Title :
Classification method for degree of lung adenocarcinoma differentiation
Author :
Murakami, Naoki ; Kanako, Toshiyuki ; Tanaka, Toshiyuki
Author_Institution :
Sch. of Fundamental Sci. & Technol., Keio Univ., Yokohama, Japan
Abstract :
The number of fatalities from lung cancer accounts for 17% of that from all cancer, and is the highest ratio. Of them, the ratio of adenocarcinoma which has the highest ratio of lung cancer is increasing yearly. On the other hands, a classification of degree of differentiation is important to estimate prognosis, to determine the most suitable remedy and to investigate the relationship between smokers and patients of adenocarcinoma. Then we proposed new method for automatically classifying degree of adenocarcinoma differentiation. In this paper, we show the effectiveness of our method with results of classification.
Keywords :
cancer; differentiation; image classification; lung; medical image processing; neural nets; lung adenocarcinoma differentiation degree classification; lung cancer; neural network topology; Accuracy; Biological neural networks; Cancer; Cavity resonators; Correlation; Feature extraction; Lungs; case classification; image processing; lung cancer; neural network;
Conference_Titel :
SICE Annual Conference (SICE), 2011 Proceedings of
Conference_Location :
Tokyo
Print_ISBN :
978-1-4577-0714-8