Automatic segmentation of organs at risk in head Magnetic Resonance Images (MRI) is a challenging task in medical image analysis. This operation is fundamental for radiotherapy treatment planning: accurate delineation of critical structures allow calibrating the radiation beam in order to hit tumour cells and preserve sane tissues, consuming a time of much lower than a radiation oncologist. In this paper we analyze the properties of head MRI and of their OARs and propose an algorithm that exploits the knowledge implied in an atlas, represented by a labelled medical image, and uses a modified version of Gradient Vector Flow Snake endowed with a parameters automatic tuning mechanism system based on Fourier Descriptors. The comparison of this method with the other traditional algorithms based on active contours showed a remarkable increase of performance.
Atlas-Based Segmentation of Organs at Risk in Radiotherapy in Head MRIs by Means of a Novel Active Contour Framework / Bevilacqua, Vitoantonio; Piazzolla, Alessandro; Stofella, Paolo (LECTURE NOTES IN COMPUTER SCIENCE). - In: Advanced Intelligent Computing Theories and Applications. With Aspects of Artificial Intelligence : 6th International Conference on Intelligent Computing, ICIC 2010, Changsha, China, August 18-21, 2010. Proceedings / [a cura di] De-Shuang Huang; Xiang Zhang; Carlos Alberto Reyes García; Lei Zhang. - STAMPA. - Berlin, Heidelberg : Springer, 2010. - ISBN 978-3-642-14931-3. - pp. 358-367 [10.1007/978-3-642-14932-0_44]
Atlas-Based Segmentation of Organs at Risk in Radiotherapy in Head MRIs by Means of a Novel Active Contour Framework
Vitoantonio Bevilacqua;
2010-01-01
Abstract
Automatic segmentation of organs at risk in head Magnetic Resonance Images (MRI) is a challenging task in medical image analysis. This operation is fundamental for radiotherapy treatment planning: accurate delineation of critical structures allow calibrating the radiation beam in order to hit tumour cells and preserve sane tissues, consuming a time of much lower than a radiation oncologist. In this paper we analyze the properties of head MRI and of their OARs and propose an algorithm that exploits the knowledge implied in an atlas, represented by a labelled medical image, and uses a modified version of Gradient Vector Flow Snake endowed with a parameters automatic tuning mechanism system based on Fourier Descriptors. The comparison of this method with the other traditional algorithms based on active contours showed a remarkable increase of performance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.