The translaminar neuromorphotopological clustering and classification of the dentate nucleus neurons.

The aim of this study is to determine whether the dentate neurons can be translaminary neuromorphotopologically classified as ventrolateral or dorsomedial type. Adult human dentate 2D binary interneuron images are used for the purposes of the analysis. The analysis is performed on the real and the virtual neuron sample. The total of 29 parameters is used. They can be divided into the classes: neuron surface, shape, length, branching and complexity. The clustering is performed through the algorithm consisted from of the steps of predictor extraction (matrix attractor analysis/non-negative matrix factorization and cluster analysis of predictor factors, separate unifactor analysis/Student's t-test and MANOVA) and multivariate cluster analysis set (cluster analysis, principal component analysis, factor analysis with pro/varimax rotation, Fisher's linear discriminant analysis and feed-forward backpropagation artificial neural networks). The separate unifactor analysis extracted as significant the following predictors: the Npd (p<0.05) on the natural sample and the Adt (p<0.05), Do (p<0.001), Ms (p<0.01), Dwdth (p<0.001), Npd (p<0.05), Nsd (p<0.001), Nt/hod (p<0.001), Nmax (p<0.01), Ds (p<0.001), Cdf(Nt/hod)st (p<0.05) on the virtual one. Considering the multidimensional analysis, except for the Fisher's linear discriminant analysis which gave the false positive result, any other analysis proclaimed the absence of the translaminar dentate neuron classification. The dentate neurons cannot be classified onto the ventrolateral/dorsomedial neuromorphotopological subtypes. Although some differences exist they are not strong enough to carry the classification. The methods of multidimensional statistical analysis are again shown to be the best for such kind of task.