ssmuБюллетень сибирской медициныBulletin of Siberian Medicine1682-03631819-3684Siberian State Medical University, the Ministry of Healthcare of the Russian Federation10.20538/1682-0363-2014-4-99-107ssmu-121Research ArticleОРИГИНАЛЬНЫЕ СТАТЬИORIGINAL PAPERSПРОГРАММНОЕ ОБЕСПЕЧЕНИЕ ДЛЯ ПОИСКА ОБЛАСТЕЙ ИНТЕРЕСА В ТРЕХМЕРНЫХ МЕДИЦИНСКИХ ИЗОБРАЖЕНИЯХSOFTWARE FOR REGIONS OF INTEREST RETRIEVAL ON MEDICAL 3D IMAGESСтромовГ. Г.StromovG. G.

Стромов Глеб Геннадьевич – аспирант кафедры медицинской и промышленной электроники

Stromov Gleb G.

ryzhkoff.d.v@gmail.comРыжковД. В.RyzhkovD. V.

Рыжков Дмитрий Владимирович – аспирант кафедры медицинской и промышленной электроники

Ryzhkov Dmitriy V.

ryzhkoff.d.v@gmail.comФокинВ. А.FokinV. A.

Фокин Василий Александрович – доктор технических наук, профессор кафедры медицинской и биологической кибернетики

Fokin Vasiliy A.

ryzhkoff.d.v@gmail.comНациональный исследовательский Томский политехнический университет, ТомскРоссияNational Research Tomsk Polytechnic University, TomskRussian FederationСибирский государственный медицинский университет, ТомскРоссияSiberian State Medical University, TomskRussian Federation20142808201413499107Copyright © Стромов Г.Г., Рыжков Д.В., Фокин В.А., 20142014Стромов Г.Г., Рыжков Д.В., Фокин В.А.Stromov G.G., Ryzhkov D.V., Fokin V.A.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://bulletin.ssmu.ru/jour/article/view/121

Цель исследования – реализация программного комплекса для организации исследований по поиску областей интереса в трехмерных медицинских изображениях и реконструкции морфологического субстрата и его апробация на примере анализа большого объема модельных МРТ-снимков.

Материал и методы

Материал и методы. Программный комплекс апробирован на модельных МРТ-снимках головного мозга, содержащих морфологический субстрат (проявление патологии – рассеянного склероза в тяжелой стадии), предоставляемых ресурсом BrainWeb. Технологический стек, выбранный для реализации предложенной схемы, базируется на кроссплатформенных решениях: для организации долговременного хранилища данных и поддержания информации в согласованном состоянии использована система управления базами данных MariaDB (open-source ветвь MySQL) и процедурное расширение SQL. Для автоматизации рутинных инфраструктурных задач использован Python версии 2.7. Расчетный модуль написан на языке программирования Java 7 с использованием библиотеки классов SpringFramework 3 и MongoDB как средства обмена данными между узлами в кластере. Версионирование кодовой базы основано на git, в качестве сборщика использован Maven 3.

Результаты

Результаты. Полученные при тестировании программы в инфраструктуре кафедры медицинской и биологической кибернетики СибГМУ результаты исследований доказывают принципиальную возможность применения технологии автоматизированного поиска областей интереса и реконструкции морфологического субстрата в трехмерных изображения МРТ на основе обобщенного анализа различий между референтной и оцениваемой группой снимков.

Background. Implementation of software for areas of interest retrieval in 3D medical images is described in this article. It has been tested against large volume of model MRIs.Material and methods. We tested software against normal and pathological (severe multiple sclerosis) model MRIs from tge BrainWeb resource. Technological stack is based on open-source cross-platform solutions. We implemented storage system on Maria DB (an open-sourced fork of MySQL) with P/SQL extensions. Python 2.7 scripting was used for automatization of extract-transform-load operations. The computational core is written on Java 7 with Spring framework 3. MongoDB was used as a cache in the cluster of workstations. Maven 3 was chosen as a dependency manager and build system, the project is hosted at Github.Results. As testing on SSMU's LAN has showed, software has been developed is quite efficiently retrieves ROIs are matching for the morphological substratum on pathological MRIs.Conclusion. Automation of a diagnostic process using medical imaging allows to level down the subjective component in decision making and increase the availability of hi-tech medicine. Software has shown in the article is a complex solution for ROI retrieving and segmentation process on model medical images in full-automated mode.We would like to thank Robert Vincent for great help with consulting of usage the BrainWeb resource.

МРТрассеянный склерозобласть интересаBrainWebjavanosqlMRIsevere multiple sclerosisinterest retrievalBrainWebjavanosqlRobert VincentRobert VincentReferencesСтромов Г.Г., Фокин В.А., Евтушенко Г.С. Интегральная оценка трехмерных биомедицинских изображений с использованием технологии распределенных вычислений // Биотехносфера, 2012. № 3–4. С. 68–72.Stromov G.G., Fokin V.A., Yevtushenko G.S. Integral'naja ocenka trehmernyh biomedicinskih izobrazhenij s ispol'zovaniem tehnologii raspredelennyh vychislenij [The integral estimation of the three-dimensional biomedical images using distributed computing technologies]. Biotehnosfera – Biotechnosphere, 2012, no. 3–4, pp. 68–72.Фокин В.А. Технология интегральной оценки состояния биомедицинских систем // Системы управления и информационные технологии. 2008. № 1.1 (31). С. 191–194.Fokin V.A. Tehnologija integral'noj ocenki sostojanija bio-medicinskih system [Technology of integrated assessment of biomedical systems]. Sistemy upravleniya i informatsionnye tekhnologii – Management Systems and information technol-ogies, 2008, no. 1.1 (31), pp. 191–194.URL: http://brainweb.bic.mni.mcgill.ca/brainweb/ (дата обращения: 07.01.2014).URL: http://brainweb.bic.mni.mcgill.ca/brainweb/ (accessed 07.01.2014).Shi and Zhang. Fast network centrality analysis using GPUs // BMC Bioinformatics. 2011. 12. 149.Shi and Zhang. Fast network centrality analysis using GPUs. BMC Bioinformatics, 2011, 12, 149.Blazewicz et al. Protein alignment algorithms with an efficient backtracking routine on multiple GPUs // BMC Bioinformatics. 2011. 12. 181.Blazewicz et al. Protein alignment algorithms with an efficient backtracking routine on multiple GPUs. BMC Bioinformatics, 2011, 12, 181.Scharfe et al. Fast multi-core based multimodal registration of 2D cross-sections and 3D datasets // BMC Bioinformatics. 2010. 11. 20.Scharfe et al. Fast multi-core based multimodal registration of 2D cross-sections and 3D datasets. BMC Bioinformatics, 2010, 11, 20.Shi et al. Parallel mutual information estimation for inferring gene regulatory networks on GPUs // BMC Research Notes. 2011. 4. 189.Shi et al. Parallel mutual information estimation for inferring gene regulatory networks on GPUs. BMC Research Notes, 2011, 4, 189.Zhang et al. Developing a multiscale, multi-resolution agent-based brain tumor model by graphics processing units // Theoretical Biology and Medical Modelling. 2011. 8. 46.Zhang et al. Developing a multiscale, multi-resolution agent-based brain tumor model by graphics processing units. Theo-retical Biology and Medical Modelling, 2011, 8, 46.Wan et al. High-performance blob-based iterative three-dimensional reconstruction in electron tomography using multi-GPUs // BMC Bioinformatics. 2012. 13 (Suppl 10). S4.Wan et al. High-performance blob-based iterative three-dimensional reconstruction in electron tomography using multi-GPUs. BMC Bioinformatics, 2012, 13, (Suppl 10), S4.Bisset et al. High-performance biocomputing for simulating the spread of contagion over large contact networks // BMC Genomics. 2012. 13 (Suppl 2). S3.Bisset et al. High-performance biocomputing for simulating the spread of contagion over large contact networks. BMC Genomics, 2012, 13 (Suppl 2), S3.Su et al. Parallel-META: efficient metagenomic data analysis based on high-performance computation // BMC Systems Biology. 2012. 6 (Suppl 1). S16.Su et al. Parallel-META: efficient metagenomic data analysis based on high-performance computation. BMC Systems Biology, 2012, 6 (Suppl 1), S16.Lewis et al. Hydra: a scalable proteomic search engine which utilizes the Hadoop distributed computing framework // BMC Bioinformatics. 2012. 13. 324.Lewis et al. Hydra: a scalable proteomic search engine which utilizes the Hadoop distributed computing framework. BMC Bioinformatics, 2012, 13, 324.Taylor. An overview of the Hadoop/MapReduce/HBase framework and its current applications in bioinformatics // BMC ioinformatics. 2010. 11 (Suppl 12). S1.Taylor. An overview of the Hadoop/MapReduce/HBase framework and its current applications in bioinformatics. BMC ioinformatics, 2010, 11 (Suppl 12), S1.Srinivasan A., Faruquie T., Joshi S. Data and task parallelism in ILP using MapReduce // Machine Learning. 2012. 86. 141–168.Srinivasan A., Faruquie T., Joshi S. Data and task parallelism in ILP using MapReduce. Machine Learning, 2012, 86, 141–168.Griffith et al. Scenario driven data modelling: a method for integrating diverse sources of data and data streams // BMC Bioinformatics. 2011. 12 (Suppl 10). S17.Griffith et al. Scenario driven data modelling: a method for integrating diverse sources of data and data streams. BMC Bioinformatics, 2011, 12 (Suppl 10), S17.Saxena et al. Implementation of a secure genome sequence search platform on public cloudleveraging open source solutions // Journal of Cloud Computing: Advances, Systems and Applications. 2012. 1. 14.Saxena et al. Implementation of a secure genome sequence search platform on public cloudleveraging open source solu-tions. Journal of Cloud Computing: Advances, Systems and Applications, 2012, 1, 14.Borozan et al. CaPSID: A bioinformatics platform for computational pathogen sequence identification in human genomes and transcriptomes // BMC Bioinformatics. 2012. 13. 206.Borozan et al. CaPSID: A bioinformatics platform for com-putational pathogen sequence identification in human genomes and transcriptomes. BMC Bioinformatics, 2012, 13, 206.Llado X. et al. Automated detection of multiple sclerosis lesions in serial brain MRI // Neuroradiology. 2012. Vol. 54. P. 787–807.Llado X. et al. Automated detection of multiple sclerosis le-sions in serial brain MRI. Neuroradiology, 2012, vol. 54, pp.787–807.Hakulinen U. et al. Repeatability and variation of region-of-interest methods using quantitative diffusion tensor MR imaging of the brain // BMC Medical Imaging. 2012. 13. 30.Hakulinen U. et al. Repeatability and variation of region-of-interest methods using quantitative diffusion tensor MR im-aging of the brain. BMC Medical Imaging, 2012, 13, 30.Udupa K.J. et al. Multiple sclerosis lesion quantification using fuzzy-connectedness principles // Medical Imaging. 1997. V. 16, Iss. 5. P. 598–609.Udupa K.J. et al. Multiple sclerosis lesion quantification using fuzzy-connectedness principles. Medical Imaging, 1997, vol. 16, iss. 5, pp. 598–609.Wells M. et al. Fully automated segmentation of multiple sclerosis lesions in multispectral MRI // Pattern Recognition and Image Analysis. 2008. V. 18, Iss. 2. P. 347–350.Wells M. et al. Fully automated segmentation of multiple sclerosis lesions in multispectral MRI. Pattern Recognition and Image Analysis, 2008, vol. 18, iss. 2, pp 347–350.

The authors declare that there are no conflicts of interest present.