DocumentCode :
2524392
Title :
PULMONARY NODULE CLASSIFICATION: SIZE DISTRIBUTION ISSUES
Author :
Jirapatnakula, A.C. ; Reevesa, A.P. ; Apanasovichb, T.V. ; Biancardia, A.M. ; Yankelevitz, D.F. ; Henschkec, C.I.
Author_Institution :
Sch. of Electr. & Comput. Eng., Cornell Univ., Ithaca, NY
fYear :
2007
fDate :
12-15 April 2007
Firstpage :
1248
Lastpage :
1251
Abstract :
Automated nodule classification systems determine a model based on features extracted from documented databases of nodules. These databases cover a large size range and have an unequal distribution of malignant and benign nodules, leading to a high correlation between malignancy and size. For two recent studies in the literature, much of the reported performance of the system may be derived from size based on analysis of their size distributions. We performed experiments to determine the effect of unequal size distribution on a nodule classification system´s performance. Preliminary results indicate that the performance across the entire dataset (a sensitivity/specificity of 0.85/0.80) does not generalize to a subset of nodules (0.50/0.80), but performance can be improved by specifically training on that subset (0.60/0.80). Additional testing with larger datasets needs to be performed, but results reported in this area are overly optimistic.
Keywords :
feature extraction; image classification; medical image processing; physiological models; pneumodynamics; feature extraction; pulmonary nodule classification; size distribution; Biomedical engineering; Cancer; Computed tomography; Educational institutions; Feature extraction; Industrial engineering; Lesions; Operations research; Sensitivity and specificity; Spatial databases;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Imaging: From Nano to Macro, 2007. ISBI 2007. 4th IEEE International Symposium on
Conference_Location :
Arlington, VA
Print_ISBN :
1-4244-0672-2
Electronic_ISBN :
1-4244-0672-2
Type :
conf
DOI :
10.1109/ISBI.2007.357085
Filename :
4193519
Link To Document :
بازگشت