Biomedical pattern discovery and visualisation based on self-adaptive neural networks

With the rapid growth of the amount of biomolecular data, there is an increasing need to develop powerful techniques to enhance pattern discovery capabilities of bioscientists and machines. Based on a self-adaptive neural network, this paper presents a new approach to biomolecular pattern identification and visualisation. The method is applied to the classification of leukaemia samples, which are described by their expression profiles. The results indicate that this framework may significantly facilitate and improve pattern discovery and visualisation tasks, in comparison to traditional algorithms such as the Kohonen self-organising map.