Growing Cell Structures

Scientists need to classify and organize complex data, such as that yielded by medical tests or analysis via GC-MS (gas chromatography-mass spectrometry). The data may be multifaceted and difficult to interpret, as different tests may conflict or yield inconclusive results. Growing cell structures may be used to assess medical data for example, such as that obtained from patient biopsies, and determine whether the test results are consistent with a diagnosis of breast cancer.1... 

Walker, A.K., Cross, S.S., and Harrison, R.F., Visualisation of biomedical datasets by use of growing cell structure networks A novel diagnostic classification technique, Lancet, 354, 1518,1999. 

A method exists that largely overcomes the problems of computational expense and uncertainty in the size of the map. This is the growing cell structure algorithm, which we explore in this chapter. 

Similar aims underlie the growing cell structure (GCS) approach, which relies on the use of lines, triangles, or, in more general terms, "dimensional hypertetrahedrons," (which, as we shall see, are far easier to use than the name suggests). 

The growing cell structure algorithm is a variant of a Kohonen network, so the GCS displays several similarities with the SOM. The most distinctive feature of the GCS is that the topology is self-adaptive, adjusting as the algorithm learns about classes in the data. So, unlike the SOM, in which the layout of nodes is regular and predefined, the GCS is not constrained in advance to a particular size of network or a certain lattice geometry. 

The building blocks of one-, two-, and three-dimensional growing cell structures. 

Wong, J.W.H. and Cartwright, H.M., Deterministic projection by growing cell structure networks for visualization of high-dimensionality datasets. /. Biomed. Inform., 38,322, 2005. 

Includes an introduction to artificial intelligence, artificial neural networks, self-organizing maps, and growing cell structures... 

Just as there are several varieties of evolutionary algorithm, so the neural network is available in several flavors. We shall consider feedforward networks and, briefly, Kohonen networks and growing cell structures, but Hop-field networks, which we shall not cover in this chapter, also find some application in science.31... 

Biomedical Datasets by Use of Growing Cell Structure Networks A Novel Diagnostic Classification Technique. 

Projection of Growing Cell Structure Networks for Visualization of High-Dimensionality... 

Fritzke, B. (1994) Growing cell structures-a self-organizing network for unsupervised and supervised learning. Neural Networks 7 1441-1460... 

---