Automatic segmentation of brain tumors in magnetic resonance imaging

Mascarenhas, Layse Ribeiro; Ribeiro Júnior, Audenor dos Santos; Ramos, Rodrigo Pereira

doi:10.31744/einstein_journal/2020AO4948

Acessibilidade / Reportar erro

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

ORIGINAL ARTICLE • Einstein (São Paulo) 18 • 2020 • https://doi.org/10.31744/einstein_journal/2020AO4948 copy

Automatic segmentation of brain tumors in magnetic resonance imaging

Authorship SCIMAGO INSTITUTIONS RANKINGS

ABSTRACT

Objective

To develop a computational algorithm applied to magnetic resonance imaging for automatic segmentation of brain tumors.

Methods

A total of 130 magnetic resonance images were used in the T1c, T2 and FSPRG T1C sequences and in the axial, sagittal and coronal planes of patients with brain cancer. The algorithms employed contrast correction, histogram normalization and binarization techniques to disconnect adjacent structures from the brain and enhance the region of interest. Automatic segmentation was performed through detection by coordinates and arithmetic mean of the area. Morphological operators were used to eliminate undesirable elements and reconstruct the shape and texture of the tumor. The results were compared with manual segmentations by two radiologists to determine the efficacy of the algorithms implemented.

Results

The correlated correspondence between the segmentation obtained and the gold standard was 89.23%.

Conclusion

It is possible to locate and define the tumor region automatically with no the need for user interaction, based on two innovative methods to detect brain extreme sites and exclude non-tumor tissues on magnetic resonance images.

Diagnostic imaging; Brain neoplasms; Image processing, computer-assisted; Magnetic resonance imaging; Computer simulation

Step 1 Pre-processing	Step 2 Segmentation	Step 3 Post-processing
Histogram equalization	MDC	Dilation
Intensity adjustment	Erosion	Hole filling
Binarization	MDA	Spatial convolution

Sequence	GS	TP	FP	FN	HR %	MR
T1c	33	27	11	6	81.81	0.65
T2	48	44	19	3	91.66	0.72
FSPGR T1c	49	45	21	4	91.83	0.70
Total	130	116	51	13	89.23	0.70

Authors	Methodology	Performance assessment
Authors	Methodology	HR (%)	MR
Alegro Mde C, Amaro Junior E, Lopes Rde D. Computerized brain tumor segmentation in magnetic resonance imaging. einstein (Sao Paulo). 2012;10(2):158-63. ⁽²³⁾	SVM	94.00	-0.04
Deng W, Luo L, Lin X, Fang T, Liu D, Dan G, et al. Head and neck cancer tumor segmentation using support vector machine in dynamic contrast-enhanced MRI. Contrast Media Mol Imaging. 2017;2017:8612519. ⁽²⁴⁾	SVM	86.00	0.89
Singh R, Agarwal P, Bhattacharya M. MR brain tumor detection employing Laplacian Eigen maps and kernel support vector machine. In: 2016 IEEE International Conference On Bioinformatics And Biomedicine (BIBM). IEEE. 2016;827-30. ⁽²⁷⁾	FCM	72.80	0.43
Hsieh TM, Liu YM, Liao CC, Xiao F, Chiang IJ, Wong JM. Automatic segmentation of meningioma from non-contrasted brain MRI integrating fuzzy clustering and region growing. BMC Medical Informatics and Decision Making. 2011;11:54. ⁽²⁸⁾	Kernel + SVM (K-SVM)	87.80	0.85
Gao F, Lin T. Application of Computer-Aided Diagnosis Technology in Brain Tumour Detection. NeuroQuantology. 2018;16(5):725-33. ⁽²⁹⁾	Machine learning	97.17	-

Authors	Methodology	Results
Isa IS, Sulaiman SN, Mustapha M, Karim NK. Automatic contrast enhancement of brain MR images using Average Intensity Replacement based on Adaptive Histogram Equalization (AIR-AHE). Biocybernetics and Biomedical Engineering. 2017;37(1):24-34. ⁽¹⁵⁾	Development of the AIR-AHE algorithm for automatic contrast enhancement of FLAIR MRI using the imadjust and stretchlim functions	The method showed good results along with other histogram equalization algorithms
Isselmou AE, Zhang S, Xu G. A novel approach for brain tumor detection using MRI Images. J Biomedical Sci Eng. 2016;9(10):44-52. ⁽¹⁷⁾	Use of the high-pass median filter and histogram equalization to improve image quality	The method improved image quality and provided excellent tumor segmentation results
Sujan M, Alam N, Noman SA, Islam MJ. A Segmentation based Automated System for Brain Tumor Detection. IJCA. 2016;153(10):41-9. ⁽³²⁾	Brain extraction on FLAIR images using MATLAB ^® morphological operations, such as binarization, erosion, dilation and structuring elements	The method contributed to better image enhancement results, and greater accuracy when compared with other algorithms in the literature
Roy S, Maji P. A simple skull stripping algorithm for brain MRI. In: Eighth International Conference On Advances In Pattern Recognition (ICAPR) [Internet]. Kolkata: (IN); 2015 [cited 2019 Aug 20]. Available from: https://ieeexplore.ieee.org/abstract/document/7050671/ ⁽³³⁾	Development of a method for skull stripping (S3) on T1-weighted images based on brain anatomy and image intensity	The performance of the algorithm is compared to BET and BSE, with satisfactory results
khandelwal P, Kaur G. Comparative study of different image enhancement technique. IJECT. 2016;7(2):116-21. ⁽³⁴⁾	Comparison of contrast enhancement techniques (subtraction, contrast adjustment, erosion, gamma correction, inversion and thresholding)	The erosion technique yielded the greatest results
Kaur R, Chawla M, Khiva NK, Ansari MD. Comparative Analysis of Contrast Enhancement Techniques for Medical Images. Pertanika J Sci Technol. 2018;26(3):965-78. ⁽³⁵⁾	Comparison of contrast enhancement techniques (neighborhood operation, median filter, imadjust and sigmoid function)	The sigmoid function and the neighborhood operation provided the greatest results
Shattuck DW, Sandor-Leahy SR, Schaper KA, Rottenberg DA, Leahy RM. Magnetic resonance image tissue classification using a partial volume model. Neuroimage. 2001;13(5):856-76. ⁽³⁶⁾	Development of a method for skull stripping (BSE) on T1-weighted images using a border detector and a series of morphological operations	A robust method, which contributes for GM and WM segmentation of the brain, respectively
Smith SM. Fast robust automated brain extraction. Hum Brain Mapp. 2002;17(3):143-55. Review. ⁽³⁷⁾	Development of a method for skull stripping (BET) on T1-weighted images	A robust, precise method applied to a range of MRI sequences
Roura E, Oliver A, Cabezas M, Vilanova JC, Rovira A, Ramió-Torrentà L, et al. MARGA: multispectral adaptive region growing algorithm for brain extraction on axial MRI. Comput Methods Programs Biomed. 2014 Feb 22;113(2):655-73. ⁽³⁸⁾	Development of a method for skull stripping (MARGA) on axial views, based on the growth of the region	The MARGA had superior results when compared with the BET and BSE approaches
Somasundaram K, Mercina JH, Magesh Kalaiselvi ST. Brain Portion Extraction Scheme using Region Growing and Morphological Operation from MRI of Human Head Scans. IJCSE. 2018;6(4):298-302. ⁽³⁹⁾	Development of a method for skull stripping based on growth of region and morphological operations (erosion, dilation and filling)	The results of the method are superior to those of existing methods (BET and BSE)
Kalavathi P, Prasath VB. Methods on Skull Stripping of MRI Head Scan Images—a Review. J Digit Imaging. 2016;29(3):365-79. Review. ⁽⁴⁰⁾	Development of a method for brain extraction on T1-weighted MRI images based on median filter and morphological operations	The method provided results comparable to those of the BET and BSE methods and showed that the median filter did not improve segmentation

Instituto Israelita de Ensino e Pesquisa Albert Einstein Avenida Albert Einstein, 627/701 , 05651-901 São Paulo - SP, Tel.: (55 11) 2151 0904 - São Paulo - SP - Brazil
E-mail: revista@einstein.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] Corresponding author: Layse Ribeiro Mascarenhas Rua do Cajueiro, CD Brauna, Torre 1, room 501 Bairro Cajueiro Zip code: 48904-083 – Juazeiro, BA, Brazil Phone: (55 74) 2102-7630 E-mail: yse.r.m@hotmail.com