Elements of information theory
Elements of information theory
Feature Selection via Discretization
IEEE Transactions on Knowledge and Data Engineering
Feature Selection Via Mathematical Programming
INFORMS Journal on Computing
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Breast cancer detection by Michaelis-Menten constants via linear programming
Computer Methods and Programs in Biomedicine
Non-linear analysis of EEG signals at various sleep stages
Computer Methods and Programs in Biomedicine
Hierarchical Classifier Design Using Mutual Information
IEEE Transactions on Pattern Analysis and Machine Intelligence
Using mutual information for selecting features in supervised neural net learning
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
We present a method for the comparative analysis of parameter groups according to their correlation to disease. The theoretical basis of the proposed method is Information Theory and Nonparametric Statistics. Normalized mutual information is used as the measure of correlation between parameters, and statistical conclusions are based on ranking. The fluorescence polarization (FP) parameter is considered as the principal diagnostic characteristic. The FP was measured in fluorescein diacetate (FDA)-stained individual peripheral blood mononuclear cells (PBMC), derived from healthy subjects and breast cancer (BC) patients, under different stimulation conditions: by tumor tissue, the mitogen phytohemagglutinin (PHA) or without the stimulants. The FP parameters were grouped according to their correlation with breast cancer. It was established that the greatest difference between cells of BC patients and healthy subjects is found in the PHA test (parameter P1).