Computational Intelligence Techniques for Classification in Microarray Analysis - Computational Intelligence in Healthcare: Advanced Methodologies - page 297

Fig. 8. Error rate for the classification process related to the iterations.
it to perform new classifications, providing a much appreciated decision support tool
for doctors. As demonstrated, the proposed system reduces the dimensionality based
on the filtering of genes with little variability and those that do not allow a separation
of individuals due to the distribution of data. It also presents a clustering technique
based on the neuronal network ESOINN, which is validated with a PAM technique.
Finally, the system incorporates a technique for knowledge extraction and presents it
to the human experts in a very intuitive format.
With the results obtained from empirical studies we can conclude that the CBR
system presented in this study provides a tool that detects genes and probes, which are
the most important factor for the detection of pathology, and facilitates a classification
and reliable diagnosis, as shown by the results presented in this chapter. The system
has been applied to classify CLL leukemia patients and allows the human expert to
obtain information about the classification process and to identify the probes consid-
ered as important or irrelevant for further classifications. Taking into account these
results, we can conclude that the incorporation of computational intelligence tech-
niques in the expression analysis can facilitate the working day of the care personnel
and provide a robust and reliable decision support tool for the prevention and detec-
tion of cancerous patterns.
Acknowledgments. Special thanks to the Institute of Cancer of Salamanca for the
information and technology provided.
References
1. Tu, Y.J., Zhou, W., Piramuthu, S.: Identifying RFID-embedded objects in pervasive
healthcare applications. Decision Support Systems 46(2), 586-593 (2008)
2. Chakraborty, D., Takahashi, H., Suganuma, T., Takeda, A., Kitagata, G., Hashimoto, K.,
Shiratori, N.: Context-aware remote healthcare support system based on overlay network.
WSEAS Transactions on Computers 7(9), 1505-1514 (2008)