Title :
Recurrent nasal tumor detection by dynamic MRI
Author :
Huang, Wen-Chen ; Hsu, Cheng-Chung ; Lee, Chungnan ; Lai, Ping-Hang
Author_Institution :
Dept. of Manage. Ing. Sci., Chia-Nan Coll. of Pharm. & Sci., Tainan, Taiwan
Abstract :
The purpose of this research is to detect and enhance the recurrent nasal tumor region by computing the relative intensity difference between consecutive MR images after using a contrast agent. In this article, we apply a relative signal increase model to recognize a recurrent nasal tumor by dynamic MR images. A robust estimation technique is used to deal with matching corresponding points among different images. The active contour technique is applied to refine automatically the region of interest and obtain a more precise definition of the area of interest. The quantitative evaluation of dynamic MR data is modeled by fitting three-parameter time-intensity curves
Keywords :
biomedical MRI; edge detection; image enhancement; image matching; image registration; medical image processing; motion compensation; tumours; active contour technique; boundary detection; consecutive MRI images; contrast agent; dynamic MRI; image matching problem; matching corresponding points; motion correction; recurrent nasal tumor detection; relative intensity difference; relative signal increase model; robust estimation technique; three-parameter time-intensity curves; Active contours; Biomedical imaging; Brain modeling; Computer displays; Hopfield neural networks; Image converters; Image segmentation; Magnetic resonance imaging; Neoplasms; Pixel;
Journal_Title :
Engineering in Medicine and Biology Magazine, IEEE
