Title :
Neural network design for optimization of the partial area under the receiver operating characteristic curve
Author :
Sahiner, Berkman ; Chan, Heang-Ping ; Petrick, Nicholas ; Gopal, S. Sanjay ; Goodsitt, Mitchell M.
Author_Institution :
Dept. of Radiol., Michigan Univ., Ann Arbor, MI, USA
Abstract :
A new backpropagation training algorithm was developed for the maximization of the area under the receiver operating characteristic (ROC) curve between two user-specified true-positive fraction thresholds. The algorithm was used to design a neural network classifier with high specificity at the high-sensitivity region of the ROC curve, which is of particular interest for computer-aided diagnosis applications. The effectiveness of the algorithm was demonstrated with a simulation study
Keywords :
backpropagation; medical diagnostic computing; neural nets; optimisation; pattern classification; ROC curve; backpropagation; computer-aided diagnosis; neural classifier; optimization; positive fraction thresholds; receiver operating characteristic; Algorithm design and analysis; Application software; Area measurement; Backpropagation algorithms; Computer aided diagnosis; Costs; Design optimization; Electronic mail; Neural networks; Radiology;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.614545