Analysis of current-voltage characteristics of bioactive points is proposed for the classification of socially significant diseases. Morphological integrity requirements can be satisfied by using an automated system that includes data-collection unit connected to a PC, communication unit and corresponding software. Polynomial of the seventh order was used for approximation of the current-voltage characteristics. It permitted to obtain the vector of informative features of bioactive point condition. Biomaterial can be analyzed turning to current-voltage characteristics of a single bioactive point or a cluster of bioactive points. Multi-agent classifiers, based on probabilistic neural networks and fuzzy neural networks, are used to classify biomaterial. The classifiers contain three macro-layers: the first one consists of three-layer probabilistic neural networks, the second and third ones consist of two-layer fuzzy neural networks. The number of modules in macro-layers is equal to the number of differentiable classes of socially significant diseases.