Stroke Tissue Pattern Recognition Based on CT Texture Analysis

Magdalena Jasionowska , Artur Nowakowski , Grzegorz Ostrek , Artur Przelaskowski , Kazimierz Szopiński

Abstract

The main objective of this paper is a texture-based solution to the problem of acute stroke tissue recognition on computed tomography images. Our proposed method of early stroke indication was based on two fundamental steps: (i) segmentation of potential areas with distorted brain tissue (selection of regions of interest), and (ii) acute stroke tissue recognition by extracting and then classifying a set of well-differentiating features. The proposed solution used various numerical image descriptors determined in several image transformation domains: 2DFourier domain, polar 2D Fourier domain, and multiscale domains (i.e., wavelet, complex wavelet, and contourlet domain). The obtained results indicate the possibility of relatively effective detection of early stroke symptoms in CT images. Selected normal or pathological blocks were classified by LogitBoost with the accuracy close to 75% with the use of our adjusted cross-validation procedure.
Author Magdalena Jasionowska WEiTI
Magdalena Jasionowska,,
- Faculty of Electronics and Information Technology
, Artur Nowakowski ZZIMN
Artur Nowakowski,,
- Department of Applied Computer Science and Computation Methods
, Grzegorz Ostrek WMiNI
Grzegorz Ostrek,,
- Faculty of Mathematics and Information Science
, Artur Przelaskowski WMiNI
Artur Przelaskowski,,
- Faculty of Mathematics and Information Science
, Kazimierz Szopiński
Kazimierz Szopiński,,
-
Pages81-90
Publication size in sheets0.5
Book Burduk Robert, Jackowski Konrad, Kurzyński Marek, Woźniak Michał, Żołnierek Andrzej (eds.): Proceedings of the 9th International Conference on Computer Recognition Systems CORES 2015, Advances in Intelligent Systems and Computing, vol. 403, 2016, Springer International Publishing, ISBN 978-3-319-26225-3, [978-3-319-26227-7], 855 p.
Keywords in EnglishHypodensity recognition, Stroke recognition, Multiscale domain
Abstract in PolishArtykuł opisywał doskonaloną metodę automatycznej detekcji tkanki mózgowia objętej udarem niedokrwiennym na podstawie cech tekstury obrazów badań tomografii komputerowej. Przetestowane zostały dotychczas wykorzystywane cechy teksturowe oraz obliczone z użyciem falek kierunkowych. Wykorzystana została metoda walidacji krzyżowej leave-one-patient-out dostosowana do niezrównoważonych zbiorów. Uzyskano, z użyciem klasyfikatora LogitBoost, dokładność klasyfikacji regionów patologicznych i norm na poziomie 75%
DOIDOI:10.1007/978-3-319-26227-7_8
URL http://www.springer.com/us/book/9783319262253
Languageen angielski
Score (nominal)0
Citation count*0
Cite
Share Share



* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back