Automated Nasopharyngeal Carcinoma Detection with Dynamic Gadolinium-Enhanced MR Imaging

Summary

Objectives: The purpose of this research is to develop an automatic medical diagnosis for segmenting nasopharyngeal carcinoma (NPC) with dynamic gadolinium-enhanced MR imaging.

Methods: This system is a multistage process, involving motion correction, head mask generation, dynamic MR data quantitative evaluation, rough segmentation, and rough segmentation refinement. Two approaches, a relative signal increase method and a slope method, are proposed for the quantitative evaluation of dynamic MR data.

Results: The NPC detection results obtained using the proposed methods had a rating of 85% in match percent compared with these lesions identified by an experienced radiologist. The match percent for the two proposed methods did not have significant differences. However, the computation cost for the slope method was about twelve times faster than the relative signal increase method.

Conclusions: The proposed methods can identify the NPC regions quickly and effectively. This system can enhance the performance of clinical diagnosis.

Keywords

Nasopharyngeal Neoplasms - Computer-Assisted Image Processing - Magnetic Resonance Imaging - Motion - Pharmacokinetics

• References

• 1 Muller-Schimpfle M, Brix G, Layer G. et al. Recurrent rectal cancer: Diagnosis with dynamic MR imaging. Radiology 1993; 189: 881-9.
• 2 Hawighorst H, Knapstein PG, Weikel W. et al. Cervical carcinoma: Comparison of standard and pharmacokinetic imaging. Radiology 1996; 201: 531-9.
• 3 Huang WC, Hsu CC, Lee CN, Lai PH. Recurrent nasal tumor detection by dynamic MRI. IEEE Engineering in Medicine and Biology 1999; 18 (Suppl. 04) 100-5.
• 4 Woods RP, Cherry SR, Mazziotta JC. Rapid automated algorithm for aligning and reslicing PET imaging. J Comput Assist Tomog 1992; 16 (Suppl. 04) 620-33.
• 5 Friston KJ, Ashburner J, Frith CD. et al. Spatial registration and normalization of images. Human Brain Mapping 1995; 3: 165-89.
• 6 Haralick RM, Joo H, Lee CN. et al. Pose Estimation from Corresponding Point Data. IEEE Trans smc 1989; 19 (Suppl. 06) 1426-46.
• 7 Meer P, Mintz D, Rosenfeld A. et al. Robust regression methods for computer vision: A review. International Journal of Computer Vision 1990; 6 (Suppl. 01) 59-70.
• 8 Nikou C, Heitz F, Armspach JP. et al. Registration of MR/MR and MR/SPECT brain images by fast stochastic optimization of robust voxel similarity measures. NeuroImage 1998; 8: 30-43.
• 9 Kass M, Witkin A, Terzopoulos D. Snakes: active contour models. Int Computer Vision 1987; 1 (Suppl. 04) 321-31.
• 10 Brix G, Semmler W, Port R. et al. Pharmacokinetic parameters in CNS Gd-DTPA enhanced MR imaging. J Comput Assist Tomog 1991; 15: 621-8.
• 11 Greenstein OS, Schnall MD, Livolsi VA. et al. Suspicious breast lesions: MR imaging with radiologic-pathologic correlation. Radiology 1994; 190: 485-93.
• 12 Alexander ME, Scarth G, Somorjai RL. An improved robust hierarchical registration algorithm. Magnetic Resonance Imaging 1997; 15 (Suppl. 04) 505-14.
• 13 Huber PJ. Robust Statistics. New York NY: John Wiley & Sons Press; 1981
• 14 Press WH, Teukolsky SA, Vetterling WT. et al. Numerical Recipes in C: The Art of Scientific Computing (2^nd ed.). Cambridge: Cambridge University Press; 1992: 683
• 15 Lee CN, Hsu CC, Yai SB. et al. Distributed Robust Image Mosaic. Proceedings of IEEE 14^th International Conference on Pattern Recognition (ICPR ’98). Brisbane: Aug. 16-20 1998: 1213-5.
• 16 Lee CN, Hsu CC, Wu MT. Robust Motion Correction in fMRI. Proceedings of Fifth International Conference on Functional Mapping of the Human Brain (HBM ’99). Dusseldorf; Germany: June 23-26, 1999. s 229.
• 17 Haralick RM, Shapiro LG. Computer and Robot Vision (vol. I). New York NY: Addison Wesley Press; 1992: 20-3.
• 18 Haralick RM, Shapiro LG. Computer and Robot Vision (vol. I). New York NY: Addison Wesley Press; 1992: 174-88.
• 19 Grace A. Optimization Toolbox User’s Guide. Mass. MA: The MathWorks Press; 1994