Metaphase recognition by the backpropagation neural network

Authors

Caesar A. Saloma ⋅ PH National Institute of Physics, University of the Philippines Diliman
Maricor N. Soriano ⋅ PH National Institute of Physics, University of the Philippines Diliman
Glenn F. Mahiya ⋅ PH National Institute of Physics, University of the Philippines Diliman

Abstract

The error-driven backpropagation algorithm for training a neural network is applied to the recognition of metaphase spreads. Images of human blood cells in the metaphase stage together with non metaphase images are used as training samples for the network. The patterns consist of unipolar gray scaled pixels in a 10x10 mesh. Several modifications to the basic architecture and algorithm are implemented such as locally connected models, combination of sigmoid and gaussian activation functions, and varying learning rate.

Downloads

Issue

1994: Proceedings of the 12th National Physics Congress of the Samahang Pisika ng Pilipinas

Article ID

SPP-1994-IP-06

Section

Instrumentation and Computational Physics

Published

1994-10-15

How to Cite

[1]

CA Saloma, MN Soriano, and GF Mahiya, Metaphase recognition by the backpropagation neural network, Proceedings of the Samahang Pisika ng Pilipinas 12, SPP-1994-IP-06 (1994). URL: https://proceedings.spp-online.org/article/view/SPP-1994-IP-06.