Feature Level Fusion for Brain Stroke Detection in MRI-CT Images

  • Abstract
  • Keywords
  • References
  • PDF
  • Abstract

    Multimodal imaging presently assumes a critical part in analytic as single modalities not ready to give adequate data to the demonstrative reason. In this work we utilize two correlative modalities figured tomography (CT) and attractive reverberation imaging (X-ray) for mind stroke location and division. The force similitudes between cerebrum injuries and some ordinary tissues result in disarray inside division calculations, particularly in the database of genuine MR pictures. The framework execution for this database, a multi-ghostly approach in light of highlight level combination is exhibited in proposed calculation. We separated two unique highlights DWT and LBP and circuit these two highlights utilizing CCA based approach. These highlights are mostly surface highlights, which are fit for catching picture data at neighborhood and worldwide levels. Neighborhood level surface data is caught utilizing Nearby Double Example (LBP) and worldwide level surface data is caught utilizing Wavelets. Despite the fact that utilizing multi-ghostly X-ray has a few disadvantages and constraints, since it makes utilization of integral data, it builds the precision of the framework.


    Here, an element level combination procedure in light of sanctioned connection examination (CCA) is proposed, CCA is connected for joining X-ray and CT successions. To portion tumors, despite the fact that information combination increments computational many-sided quality of the division calculation, and it results in a higher precision. Despite the fact that information combination increments computational multifaceted nature of the division calculation, it results in a higher precision (93.3%), affectability (94.2%), and F1 Score (93.76). In the wake of coordinating use k-implies closest grouping for Division was found as k=2 for unusual stroke and typical district of stroke.



  • Keywords

    DWI, LBP, CCA, SVD, K-means, DWT, ADC, Image fusion.

  • References

      [1] D Le Bihan, E Breton, D Lallemand, P Grenier,E Cabanis and M Laval-Jeantet. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology, vol. 161, no. 2, pages 401–7,Nov 1986.

      [2] Moseley, J. Kucharczyk, J. Mintorovitch, Y. Cohen,J. Kurhanewicz, N. Derugin, H. Asgari and D. Norman.Diffusion-weighted MR imaging of acute stroke: correlation with T2-weighted and magnetic usceptibility-enhanced MR imaging in cats. American Journal of Neuroradiology, vol. 11, no. 3, pages 423–429, 1990.

      [3] C. Rosso, N. Hevia-Montiel, S. Deltour, E. Bardinet,D. Dormont, S. Crozier, S. Baillet and Y. Samson. Prediction of infarct growth based on apparent diffusion coefficients: penumbral assessment without intravenous contrast material. Radiology, vol. 250, no. 1, pages 184–92, Jan 2009.

      [4] Frederik Maes, Andre Collignon, Dirk Vandermeulen, Guy Marchal and Paul Suetens, “Multi -Modality Image Registration by Maximzation of Mutual Information”, IEEE Transactions on Medical Imaging, Vol.16, NO.2, APRIL 1997.

      [5] Amarjot Singh, Srikrishna Karanam, Shivesh Bajpai, Akash Choubey, Thaluru Raviteja, “Malignant Brain Tumor Detection”, in 4th IEEE International Conference on Computer Science and Information Technology (ICCSIT 2011) (IEEE Xplore) Vol. 1, pp. 163-167, 2011.

      [6] Ojala, T., Pietikäinen, M. and Mäenpää, T. (2002), Multiresolution Gray-scale and Rotation Invariant Texture Classification with Local Binary Patterns. IEEE Trans. Pattern Analysis and Machine Intelligence 24(7): 971-987. [7] Tingkai Sun, Songcan Chen; Locality preserving CCA with applications to data visualization and pose estimation; Image and Vision Computing 25 (2007) 531–543.

      [7] Kenya Murase, Masaaki Shinohara, and Youichi Yamazaki; “Accuracy of deconvolution analysis based on singular value decomposition for quantification of cerebral blood flow using dynamic susceptibility contrast-enhanced magnetic resonance imaging; Phys. Med. Biol. 46 (2001) 3147–3159

      [8] N.Hemarajini, R.Bhavani, “Automatic Detection and Classification of Ischemic Stroke Using K-Means Clustering and Texture Features,” Emerging Technologies in Intelligent Applications for Image and Video Processing IGI Global, 2016




Article ID: 19262
DOI: 10.14419/ijet.v7i3.27.19262

Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.