, Russian Federation
, Russian Federation
Tomsk, Tomsk, Russian Federation
GRNTI 50.07 Теоретические основы вычислительной техники
BBK 3297 Вычислительная техника
This research is devoted to the segmentation of heart and brain anatomical structures. In the study, we present a segmentation algorithm based on the square blocks (superpixels) propagation. The square blocks propagation algorithm checks two criteria. For the first criteria, the current intensity of the pixel is compared to the average intensity of the segmented region. For the second criterion, the intensity difference of the pixels lying on the superpixel sides is compared to the threshold. Once these criteria are successfully checked, the algorithm merges homogeneous superpixels into one region. Then the following superpixels are attached to the final superpixel set. The last step of the proposed method is the spline generation. The spline delineates the borders of the region of interest. The main parameter of the algorithm is the size of a square block. The cardiac MRI dataset of the University of York and the brain tumor dataset of Southern Medical University were used to estimate the segmentation accuracy and processing time. The highest Dice similarity coefficients obtained by the presented algorithm for the left ventricle and the brain tumor are 0.93±0.03 and 0.89±0.07 respectively. One of the most important features of the border detection step is its scalability. It allows implementing different one-dimensional methods for border detection.
square blocks propagation, superpixels, region growing, left ventricle segmentation, brain tumor segmentation
1. Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P. andSüsstrunk, S. 2012. SLIC superpixels compared to stateof-the-art superpixel methods. IEEE Transactions onPattern Analysis and Machine Intelligence. 34, 11 (2012),2274–2281.DOI:https://doi.org/10.1109/TPAMI.2012.120.
2. Adams, R. and Bischof, L. 1994. Seeded Region Growing.IEEE Transactions on Pattern Analysis and MachineIntelligence. 16, 6 (1994), 641–647.DOI:https://doi.org/10.1109/34.295913.
3. Andreopoulos, A. and Tsotsos, J.K. 2008. Efficient andgeneralizable statistical models of shape and appearancefor analysis of cardiac MRI. Medical Image Analysis. 12,3 (Jun. 2008), 335–357.DOI:https://doi.org/10.1016/j.media.2007.12.003.
4. Ballard, D.H. and Brown, C.M. 1982. Computer Vision.
5. Bankman, I.N. 2000. Handbook of Medical Imaging.
6. Cheng, J., Huang, W., Cao, S., Yang, R., Yang, W., Yun,Z., Wang, Z. and Feng, Q. 2015. Enhanced performanceof brain tumor classification via tumor regionaugmentation and partition. PLoS ONE. 10, 10 (2015).DOI:https://doi.org/10.1371/journal.pone.0140381.
7. Cheng, J., Yang, W., Huang, M., Huang, W., Jiang, J.,Zhou, Y., Yang, R., Zhao, J., Feng, Y., Feng, Q. and Chen,W. 2016. Retrieval of Brain Tumors by Adaptive SpatialPooling and Fisher Vector Representation. PLoS ONE. 11,6 (2016).DOI:https://doi.org/10.1371/journal.pone.0157112.
8. Chiverton, J., Wells, K., Lewis, E., Chen, C., Podda, B.and Johnson, D. 2007. Statistical morphological skullstripping of adult and infant MRI data. Computers inBiology and Medicine. 37, 3 (2007), 342–357.DOI:https://doi.org/10.1016/j.compbiomed.2006.04.001.
9. Crommelinck, S., Bennett, R., Gerke, M., Koeva, M.N.,Yang, M.Y. and Vosselman, G. 2017. SLIC Superpixelsfor Object Delineation from UAV Data. ISPRS Annals ofthe Photogrammetry, Remote Sensing and SpatialInformation Sciences (2017), 9–16.
10. Csillik, O. 2017. Fast segmentation and classification ofvery high resolution remote sensing data using SLICsuperpixels. Remote Sensing. 9, 3 (2017).DOI:https://doi.org/10.3390/rs9030243.
11. Dehmeshki, J., Amin, H., Valdivieso, M. and Ye, X. 2008.Segmentation of pulmonary nodules in thoracic CT scans:A region growing approach. IEEE Transactions onMedical Imaging. 27, 4 (2008), 467–480.DOI:https://doi.org/10.1109/TMI.2007.907555.
12. Duong, T.H. and Hoberock, L.L. 2018. DUHO imagesegmentation based on unseeded region growing onsuperpixels. 2018 IEEE 8th Annual Computing andCommunication Workshop and Conference (CCWC) (Jan.2018), 558–563.
13. Isa, N.A.M., Sabarudin, S., Ngah, U.K. and Zamli, K.Z.2005. Automatic detection of breast tumours fromultrasound images using the modified seed based regiongrowing technique.
14. Kamnitsas, K., Ledig, C., Newcombe, V.F.J., Simpson,J.P., Kane, A.D., Menon, D.K., Rueckert, D. and Glocker,B. 2017. Efficient multi-scale 3D CNN with fullyconnected CRF for accurate brain lesion segmentation.Medical Image Analysis. 36, (2017), 61–78.DOI:https://doi.org/10.1016/j.media.2016.10.004.
15. Kang, H.C., Lee, J. and Shin, J. 2016. Automatic fourchamber segmentation using level-set method and splitenergy function. Healthcare Informatics Research. 22, 4(2016), 285–292.DOI:https://doi.org/10.4258/hir.2016.22.4.285.
16. Litjens, G., Kooi, T., Bejnordi, B.E., Setio, A.A.A.,Ciompi, F., Ghafoorian, M., van der Laak, J.A.W.M., vanGinneken, B. and Sánchez, C.I. 2017. A survey on deeplearning in medical image analysis. Medical ImageAnalysis.
17. Oktay, O., Ferrante, E., Kamnitsas, K., Heinrich, M., Bai,W., Caballero, J., Cook, S.A., De Marvao, A., Dawes, T.,O’Regan, D.P., Kainz, B., Glocker, B. and Rueckert, D.2018. Anatomically Constrained Neural Networks(ACNNs): Application to Cardiac Image Enhancementand Segmentation. IEEE Transactions on MedicalImaging. 37, 2 (2018), 384–395.DOI:https://doi.org/10.1109/TMI.2017.2743464.
18. Park, J.G. and Lee, C. 2009. Skull stripping based onregion growing for magnetic resonance brain images.NeuroImage. 47, 4 (2009), 1394–1407.DOI:https://doi.org/10.1016/j.neuroimage.2009.04.047.
19. Pham, D.L., Xu, C. and Prince, J.L. 2000. CurrentMethods in Medical Image Segmentation. Annual Reviewof Biomedical Engineering. 2, 1 (2000), 315–337.DOI:https://doi.org/10.1146/annurev.bioeng.2.1.315.
20. Pinto, A., Alves, V. and Silva, C.A. 2016. Brain TumorSegmentation using Convolutional Neural Networks inMRI Images. IEEE Transactions on Medical Imaging. 35,5 (2016), 1240–1251.DOI:https://doi.org/10.1109/TMI.2016.2538465.
21. Ren, X. and Malik, J. 2003. Learning a classificationmodel for segmentation. Proceedings Ninth IEEEInternational Conference on Computer Vision. 1, c(2003), 10–17 vol.1.DOI:https://doi.org/10.1109/ICCV.2003.1238308.
22. Rogowska, J. 2009. Overview and fundamentals ofmedical image segmentation. Handbook of Medical ImageProcessing and Analysis. 73–90.
23. Roy, S. and Maji, P. 2015. A simple skull strippingalgorithm for brain MRI. ICAPR 2015 - 2015 8thInternational Conference on Advances in PatternRecognition (2015).
24. Saxen, F. and Al-Hamadi, A. 2014. Superpixels for SkinSegmentation. Www-E.Uni-Magdeburg.De. (2014).
25. Soliman, A., Khalifa, F., Elnakib, A., El-Ghar, M.A.,Dunlap, N., Wang, B., Gimel’farb, G., Keynton, R. andEl-Baz, A. 2017. Accurate lungs segmentation on CTchest images by adaptive appearance-guided shapemodeling. IEEE Transactions on Medical Imaging. 36, 1(2017), 263–276.DOI:https://doi.org/10.1109/TMI.2016.2606370.
26. Tang, J.T.J. 2010. A color image segmentation algorithmbased on region growing. Computer Engineering andTechnology (ICCET), 2010 2nd International Conferenceon. 6, (2010), 634–637.DOI:https://doi.org/10.1109/ICCET.2010.5486012.
27. Tsechpenakis, X.H.G. 2013. Medical ImageSegmentation. Advanced Materials Research. i (2013), 1–35. DOI:https://doi.org/10.1201/9781420090413-c10.
28. Wang, L., Pei, M., Codella, N.C.F., Kochar, M., Weinsaft,J.W., Li, J., Prince, M.R. and Wang, Y. 2015. Leftventricle: Fully automated segmentation based onspatiotemporal continuity and myocardium information incine cardiac magnetic resonance imaging (LV-FAST).BioMed Research International. (2015).DOI:https://doi.org/10.1155/2015/367583.
