Neural Networks in Automatic Diagnosis Malignant Brain Tumors.

Conference PaperinLecture Notes in Computer Science · January 1999with4 Reads
Impact Factor: 0.51 · DOI: 10.1007/BFb0100545 · Source: DBLP
Conference: Engineering Applications of Bio-Inspired Artificial Neural Networks, International Work-Conference on Artificial and Natural Neural Networks, IWANN '99, Alicante, Spain, June 2-4, 1999, Proceedings, Volume II
Abstract

Automatic Proton Nuclear Magnetic Resonance (1H NMR) spectral based diagnosis malignant brain tumors is analyzed, using Neural Networks for classifying purposes. The Pattern Recognition has been split into its principal parts or subproblems, adapting each one of them to the special task of analyzing 1H NMR spectra. We study the principal algorithms needed for solving those problems, and finally a distributed object oriented classifying system is proposed, which attempts to solve the principal problems mentioned in each section.