ISSN 2071-8594

Russian academy of sciences

Editor-in-Chief

Gennady Osipov

V.P. Fralenko, M.V. Khachumov, M.V. Shustova The tools for automatically finding and visualization of interest areas in the MRI data to support of medical researchers decision-making

Abstract.

The article gives a detailed description of the techniques that developed by the authors for primary and deep processing of magnetic resonance imaging, directed on detecting areas ischemic defeat in the brain of rats. The composition tools include techniques to bring MRI images of different samples to the normalized form (size, shape, brightness). Another set of tools associated with the detection of anomalies by T2 and MDC images with using of artificial neural networks and specific metrics. To improve the accuracy of recognition tools are configured into sections based on atlases and anatomical features of the rat brain. Tools providing cognitive illumination of the interest zones, in particular ischemic lesion areas. It is assumed that created algorithms and programs will be part of the developing research software system that oriented to support of medical researchers decision-making.

Keywords:

magnetic resonance tomography, brain, ischemic lesion, image recognition, visualization, metric, convolutional neural network.

PP. 27-37.

REFERENCES

1. NITRC: Robust Brain Extraction (ROBEX): Tool/Resource Info [Elektronnyy resurs]. URL: http://www.nitrc.org/projects/robex (data obrashcheniya: 26.07.2016).
2. BET – FslWiki [Elektronnyy resurs]. URL: tp://fsl.fmrib.ox.ac.uk/fsl/fslwiki/BET ht (data obrashcheniya: 26.07.2016).
3. NiftySeg – TIG [Elektronnyy resurs]. URL: http://cmictig.cs.ucl.ac.uk/wiki/index.php/NiftySeg (data obrashcheniya: 26.07.2016).
4. Cabezas M., Oliver A., Llado X., Freixenet J., Cuadra M.B. A Review of Atlas-based Segmentation for Magnetic Resonance Brain Images // Computer Methods and Programs in Biomedicine. Dec 2011. – P. 314-355.
5. Wei Huang. Automatic Affine and Elastic Registration Strategies for Multi-Dimensional Medical Images // A Dissertation in partial fulfillment of the requirements for the Degree of Doctor of Philosophy, Worcester Polytechnic Institute. - 2007. – 140 p.
6. Khachumov M.V. Algoritmy tselochislennoy arifmetiki v zadachakh rascheta traektornogo dvizheniya // Sbornik nauchnykh trudov II Vserossiyskogo nauchno-prakticheskogo seminara «Bespilotnye transportnye sredstva elementami iskusstvennogo intellekta» (BTS-II-2015) (8-9 oktyabrya 2015 g., Sankt-Peterburg). - 2015. – S. 130-135.
7. Braccini C., Cocurullo F., Lavagetto F. A Fast Algorithm for High Quality Vector Quantization Codebook Design // Proceeding ICIAP '95 Proceedings of the 8th International Conference on Image Analysis and Processing. – Springer-Verlag London, UK. - 1995. – P. 643-648.
8. Abbadi N.K., Dahir N.S., Alkareem Z.A. Skin texture recognition using neural networks // Proc. Int’l Arab Conf. on Information Technology. - 2008. – P. 1-4.
9. Haralick R.M., Shanmugam K., Dinstein I. Textural features for image classification // IEEE Transactions on Systems, Man and Cybernetics. - 1973. Vol.3. – P. 610-621.
10. Amelkin S.A., Zakharov A.V., Khachumov V.M. Obobshchennoe rasstoyanie Yevklida-Makhalanobisa i ego svoystva // Informatsionnye tekhnologii i vychislitelnye sistemy. – 2006. – №4. – S. 40-44.
11. Derek K. Jones. Diffusion MRI: Theory, Methods, and Applications // New York: Oxford University Press. - 2011. – 767 p.
12. Talalaev A.A., Tishchenko I.P., Fralenko V.P., Khachumov V.M. Analiz effektivnosti primeneniya iskusstvennykh neyronnykh setey dlya resheniya zadach raspoznavaniya, szhatiya i prognozirovaniya // Iskusstvennyy intellekt i prinyatie resheniy. – 2008. –№2. – S. 24-33.
13. Khachumov M.V. Primenenie iskusstvennykh neyronnykh setey dlya avtomaticheskoy klassifikatsii stepeni zabolevaniya // Sbornik nauchnykh trudov po materialam III Mezhdunarodnoy nauchno-prakticheskoy konferentsii «Teoreticheskie i prikladnye aspekty sovremennoy nauki» (Belgorod, 30 sentyabrya 2014). Ch.1. – 2014. – S. 203-208.
14. Molodchenkov A.I., Fralenko V.P., Khachumov V.M. Klassifikatsiya stepeni tyazhesti zabolevaniya na osnove iskusstvennykh neyronnykh setey // Vestnik RUDN. Seriya Matematika. Informatika. Fizika. – 2014. – №2. – S. 150-156.
15. LeCun Y., Bengio Y., Hinton G. Deep learning // Nature. - 2015, 521 (7553): P.436-444.
16. Maxout Networks [Elektronnyy resurs]. URL: http://colinraffel.com/wiki/maxout_networks (data obrashcheniya: 26.07.2016).
17. Srivastava N., Hinton G., Krizhevsky A., Sutskever I., Salakhutdinov R. Dropout: A Simple Way to Prevent Neural Networks from Overfitting, 2014 [Elektronnyy resurs]. URL: https://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf (data obrashcheniya: 26.07.2016).
18. GitHub – milakov/nnForge: Convolutional neural networks C++ framework with CPU and GPU (CUDA) backends [Elektronnyy resurs]. URL: https://github.com/milakov/nnForge (data obrashcheniya: 26.07.2016).
19. Lebedev A.S., Fralenko V.P., Chen G.S., Chzhan G.L. Eksperimentalnoe issledovanie zadachi poiska slozhnykh rigidnykh obektov v sisteme obrabotki kosmicheskikh snimkov «PS NSKiD» // Sovremennye problemy nauki i obrazovaniya. – 2015. – №1 [Elektronnyy resurs]. URL: http://www.science-education.ru/pdf/2015/1/1194.pdf (data obrashcheniya: 26.07.2016).
20. Fralenko V.P., Suvorov R.Ye., Ovcharenko R.I., Tikhomirov I.A. Avtomaticheskaya klassifikatsiya izobrazheniy v zadachakh filtratsii kontenta // Informatsionnye tekhnologii i vychislitelnye sistemy. – 2015. – №3. – S. 3-11.
21. Agronik A.Yu., Fralenko V.P. Biblioteka algoritmov vysokoproizvoditelnoy obrabotki dannykh ot sistemy tekhnicheskogo zreniya bespilotnogo letatelnogo apparata // Programmnye sistemy: teoriya i prilozheniya. – 2016. – №2. – S. 61-71.
22. Ciresan D.C., Meier U., Masci J. Gambardella L.M., Schmidhuber J. Flexible, High Performance Convolutional Neural Networks for Image Classification // International Joint Conference on Artificial Intelligence. - 2011. – P. 1237-1242.