CT AND MR ASSESSMENT OF TUMORS OF THE NOSE AND PARANASAL SINUSES, THE NASOPHARYNX AND THE PARAPHARYNGEAL SPACE USING ROC METHODOLOGY

被引:18
作者
HUNINK, MGM
DESLEGTE, RGM
GERRITSEN, GJ
SPEELMAN, H
机构
[1] FREE UNIV AMSTERDAM HOSP,DEPT DIAGNOST RADIOL,AMSTERDAM,NETHERLANDS
[2] FREE UNIV AMSTERDAM HOSP,DEPT EAR NOSE & THROAT SURG,AMSTERDAM,NETHERLANDS
[3] ERASMUS UNIV,HOSP DIJKZIGT,CTR CLIN DECIS ANAL,ROTTERDAM,NETHERLANDS
关键词
CT; comparative study; MRI; Receiver operating characteristic methodology; Tumors; nose and paranasal sinuses; nasopharynx; parapharyngeal space;
D O I
10.1007/BF00589116
中图分类号
R74 [神经病学与精神病学];
学科分类号
摘要
Neoplastic disease of the nose and paranasal sinuses, the nasopharynx and the parapharyngeal space requires thorough assessment of location and extension in order to plan appropriate treatment. This study evaluates computer tomography and magnetic resonance imaging in the workup of malignant and non-malignant tumors of the nose and paranasal sinuses, the nasopharynx and the parapharyngeal space in 76 patients. An attempt is made to characterize histopathology on magnetic resonance images by analyzing the signal intensities on T1- and T2-weighted images relative to muscle and brain tissue. The test performance of computer tomography and magnetic resonance imaging in the assessment of tumor extension are compared with receiver operating characteristic methodology. Although no definitive conclusions can be made as to the histopathology on the basis of the signal intensities on magnetic resonance imaging, some tumors show characteristic images. Receiver operating characteristic analysis of the performance of computer tomography and magnetic resonance imaging in the assessment of extension of neoplastic disease of the nose and paranasal sinuses, the nasopharynx and the parapharyngeal space, demonstrates no statistically significant difference in overall test performance. However, in evaluating regions involving predominantly soft tissue structures and comparatively large bony structures magnetic resonance imaging is superior to computer tomography, whereas in evaluating regions involving thin bony structures, computer tomography performs better than magnetic resonance imaging. © 1990 Springer-Verlag.
引用
收藏
页码:220 / 225
页数:6
相关论文
共 15 条
[1]   ASSESSMENT OF DIAGNOSTIC-TESTS WHEN DISEASE VERIFICATION IS SUBJECT TO SELECTION BIAS [J].
BEGG, CB ;
GREENES, RA .
BIOMETRICS, 1983, 39 (01) :207-215
[2]   ASSESSMENT OF RADIOLOGIC TESTS - CONTROL OF BIAS AND OTHER DESIGN CONSIDERATIONS [J].
BEGG, CB ;
MCNEIL, BJ .
RADIOLOGY, 1988, 167 (02) :565-569
[3]  
CARTER BL, 1988, COMPUT TOMOGR, V1, P445
[4]  
Centor R M, 1985, Med Decis Making, V5, P139, DOI 10.1177/0272989X8500500203
[5]  
CENTOR RM, 1985, MED DECIS MAKING, V5, P149
[6]  
DESLEGTE RGM, 1988, ESSENTIALS CLIN MRI
[7]  
DESPLANTES BGZ, 1988, MAGNETIC RESONANCE I, V1, P308
[8]   MAXIMUM LIKELIHOOD ESTIMATION OF PARAMETERS OF SIGNAL DETECTION THEORY - A DIRECT SOLUTION [J].
DORFMAN, DD ;
ALF, E .
PSYCHOMETRIKA, 1968, 33 (01) :117-&
[9]   A METHOD OF COMPARING THE AREAS UNDER RECEIVER OPERATING CHARACTERISTIC CURVES DERIVED FROM THE SAME CASES [J].
HANLEY, JA ;
MCNEIL, BJ .
RADIOLOGY, 1983, 148 (03) :839-843
[10]   ALTERNATIVE APPROACHES TO RECEIVER OPERATING CHARACTERISTIC ANALYSES [J].
HANLEY, JA .
RADIOLOGY, 1988, 168 (02) :568-570