DOI: 10.17587/prin.10.125-134

Classification of Sensory Images using Tactile Information Obtained from Mechanoreceptor

V. M. Staroverov, staroverovvl@yandex.ru, M. V. Lomonosov Moscow State University, Faculty of Mechanics and Mathematics, Moscow, 119991, Russian Federation

Corresponding author: Staroverov Vladimir M., Associate Professor, M. V. Lomonosov Moscow State University, Faculty of Mechanics and Mathematics, Moscow, 119991, Russian Federation, Е-mail: staroverovvl@yandex.ru

The article is devoted to the automatic analysis of tactile images obtained with the help of Medical Tactile Endo-surgical Complex (MTEC). The author proposes an algorithm for tactile samples classification that can identify and classify the heterogeneities in the tissues studied. To test the algorithm, the author creates a library of sensor heterogeneities, which consists of silicone parallelepipeds with solid inclusions. On the basis of this library, test samples of clicks was made. In the article the decision on belonging of the tested sample to one of the pre-defined classes is based on the proximity function between the given sample and the samples from the pre-defined classes. The final decision on the corresponding of the sample to the predefined class is made on the basis of some kind of "voting" method. Testing of the algorithm is carried out on the solution of two applied problems: 1) determination of belonging of the tested sample to one of the six types of heterogeneities; 2) determination of the presence of the inhomogeneity of the undefined type in the tested sample. The article deals with heterogeneities based on artificial samples simulating homogeneous tissue, tissue with pathologic inclusions and tissue with a blood vessel. In the solving the problems the method of cross-validation is actively used in the article. Detailed results of calculations posted on the Internet at the link provided with the article.

Keywords: Artificial tactile sensing, Medical Tactile Endosurgical Complex, library of sensory images

pp. 125–134

For citation:
Staroverov V. M. Classification of Sensory Images using Tactile Information Obtained from Mechanoreceptor, Programmnaya Ingeneria, 2019, vol. 10, no 3, pp. 125—134.

The research was supported by the Russian Science Foundation (project 16-11-00058 "The development of methods and algorithms for automated analysis of medical tactile information and classification of tactile images")