论文信息 - A classifier with the fuzzy Hopfield network for human chromosomes

A classifier with the fuzzy Hopfield network for human chromosomes

This paper focuses on automatic classification of human chromosomes and develops a classifier based on the fuzzy Hopfield network, a dynamic neuro-fuzzy system. The classifier is a neurocomputing model as well as a fuzzy computing model. It holds fuzzy clustering capability, and is able to acquire knowledge about human chromosome data from some samples. Its task is to identify human chromosomes and assign each of them to one of the 24 human chromosome classes (22 autosomes and two sex chromosomes). The research results show that the classifier has a lower unidentified rate and the wrong identification rate of zero. The classifier offers us a promising computerized intelligent machine for automatic analysis of human chromosomes.

Xiaogang Ruan

[1] Xiaogang Ruan,et al. A dynamic neuro-fuzzy system configuration, stability, and fuzzy operational function , 1999, Fuzzy Sets Syst..

[2] Arnold W. M. Smeulders,et al. Human chromosome classification based on local band descriptors , 1989, Pattern Recognit. Lett..

[3] Jim Piper,et al. Automation of chromosome analysis , 1980 .

[4] Boaz Lerner,et al. Toward a completely automatic neural-network-based human chromosome analysis , 1998, IEEE Trans. Syst. Man Cybern. Part B.

[5] J L Simpson,et al. Incidence and timing of pregnancy losses: relevance to evaluating safety of early prenatal diagnosis. , 1990, American journal of medical genetics.