1. Abstract Breast DCE-MRI segmentation and lesion detection is a crucial image analysis task for cancer detection and tissue characterization. In this paper, a hard-clustering technique with Grammatical Fireworks algorithm (GFWA) is proposed to segment the breast MR images for lesion detection. GFWA is Swarm Programming (SP) method developed for automatic computer program generation in any arbitrary language
2. Introduction According to World Health Organization (WHO)’s report (WHO cancer prevention diagnosis screening breast cancer, https://www. who.int/cancer/prevention/diagnosisscreening/ breast-cancer/en/), it is estimated that 6,27,000 women in the world have died from breast cancer and it is approximately 15% of all types of cancer deaths among women in 2018. Breast cancer is the most common cancer in women in India and accounts for 14% of all cancers in women [1, 2]. Organized and opportunistic screening programs in the developed countries result in a significant decrease in mortality caused due to breast cancer [3]. Recently, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is widely used for breast cancer detection, diagnosis, and treatment planning or surgery. Several methods have been developed for lesion detection and its characterization in breast DCE-MRI in the recent past. [4] proposed a fuzzy c-means (FCM) clustering-based segmentation method for the detection of breast lesions in 3D DCE-MR images. FCM was applied to an enhanced region of interest (ROI) selected by a human operator manually. After clustering, binarization of the lesion membership map was done and lesions were finally selected followed by connected- omponent labeling [5]. developed a lesion segmentation and characterization methodology. First, Laplacian filter was used to enhance the lesions in ROI selected by a human operator manually. Then, extracted morphology and texture features from lesions were utilized for characterization in benign and alignant lesions using a Multilayer perceptron (MLP) [6].
The Iterative Conditional Mode (ICM) method was used to obtain the maximum a posteriori (MAP) estimate of the class membership of lesion and non-lesion [8]. proposed Improved Markov Random Field (IMRF) for lesion segmentation in breast DCE-MRI. The prior distributions of the class members were modeled as a ratio of conditional probability distributions of similar pixels and non-similar ones in a neighborhood. An adaptive moment preserving method was proposed by [9] (Wei et al., 2012) to segment the fibroglandular tissue in breast DCE-MRI. [10] (Chang et al., 2012) developed a computer-aided diagnosis (CAD) system for the characterization of breast mass lesions in benign and malignant breast tumors in DCE-MRI. [11] (Jayender et al., 2013) developed a statistical learning algorithm for tumor segmentation using Hidden Markov Models (HMMs) to auto-segment the angiogenesis corresponding to a tumor in breast DCE-MRI. [12] (Wang et al., 2013) proposed a hierarchical SVM-based segmentation method for breast DCE-MRI. 3D multi-parametric features from T1-weighted (T1-w), T2-weighted (T2-w), PD-w, and threepoint Dixon water-only and fat-only MRIs were used as inputs to the SVM to classify the breast tissues into fatty, fibroglandular, lesion, and skin. [13] (Milenkovi´c et al., 2013) applied logistic regression, the least-square minimum-distance classifier (LSMD), and least-squares support vector machine (LS-SVM) classifiers on breast DCE-MRI for classification of malignant and benign breast lesions in the breast. [14] (McClymont et al., 2014) used meanshift and graph-cuts algorithm for breast lesion segmentation in DCE-MRI. (Wang et al., 2014) used modifying FCM for clustering for breast tumor segmentation in DCE-MRI and further lesions are characterized using a pharmacokinetic model. A computer-aided detection auto probing (CADAP) system was developed by [15] (Sim et al., 2014) for lesion detection in breast DCE-MRI utilizing a spatial-based discrete Fourier transform and further characterized in benign, suspicious, or malignant.
3. Contribution of this Article Now, contributions of this article may be summarized as follows: 1. A GFWA-based segmentation methodology for breast lesion detection in DCE-MRI is proposed. A hard-clustering technique with GFWA is proposed to segment the breast DCE-MRI. 2. GFWA is not used previously in the segmentation of breast DCE-MRI as well as in any kind of image segmentation. Hence, the application of GFWA in breast lesion detection is another novelty of this article. 3. Finally, a comparative study of the proposed method is made with Grammatical Swarm based clustering technique [16, 18] (Si et al., 2014, 2015b) and K-means algorithm [19] (MacQueen, 1967).
References 1. Bray F, Ren, Jian-Song, Masuyer E. and Ferlay J. Estimates of global cancer prevalence for 27 sites in the adult population in 2008. Int J Cancer. 2013; 132(5): pp. 1133-45.
2. Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C.et. Al GLOBOCAN 2012 v1.0 Cancer Incidence and MortalityWorldwide : IARC Cancer Base. 2013; 11, F. Lyon :International Agency for Research on Cancer.
3. Shetty M.K. Breast Cancer Screening and Diagnosis - A Synopsis. Springer Science + Business Media, New York. 2015.
4. Chen W, Giger M.L. and Bick U. A Fuzzy C-Means (FCM)-Based approach for computerized segmentation of breast lesions in dynamic contrast-enhanced MR images. Acad Radiol, 2006; 13: 63-72.
5. Nie K, Chen J.H, Yu H.J, Chu Y, Nalcioglu O. and Su M.Y. Quantitative Analysis of Lesion Morphology and Texture Features for Diagnostic Prediction in Breast MRI. Acad. Radiol, 2008; 15: pp. 1513-1525.
6. Yao J, Chen J. and Chow C. Breast Tumor Analysis in Dynamic Contrast Enhanced MRI Using Texture Features and Wavelet Transform. IEEE Journal of Selected Topics in Signal Processing, 2009; 3(1): pp. 94-100.
7. Wu Q, Salganicoff M, Krishnan A, Fussell D.S. and Markey M.K. Interactive lesion segmentation on dynamic contrast enhanced breast MR using a Markov Model. Medical Imaging 2006: Image Processing, Eds. Joseph M. Reinhardt, Josien P.W. Pluim, Proc. of SPIE 2016; 6144: 61444M
8. Azmi R and Norozi N. A new markov random field segmentation method for breast lesion segmentation in MR images. J Med Signals Sens. 2011; 1: 3.
9. Wei Chia-hung, Li Y, Huang P.J, Gwo Chih-ying. and Harms S.E. Estimation of breast density: An adaptive moment preserving method for segmentation of fibroglandular tissue in breast magnetic resonance images. European Journal of Radiology, 2012; 81(4): pp.e618-e624.
10. Chang Yeun-Chung, Huang Yan-Hao, Huang Chiun-Sheng, Chang Pei-Kang, Chen Jeon-Hor. and Chang Ruey-Feng. Classification of breast mass lesions using model-based analysis of the characteristic kinetic curve derived from fuzzy c-means clustering. Magnetic Resonance Imaging, 2012; 30(3): pp. 312-322.
11. Jayender J, Gombos E, Chikarmane S, Dabydeen D, Jolesz F.A. and Vosburgh K.G. Statistical Learning Algorithm for in situ and invasive breast carcinoma segmentation. Computerized Medical Imaging and Graphics, 2013; 37: pp. 281-292.
12. Wang Y, Morrell G, Heibrun M.E, Payne A. and Parker D.L. 3D Multiparametric Breast MRI Segmentation Using Hierarchical Support Vector Machine with Coil Sensitivity Correction. Acad. Radiol. 2013; 20: pp. 137-147.
13. Milenkovi´c J, Hertl K., Košir A, Žibert J. and Tasiˇc J.F. Characterization of spatiotemporal changes for the classification of dynamic contrast-enhanced magneticresonance breast lesions. Artificial Intelligence in Medicine, 2013; 58(2): pp. 101-114.
14. McClymont D, Mehnert A, Trakic A, Kennedy D. and Crozier S. Fully Automatic Lesion Segmentation in Breast MRI Using Mean-Shift and Graph-Cuts on a Region Adjacency Graph. J. Magn. Reson. Imaging, 2014; 39: pp. 795-804.
15. Sim K.S, Chia F.K, Nia, M.E, Tso C.P, Chong A.K, Abbas S.F. and Chong S.S. Breast cancer detection from MR images through an auto-probing discrete Fourier transform system. Computers in Biology and Medicine, 2014; 49: pp. 46- 59.
16. Si T, De A. and Bhattacharjee A.K. Brain MRI segmentation for tumor detection using Grammatical Swarm based clustering algorithm. In Proc. of IEEE International onference on Circuits, Power and Computing Technologies, (ICCPCT-2014 ), Nagercoil, India.
17. Si T, De A. and Bhattacharjee A.K. Grammatical Swarm based Segmentation Methodology for Lesion Segmentation in Brain MRI. International Journal of Computer Applications, 2015; 121(4): pp. 1-8.
18. Si T. Grammatical Evolution Using Fireworks Algorithm. In: Pant M., Deep K., Bansal J., Nagar A., Das K. (eds) Proceedings of Fifth International Conference on Soft Computing for Problem Solving. Advances in Intelligent Systems and Computing, 2015; 436. Springer, Singapore.
19. MacQueen, J.B. Some Methods for classification and Analysis of Multivariate Observations. Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, University of California Press, 1967; pp. 281-297.
20. Balafar M.A, Ramli A.R. and Mashohor S. A new method for MR grayscale inhomogeneity correction. Artif. Intell. Rev. 2010; 34: pp. 195-204.
21. Brinkmann B.H, Manduca A. and Robb R.A. Optimized homomorphic un-sharp masking for MR greyscale inhomogeneity correction. IEEE Trans Med Imaging. 1998; 17(2); pp. 161-171.
22. Clark K, Vendt B, Smith K, Freymann J, Kirby J, Koppel P, et.Al The cancer imaging archive (TCIA): maintaining and operating a public information repository. J Digit Imaging, 2013; 26(6): pp. 1045-1057.
23. Davies D.L. and Bouldin D.W. A cluster separation measure. IEEE Trans. Pattern Anal. Mach. Intell, 1979; 1(2): pp. 224-227.
24. Derrac J, Garc´la S, Molina D. and Herrera F. A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm and Evolutionary Computation, 2011; 1: pp.3-18.
25. Lingle W, Erickson BJ, Zuley ML, Jarosz R, Bonaccio E, Filippini J. and Gruszauskas N. Radiology Data from The Cancer Genome Atlas Breast Invasive Carcinoma [TCGA-BRCA] collection. The Cancer Imaging Archive. 2016.
26. Mohana J, Krishnavenib V. and Guo Y. A survey on the magnetic resonance image denoising methods. Biomedical Signal Processing and Control, 2014; 9: pp. 56-69.
27. Perona P. and Malik J. Scale-Space and Edge Detection Using Anisotropic Diffussion. IEEE Trans. Pattern Anal. Mach. Intell., 1990; 12(7): pp. 629-639.
28. Si T, De A. and Bhattacharjee A.K. Segmentation of Brain MRI Using Wavelet Transform and Grammatical Bee Colony. Journal of Circuits, Systems and Computers, 2018; 27: 07, doi: 10.1142/ S0218126618501086.
29. Tan Y. and Zhu Y. Fireworks algorithm for optimization. In: Tan Y. et al. (eds.) ICSI 2010, Part I, LNCS, 2010; 6145: pp. 355-364. Springer, Heidelberg.
30. Wang T.C, Huang Y.H, Huang C.S, Chen J.H, Huang G.Y, Chang Y.C. and Chang R.F. Computer-aided diagnosis of breast DCE-MRI using pharmacokinetic model and 3-D morphology analysis. Magn Reson Imaging, 2014; 32(3): pp. 197-205.
Mondal Sukumar. Breast DCE-MRI Segmentation for Lesion Detection Using Grammatical Fireworks Algorithm. Annals of Clinical and Medical Case Reports 2021