Diagnosis of Breast Cancer by K-Mean Clustering and Otsu Thresholding Segmentation Methods

Abstract

Breast cancer has increased decidedly among women. But with early diagnosis, a positive response to treatment can be given. Researchers are conducting various studies in imaging methods to detect the disease early and accurately. In this study, 9 cancerous images taken from the TCİA image data bank were detected by K-mean clustering and the Otsu threshold method. Performance metrics were evaluated by comparing them with marked reference images (ground truth) by the radiologist. For the clustering process, TPR (True Positive Rate) 0.89, FPR (False Positive Rate) 0.14, similarity 0.67, accuracy 0.87, sensitivity 0.89, exact hit ratio 0.86, specificity 0.87, F Score 0.87 were found, respectively. For Otsu, TPR (True Positive Rate) 0.84, FPR (False Positive Rate) 0.12, similarity 0.73, accuracy 0.84, sensitivity 0.84, exact hit 0.86, specificity 0.87, F Score 0.84 were calculated. The aim of this study is to determine the tumor boundaries more accurately and to use them in imaging devices in the field of health with pixel-based segmentation. As a result, both methods were successful can be used in the field and close to each other.

Keywords

• Mammography
• Image processing
• K-mean clustering

Meme Kanserinin K-Ortalama Kümeleme ve Otsu Eşikleme Segmentasyon Yöntemleri İle Teşhisi

Öz

Meme kanseri, kadınlar arasında büyük oranda artış göstermiştir. Ancak erken teşhisiyle, tedaviye olumlu cevap verilebilmektedir. Araştırmacılar, hastalığı erken ve doğru tespit edebilme adına görüntüleme yöntemlerinde çeşitli çalışmalar yapmaktadır. Bu çalışmada; TCİA görüntü veri bankasından alınan 9 kanserli görüntüde K ortalama kümeleme ve otsu eşikleme yöntemi ile tümör tespiti yapılmıştır. Radyolog tarafından işaretli referans görüntüleri ile (ground truth) ile karşılaştırarak, başarım (performans) metrikleri değerlendirilmiştir. Kümeleme işlemi için sırasıyla TPR (Doğru Pozitif Oranı) 0.89, FPR (Yanlış Pozitif Oranı) 0.14, benzerlik 0.67, doğruluk 0.87, duyarlılık 0.89, hassasiyet 0.86, özgüllük 0.87, F puanı 0.87 bulunmuştur. Otsu için TPR (Doğru Pozitif Oranı) 0.84, FPR (Yanlış Pozitif Oranı) 0.12, benzerlik 0.73, doğruluk 0.84, duyarlılık 0.84, hassasiyet 0.86, özgüllük 0.87, F puanı 0.84 olarak hesaplanmıştır. Bu çalışmada, daha az veri kümesi ile daha kısa sürede, görüntü işleme yöntemlerini kullanarak, piksel tabanlı segmentasyon ile tümör sınırlarının daha doğru belirlenmesi, insana duyulan ihtiyacın azalması ve sağlık alanında sahada görüntülemede kullanılan tıbbi cihazların bilgisayar destekliyazılımlarla geliştirilmesi, mamografik tarama sistemlerinin doğru ve hızlı bir şekilde yapılabilmesi amaçlanmıştır. Sonuç olarak, her iki yöntem de başarılı, sahada kullanılabilir ve birbirine yakın başarım değerleri bulunmuştur.

Anahtar Kelimeler

• Mamografi
• Görüntü işleme
• K ortalama kümeleme

Kaynakça

1. [1] Cancer Facts and Figures. American Cancer Society. Atlanta 2020; 1–76.
2. [2] Canadian Cancer Statistics Advisory Committee. Canadian Cancer Statistics 2018.
3. [3] Toronto ON: Canadian Cancer Society 2020. Available at: cancer.ca/Canadian-Cancer-Statistics-2018-EN.
4. [4] Sun L., Legood R., Sadique Z. Dos-Santos-Silva I., Yang L. Cost–effectiveness of risk-based breast cancer screening programme China. Bull World Health Organ 2018; 96: 568-577.
5. [5] Malvia S., Bagadi SA., Dubey US., Saxena S. Epidemiology of breast cancer in Indian women. Asia-Pacific J Clin Oncol 2017; 13(4): 289–295. [6] Ng EYK., Kee EC. Advanced integrated technique in breast cancer thermography. J Med Eng Technol 2008; 32: 103–114.
6. [7] Mentari BA., Rasyid Y., Fitri A., Khairul M. Histogram statistics and GLCM features of breast thermograms for early cancer detection, 15th International Conference on Electrical Engineering/Electronics. Computer, Telecommunications and Information Technology (ECTI-NCON2018) 2018; 120- 124.
7. [8] Etehadtavakol M., Ng EYK. Survey of numerical bioheat transfer modelling for accurate skin surface measurements. Therm Sci Eng Prog J 2020; 20: 2451–9049.
8. [9] Das S., Abraham A., Konar A. Automatic clustering using an ımproved differential evolution algorithm. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans 2008; 38: 218-237.

9. [10] Shokrgozar N., Sobhani FM. Customer segmentation of bank based on iscovering of their transactional relation by using data mining algorithms. Modern Applied Science 2016; 10(10): 283-286.
10. [11] Khan SS., Ahmad A. Cluster centre initialization algorithm for k-means cluster. In Pattern Recognition Letters 2004; 1293–1302.
11. [12] Kaur A. Comparative analysis of segmentation algorithms for brain tumor detection in MR images 2017.
12. [13] Yang MS., Sinaga KP. A feature-reduction multi-view K-means clustering algorithm. IEEE 2019; 7: 114472-114486.
13. [14] Lin H., Ji Z. Breast cancer prediction based on K-Means and SOM Hybrid Algorithm. In Journal of Physics: Conference Series. IOP Publishing 2020; 1624(4): 1-7.
14. [15] Aswathy MA., Jagannath M. Performance analysis of segmentation algorithms for the detection of breast cancer. Procedia Computer Science 2020; 167: 666-676.
15. [16] Çiklaçandir FGY., Ertaylan A., Bınzat U., Kut A. Lesion detection from the ultrasound images using k-means algorithm. 2019 Medical Technologies Congress (TIPTEKNO) 2019; 1-4.
16. [17] Bottou L., Lin CJ. Support vector machine solvers. Large Scale Kernel Mach 2007 ;3(1): 301–320.
17. [18] Tang T., Chen S., Zhao M., Huang W., Luo J. Very large-scale data classification based on K-means clustering and multi-kernel SVM. Soft Computing 2019; 23(11): 3793-3801.
18. [19] Katz E., Barness Y. Comparison of SNR and Peak-SNR (PSNR) as performance measures and signals for peak-limited two-dimensional (2D) pixelated optical wireless communication, in: Conference on Signals, Systems & Computers. IEEE 2015; 1880–1884.
19. [20] Et-taleby A, Boussetta M, Benslimane M. Faults detection for photovoltaic field based on K-Means, Elbow, and Average Silhouette Techniques through the Segmentation of a Thermal Image. International Journal of Photoenergy 2020.
20. [21] Tang T, Chen S, Zhao M, Huang W, Luo J. Very large-scale data classification based on K-means clustering and multi-kernel SVM. Soft Computing 2019;23(11):3793- 3801.
21. [22] Mittal, H., Saraswat, M. An optimum multi-level image thresholding segmentation using non-local means 2D histogram and exponential Kbest gravitational search algorithm, Engineering Applications of Artificial Intelligence 2018; 71: 226–235.
22. [23] Bradley PS., Fayyad UM. Refining ınitial points for k -means clustering, 15th International Conference on Machine Learning, San Francisco-ABD 1998; 91-99.
23. [24] Karen SJ., Emily FC., Mary SS. Molecular subtypes of breast cancer: A review for breast radiologists. Journal of Breast Imaging 2021; 3(1): 12–24.
24. [25] Ghosh S., Dubey KS. Comparative analysis of k-means and fuzzy c means algorithms. ((IJACSA). International Journal of Advanced Computer Science and Applications 2013; (4)4: 35-39. [26] Banerjee S., Choudhary A., Pal S. Empirical evaluation of k-means, k-means bisection, fuzzy c-means and genetic k-means clustering algorithms. IEEE international WIE conference on electrical and computer engineering in 2015; 168-172.
25. [27] Kaygısız H., Çakır A. FPGA kullanılarak görüntülerin gerçek zamanlı olarak OTSU metodu ile bölütlenmesi. Avrupa Bilim ve Teknoloji Dergisi 2020; 18: 911-917.
26. [28] Kapoor A., Singhal A. A comparative study of k-means, k-means++ and fuzzy c-means clustering algorithms. 2017 3rd International Conference on Computational Intelligence & Communication Technology (CICT) 2017; 1-6.
27. [29] Dallali A., el Khediri S., Amel SA., Kachouri A. Breast tumors segmentation using Otsu method and K-means. ATSIP 2018; 1-6.
28. [30] Tianwen X., Qiufeng Z., Caixia F., Qianming B., Xiaoyan Z., Lihua L., Robert G., Li L., Yajia G., Weijun P. In MRI, complete tumor histogram analysis and otsu threshold method to distinguish breast cancer from others 2019.
29. [31] Malali HE., Assir A., Harmouchi M., Rattal M., Lyazidi A., Mouhsen A. Adaptive local gray-level transformation based on variable s-curve for contrast enhancement of mammogram images. Embedded Systems and Artificial Intelligence 2020; 671-679.
30. [32] Aksebzeci B. Computer-aided classification of breast cancer histopathological images; 2017.
31. [33] Dubey KA., Gupta U., Jain S. Comparative study of k-means and fuzzy c-means algorithms on the breast cancer data. International Journal on Advanced Science Engineering Information Technology 2018; 8(1): 18-29.