmentation; Image processing; Fast Fourier transformation; Magnetic resonance; Expert system; Anatomy; Acquisition; Modeling; Geometrical model; Algorithm performance; Genetic algorithm; Tridimensional image
• ISSN: 0031-3203; 1873-5142
• DOI: 10.1016/S0031-3203(96)00074-X

This work applies 3D Fourier Descriptors (FDs) and Genetic Algorithms (GAs) to the optimisation of the shape and position of models of anatomical objects within the human brain. Using magnetic resonance image data, we perform an approximate segmentation on one lateral ventricle and use the FDs from this as seeding values for the GAs to search for the left and right lateral ventricles in subsequent 3I) image data sets, showing that the method is capable of coping with normal biological variation within and between individuals. Finally, we compare the GA-guided segmentation with a manual region growing method and find an agreement of 79.9±5.8% in voxel classification with a corresponding mean edge placement error of 2.1±0.4 mm.