Distinguishing recurrent nasal tumor from post-treatment changes by dynamic MR images

The purpose of this research is to detect and enhance the tumor. region by computing the relative intensity difference between consecutive MR images after using contrast agents. We apply a relative signal increase (RSI) model to recognize the recurrent nasal tumor from dynamic MR images. Image registration is required to deal with the matching corresponding points among different images. The phase correlation technique and function minimization is used for image registration. The active contour technique is used to refine the region of interest automatically and more precise to define the interested a,sa. The quantitative evaluation of dynamic MR data is modeled by fitting the three-parameter time-intensity curves. From The color distribution of rime constant for arrival of contrast intensity and enhancement amplitude and the color representation in the final images show that our proposed method can identify the nasal tumor effectively. The results enhance the clinical diagnosis function and greatly increase the precision of the recurrent nasal tumor detection.