Title :
Neural networks trained by a genetic algorithm for visual field diagnosis
Author :
Stefano, G. Di ; Capozza, M. ; Accornero, N.
Author_Institution :
Dipt. di Inf. e Sist., Univ. La Sapienza, Rome, Italy
fDate :
Oct. 29 1992-Nov. 1 1992
Abstract :
A population of feed-forward multi-layer neural networks has been trained by a genetic algorithm in order to produce a neural network able to diagnose visual field defects from digital images obtained with an automatic visual field analyzer. The algorithm has been implemented on a system based on a network of five transputers. The results obtained by two different experiments are discussed.
Keywords :
genetic algorithms; neural nets; patient diagnosis; vision defects; digital images; feed forward multi layer neural networks; genetic algorithm; visual field defect diagnosis; visual field diagnosis; Neurons;
Conference_Titel :
Engineering in Medicine and Biology Society, 1992 14th Annual International Conference of the IEEE
Conference_Location :
Paris
Print_ISBN :
0-7803-0785-2
Electronic_ISBN :
0-7803-0816-6
DOI :
10.1109/IEMBS.1992.5761235
