A Multi-Level Thresholding Method based on Histogram Derivatives for Accurate Brain MRI Segmentation

María Gabriela Pérez, Angel Sanchez Calle, Ana Belen Moreno, Eldman Nunes, Víctor Andaluz

Resumen



Resumen: Este trabajo describe la implementación y evaluación cuantitativa de un método automático para la segmentación precisa de imágenes cerebrales de resonancia magnética (RM). El método se basa en la umbralización adaptativa multinivel del histograma de la imagen con el fin de clasificarla en un número variable de clases de tejidos cerebrales. Inicialmente, se realiza una etapa de preproceso para eliminación de ruido y realzado de la imagen. Después, se calcula el histograma de la imagen que es suavizado usando un filtro piramidal. La derivada de dicho histograma se usa para determinar una lista de picos y valles en la correspondiente función. El número de clases de tejidos cerebrales a segmentar se corresponde al número de umbrales buscados en el histograma más uno, y dichos umbrales se determinan usando los valores los valles de la función derivada que minimizan los errores en la clasificación de los píxeles de la imagen. El método propuesto se usó para segmentar cuatro clases de tejidos en las imágenes cerebrales (materia blanca, materia gris, líquido cefalorraquídeo y fondo, respectivamente) correspondientes a un conjunto de imágenes sintéticas de resonancia magnética cerebral obtenidas usando la base de datos BrainWeb. El método propuesto se comparó con otros dos métodos de segmentación implementados: el primero basado en modelos de mezcla de gaussianas y el segundo basado en el algoritmo de las k medias. Nuestra propuesta produjo resultados de clasificación correcta por encima del 95%, que son equivalentes a los de los algoritmos comparados.

 

Abstract: This work describes the implementation and quantitative evaluation of an automatic and accurate brain magnetic resonance image segmentation method. This is based on adaptive multi-level thresholding to classify the images into variable number brain tissue classes of interest. The method includes a denoising and enhancement image preprocessing stage. After that, the image histogram is computed and smoothed using a pyramid filter. Then, this histogram is differenced to determine a list of peaks and valleys (i.e. local minima) on it. As the number of considered tissue classes to segment is the number of searched histogram thresholds plus one, the histogram thresholds were chosen using the values of valleys that minimize the classification errors. The proposed method was used to segment four tissue classes (i.e. white matter, gray matter, cerebrospinal fluid and background, respectively) in a collection of synthetic brain MR slice from BrainWeb database. The method was compared to other two implemented segmentation approaches: one based on Gaussian mixture models and other one based on k-means clustering. Our multi-level thresholding segmentation algorithm produced equivalent correct classification results (above 95%) than the other two compared methods of the literature.


Citas


P.A. Bandettini. “What's new in neuroimaging methods?”. Ann. N.Y. Acad. Sci. 1156, pp. 260–293. New York. USA. 2009.

P. Bright (ed). “Neuroimaging Methods”. InTech. 358 pp, ISBN: 978-953-51-00973. 2012.

E. Pagani, A. Bizzi, F. di Salle, N. de Stefano and M. Filippi. “Basic concepts of advanced MRI techniques”. Journal of the Neurological Sciences. 29(3), pp. 290-295. 2008. ISSN: 0022-510X. DOI: 10.1007/s10072-008-1001-7.

S. Bauer, R. Wiest, L.-P Nolte and M. Reyes. “A survey of MRI-based medical image analysis for brain tumor studies”. Physics in Medicine and Biology. 58(13), pp. 97-129. 2013. ISSN: 0031-9155. DOI:10.1088/0031-9155/58/13/R97

Y. Zhang. M. Brady and S. Smith. “Segmentation of Brain MR Images through a Hidden Markov Random Field Model and the Expectation-Maximization Algorithm”. IEEE Transactions on Medical Imaging. 20(1), pp. 45-57, 2001. ISSN: 0278-0062. DOI: 10.1109/42.906424.

R.A. Heckemann, J.V. Hajnal, P. Aljabar, D. Rueckert and A. Hammers. “Automatic anatomical brain MRI segmentation combining label propagation and decision fusion”. NeuroImage. 33(1), pp. 115–126. 2006. ISSN: 1053-8119. DOI: 10.1016/j.neuroimage. 2010.01.072.

A.T. Thankam and A.K.S. Viji. “Abnormality Segmentation of MRI Brain Images using Fuzzy Nearest Neighbour Approach”. International Journal of Computer Applications 67(19), pp. 6-12. 2013. ISSN: 0975 -8887. DOI: 10.5120/11501-7215.

A. Ortiz, J.M. Gorriz, J. Ramirez and D. Salas-Gonzalez. “Unsupervised Neural Techniques Applied to MR Brain Image Segmentation”. Advances in Artificial Neural Systems. Vol. 2012, 7 pages. 2012. ISSN: 1687-7594. DOI: /10.1155/2012/457590.

A. Zavaljevski, A.P. Dhawan, M. Gaskil, W. Ball and J.D. Johnson. “Multi-level adaptive segmentation of multi-parameter MR brain images”. Computerized Medical Imaging and Graphics. 24(2), pp. 87-98, 2000. ISSN: 0895-6111. DOI: 10.1148/rg.282075083.

P.D. Sathya and R. Kayalvizhi. “Optimal segmentation of brain MRI based on adaptive bacterial foraging algorithm”. Neurocomputing. 74(14-15), pp. 2299-2313. 2011. ISSN: 0925-2312. DOI: 10.1016/j.neucom.2011.03.010.

R.C. Gonzalez and R.E. Woods. “Digital Image Processing”. Prentice Hall, Third Edition, Upper Saddle River, New Jersey, USA. ISBN: 9780131687288. 2008.

M. Sonka, V. Hlavac and R Boyle. “Image Processing, Analysis, and Machine Vision”, Second Edition. PWS Publishers, Pacific Grove, USA. ISBN: 1133593607. 2000.

M. Sezgin and B. Sankur. “Survey over image thresholding techniques and quantitative performance evaluation”. Journal of Electronic Imaging. 13(1), pp. 146–165. 2004. ISSN: 1017-9909. DOI: 10.1117/1.1631315.

D. M. Tsai and Y. H. Chen. “A fast histogram-clustering approach for multilevel thresholding”. Pattern Recognition Letters. 13(4), pp. 245-252, 1992. ISSN: 0167-8655. DOI: 10.1016/0167-8655(92)90075-B.

W. Tsai. “Moment-Preserving Thresholding: A New Approach”. Computer Vision, Graphics and Image Processing. 29, pp. 377-393. 1985. ISSN: 0734-189X. DOI: 10.1145/321119.321123.

P. Bolton et al. “Image Analysis: Filtering and Segmentation” (In French). Ed. Masson. Paris. France. 1995.

G.X. Ritter and J.N. Wilson. “Handbook of computer vision algorithms in image algebra”. CRC Press, Boca Raton, USA. ISBN: 0849326362. 1996.

M. Cheriet, J.N. Said and C.Y. Suen. “A Recursive Thresholding Technique for Image Segmentation”. IEEE Transactions on Image Processing. 7(6), pp. 918-921, 1998. ISSN: 1057-7149. DOI: 10.1109/83.679444.

H. Yan. “Unified Formulation of a Class of Image Thresholding Techniques”. Pattern Recognition. 29(12), pp. 2025-2031. 1996. ISSN: 0031-3203. DOI: 10.1016/S0031-3203(96)00050-7

S. R. AboudNeta, L. V. Dutra and G. J. Erthal. “Automatic Thresholding in Multi-Modal Histograms” (In Portuguese). Proc. 7th Brazilian Conference on Dynamics, Control and Applications. Sao Paulo. 2008.

C.C. Chang and L.L. Wang. “A Fast Multilevel Thresholding Method Based on Lowpass and Highpass Filtering”. Pattern Recognition Letters. 18, pp. 1469-1478. 1997. ISSN: 0167-8655. DOI: 10.1016/S0167-8655(97)00134-7.

E. Fletcher, O. Carmichael, and C. DeCarli. “MRI Non-Uniformity Correction through Interleaved Bias Estimation and B-Spline Deformation with a Template”. Proc. Intl. Conf. IEEE Engineering in Medicine and Biology Society. pp. 106-109. San Diego (USA). 2012.

E. de O. Nunes and G. Perez. “Medical Image Segmentation by Multilevel Thresholding Based on Histogram Difference”. Proc. 17th International Conference on Systems, Signals and Image Processing (IWSSIP „10). pp. 284-287. Rio de Janeiro, Brazil. 2010.

M. Mignotte, J. Meunier, J.-P. Soucy, and C. Janicki. “Segmentation and classification of brain SPECT images using 3D Markov random field and density mixture estimations”. Proc. 5th World Multi-Conf. Systemics, Cybernetics and Informatics. Orlando. USA. Vol. X, pp. 239-244. 2001.

P. Sneath and R. Sokal. “Numerical Taxonomy: The Principles and Practice of Numerical Classification”. Freeman. San Francisco. USA. ISBN: 0-7167-0621-0. 1973.

BrainWeb: Simulated Brain Database. URL: brainweb.bic.mni.mcgill.ca/brainweb. Accessed: 15-04-2014.

C.A. Cocosco, V. Kollokian, R.K.-S. Kwan and A.C. Evans. "BrainWeb: Online Interface to a 3D MRI Simulated Brain Database". NeuroImage. 5(4), p. 425, 1997. ISSN: 1053-8119. DOI: 10.1.1.51.3917.

Statistical Parametric Mapping (SPM). SPM8 Release Notes. 2009. URL: http://www.fil.ion.ucl.ac.uk/spm/software/spm8/Accessed: 15-03-2014.

J. Ashburner and K.J. Friston. “Unified segmentation”. NeuroImage. 26, pp. 839-851. 2005. ISSN: 1053-8119. DOI: 10.1016/j.neuroimage.2005.02.018.

[30] M. A. Balafar, A.R. Ramli, M.I. Saripan and S. Mashohor. “Review of brain MRI image segmentation methods”. Artificial Intelligence Review. 33(3), pp 261-274. 2010. ISSN: 0269-2821. DOI :10.1007/s10462-010-9155-0.

Z. Ji, Y. Xia, Q. Sun, Q. Chen, D. Xia and D.D. Feng. “Fuzzy local Gaussian mixture model for brain MR image segmentation“. IEEE Transactions on Information Technology in Biomedicine. 16(3), pp. 339-347. 2012. ISSN. 1089-7771. DOI: 10.1109/TITB.2012.2185852.

María G. Pérez, “Métodos de eliminación de ruido y segmentación de tejidos cerebrales en imágenes médicas de resonancia magnética nuclear,”Ph.D. dissertation, Dept. of ComputerScience, Rey Juan Carlos Univ., Madrid, Spain, 2012.


Texto completo: PDF

Refbacks

  • No hay Refbacks actualmente.


Creative Commons License
Este trabajo está licenciado bajo la licencia Creative Commons Attribution 3.0 .

El material presente en este sitio es de acceso abierto, es permitido su reproducción siempre y cuando se referencie a esta revista como su autor.