Criteria for Analysis and Comparison of Experimental Data under Conditions of Uncertainty

The paper deals with investigation of the important problem of processing the ophthalmic data on the post-operation status of patients. The groups of patients differ by the type (technology) of fixing the intraocular lenses (IOL). Validity of each type of technology is estimated by computation of criteria for distinction of data between groups. The initial information comprises measurements of several ophthalmic indices. The samples on each index are very short; in each index, as a rule, the samples of patients' groups overlap each other; any probabilistic characteristics of the measuring indices are unknown; any probabilistic characteristics of the measuring errors are also unknown. So, the standard methods of mathematical statistics can be applied only in the formal way and have shown to be inefficient. In contrast, the Hausdorff distance (from the set theory) as the criterion of distinction between two samples (both for one- and, especially, for two-dimensional indices) demonstrated to be reliable to distinct the patient's status. Computations of the Hausdorff distance are valid for any relative location of point sets under comparison.