Bounded-abstaining classification for breast tumors in imbalanced ultrasound images

• Autorzy: Guan Hongjiao , Zhang Yingtao , Cheng Heng-Da , Tang Xianglong

Identyfikatory

DOI

10.34768/amcs-2020-0025

• Języki publikacji: EN

Abstrakty

EN

Computer-aided breast ultrasound (BUS) diagnosis remains a difficult task. One of the challenges is that imbalanced BUS datasets lead to poor performance, especially with regard to low accuracy in the minority (malignant tumor) class. Missed diagnosis of malignant tumors can cause serious consequences, such as delaying treatment and increasing the risk of death. Moreover, many diagnosis methods do not consider classification reliability; thus, some classifications may have a large uncertainty. To resolve such problems, a bounded-abstaining classification model is proposed. It maximizes the area under the ROC curve (AUC) under two abstention constraints. A total of 219 (92 malignant and 127 benign) BUS images are collected from the First Affiliated Hospital of Harbin Medical University, China. The experiment tests BUS datasets of three imbalance levels, and the performance contours are analyzed. The results demonstrate that AUC-rejection curves are less affected by class imbalance than accuracy-rejection curves. Compared with the state-of-the-art, the proposed method yields a significantly larger AUC and G-mean using imbalanced BUS datasets.

Słowa kluczowe

EN

breast ultrasound image; reliable diagnosis; abstaining classification; imbalanced dataset

PL

ultrasonografia piersi; diagnoza wiarygodna; zbiór danych niezrównoważony

• Wydawca: Oficyna Wydawnicza Uniwersytetu Zielonogórskiego
• Czasopismo: International Journal of Applied Mathematics and Computer Science
• Rocznik: 2020
• Tom: Vol. 30, no. 2
• Strony: 325--336
• Opis fizyczny: Bibliogr. 51 poz., rys., tab., wykr.

Twórcy

autor

Guan Hongjiao

• School of Cyber Security, Qilu University of Technology (Shandong Academy of Sciences), 3501 Daxue Road, Changqing District, Jinan 250353, China; Shandong Key Laboratory of Computer Networks, 3501 Daxue Road, Changqing District, Jinan 250353, China; School of Computer Science and Technology, Harbin Institute of Technology, 92 West Dazhi Street, Nangang District, Harbin 150001, China

autor

Zhang Yingtao

• School of Computer Science and Technology, Harbin Institute of Technology, 92 West Dazhi Street, Nangang District, Harbin 150001, China

autor

Cheng Heng-Da

• School of Computer Science, Utah State University, Logan, UT 84322, USA

autor

Tang Xianglong

• School of Computer Science and Technology, Harbin Institute of Technology, 92 West Dazhi Street, Nangang District, Harbin 150001, China

Bibliografia

• [1] Abdel-Nasser, M., Melendez, J., Moreno, A., Omer, O.A. And Puig, D. (2017). Breast tumor classification in ultrasound images using texture analysis and super-resolution methods, Engineering Applications of Artificial Intelligence 59: 84–92.
• [2] Acharya, U.R., Ng, W.L., Rahmat, K., Sudarshan, V.K., Koh, J.E., Tan, J.H., Hagiwara, Y., Yeong, C.H. And Ng, K.H. (2017). Data mining framework for breast lesion classification in shear wave ultrasound: A hybrid feature paradigm, Biomedical Signal Processing and Control 33: 400–410.
• [3] Cai, L.,Wang, X.,Wang, Y., Guo, Y., Yu, J. andWang, Y. (2015). Robust phase-based texture descriptor for classification of breast ultrasound images, Biomedical Engineering Online 14(1): 26.
• [4] Chang, J.M., Moon, W.K., Cho, N., Yi, A., Koo, H.R., Han, W., Noh, D.-Y., Moon, H.-G. and Kim, S.J. (2011). Clinical application of shear wave elastography (SWE) in the diagnosis of benign and malignant breast diseases, Breast Cancer Research and Treatment 129(1): 89–97.
• [5] Chen, S.-C., Cheung, Y.-C., Su, C.-H., Chen, M.-F., Hwang, T.-L. and Hsueh, S. (2004). Analysis of sonographic features for the differentiation of benign and malignant breast tumors of different sizes, Ultrasound in Obstetrics and Gynecology 23(2): 188–193.
• [6] Cheng, H.-D., Shan, J., Ju, W., Guo, Y. and Zhang, L. (2010). Automated breast cancer detection and classification using ultrasound images: A survey, Pattern Recognition 43(1): 299–317.
• [7] Daoud, M.I., Bdair, T.M., Al-Najar, M. and Alazrai, R. (2016). A fusion-based approach for breast ultrasound image classification using multiple-ROI texture and morphological analyses, Computational and Mathematical Methods in Medicine 2016: 6740956.
• [8] Fawcett, T. (2004). ROC graphs: Notes and practical considerations for researchers, Machine Learning 31(1): 1–38.
• [9] Fawcett, T. (2006). An introduction to ROC analysis, Pattern Recognition Letters 27(8): 861–874.
• [10] Fischer, L., Hammer, B. and Wersing, H. (2015). Efficient rejection strategies for prototype-based classification, Neurocomputing 169: 334–342.
• [11] Fu, J., Li, Y., Li, N. and Li, Z. (2018). Comprehensive analysis of clinical utility of three-dimensional ultrasound for benign and malignant breast masses, Cancer Management and Research 10: 3295–3303.
• [12] Gai, S., Zhang, B., Yang, C. and Yu, L. (2018). Speckle noise reduction in medical ultrasound image using monogenic wavelet and Laplace mixture distribution, Digital Signal Processing 72: 192–207.
• [13] Garcia-Closas, M. et al. (2008). Heterogeneity of breast cancer associations with five susceptibility loci by clinical and pathological characteristics, PLoS Genetics 4(4): e1000054.
• [14] Guan, H., Zhang, Y., Cheng, H., Xian, M. and Tang, X. (2019). Ba2cs: Bounded abstaining with two constraints of reject rates in binary classification, Neurocomputing 357: 125–134.
• [15] Guyon, I. and Elisseeff, A. (2003). An introduction to variable and feature selection, Journal of Machine Learning Research 3(Mar): 1157–1182.
• [16] Haralick, R.M., Shanmugam, K. and Dinstein, I. (1973). Textural features for image classification, IEEE Transactions on Systems, Man, and Cybernetics SMC-3(6): 610–621.
• [17] Hofmann, T. (1999). Probabilistic latent semantic analysis, Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence, Stockholm, Sweden, pp. 289–296.
• [18] Hong, X., Chen, S. and Harris, C.J. (2007). A kernel-based two-class classifier for imbalanced data sets, IEEE Transactions on Neural Networks 18(1): 28–41.
• [19] Kang, S., Cho, S., Rhee, S.-j. and Yu, K.-S. (2017). Reliable prediction of anti-diabetic drug failure using a reject option, Pattern Analysis and Applications 20(3): 883–891.
• [20] Lee, J., Nishikawa, R.M., Reiser, I. and Boone, J.M. (2018). Relationship between computer segmentation performance and computer classification performance in breast CT: A simulation study using RGI segmentation and LDA classification, Medical Physics 45(8): 3650–3656.
• [21] Li, L., Zhou, X., Zhao, X., Hao, S., Yao, J., Zhong, W. and Zhi, H. (2017). B-mode ultrasound combined with color Doppler and strain elastography in the diagnosis of non-mass breast lesions: A prospective study, Ultrasound in medicine & biology 43(11): 2582–2590.
• [22] Liberman, L. and Menell, J.H. (2002). Breast imaging reporting and data system (BI-RADS), Radiologic Clinics of North America 40(3): 409–430.
• [23] Lin, C.-M., Hou, Y.-L., Chen, T.-Y. and Chen, K.-H. (2014). Breast nodules computer-aided diagnostic system design using fuzzy cerebellar model neural networks, IEEE Transactions on Fuzzy Systems 22(3): 693–699.
• [24] Liu, Y., Cheng, H., Huang, J., Zhang, Y., Tang, X., Tian, J.-W. And Wang, Y. (2012). Computer aided diagnosis system for breast cancer based on color Doppler flow imaging, Journal of Medical Systems 36(6): 3975–3982.
• [25] López, V., Fernández, A., García, S., Palade, V. and Herrera, F. (2013). An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics, Information Sciences 250: 113–141.
• [26] Monticciolo, D.L., Newell, M.S., Hendrick, R.E., Helvie, M.A., Moy, L., Monsees, B., Kopans, D.B., Eby, P.R. and Sickles, E.A. (2017). Breast cancer screening for average-risk women: Recommendations from the ACR commission on breast imaging, Journal of the American College of Radiology 14(9): 1137–1143.
• [27] Moon, W.K., Chen, I.-L., Yi, A., Bae, M.S., Shin, S.U. and Chang, R.-F. (2018). Computer-aided prediction model for axillary lymph node metastasis in breast cancer using tumor morphological and textural features on ultrasound, Computer Methods and Programs in Biomedicine 162: 129–137.
• [28] Mousania, Y. and Karimi, S. (2019). Contrast improvement of ultrasound images of focal liver lesions using a new histogram equalization, in K.S. Montaser (Ed.), Fundamental Research in Electrical Engineering, Springer, Singapore, pp. 43–53.
• [29] Pietraszek, T. (2007). On the use of ROC analysis for the optimization of abstaining classifiers, Machine Learning 68(2): 137–169.
• [30] Prati, R.C., Batista, G. and Monard, M.C. (2011). A survey on graphical methods for classification predictive performance evaluation, IEEE Transactions on Knowledge and Data Engineering 23(11): 1601–1618.
• [31] Rahmawaty, M., Nugroho, H. A., Triyani, Y., Ardiyanto, I. And Soesanti, I. (2016). Classification of breast ultrasound images based on texture analysis, International Conference on Biomedical Engineering (IBIOMED), Yogyakarta, Indonesia, pp. 1–6.
• [32] Rawashdeh, M., Lewis, S., Zaitoun, M. and Brennan, P. (2018). Breast lesion shape and margin evaluation: BI-RADS based metrics understate radiologists’ actual levels of agreement, Computers in Biology and Medicine 96: 294–298.
• [33] Rodriguez-Cristerna, A., Gomez-Flores, W. and de Albuquerque Pereira, W.C. (2018). A computer-aided diagnosis system for breast ultrasound based on weighted bi-rads classes, Computer Methods and Programs in Biomedicine 153: 33–40.
• [34] Roffo, G., Melzi, S., Castellani, U. and Vinciarelli, A. (2017). Infinite latent feature selection: A probabilistic latent graph-based ranking approach, Proceedings of the IEEE International Conference on Computer Vision (ICCV 2017), Venice, Italy, pp. 1407–1415.
• [35] Roffo, G., Melzi, S. and Cristani, M. (2015). Infinite feature selection, Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile, pp. 4202–4210.
• [36] Roy, R., Ghosh, S. and Ghosh, A. (2018). Speckle de-noising of clinical ultrasound images based on fuzzy spel conformity in its adjacency, Applied Soft Computing 73: 394–417.
• [37] Shan, J., Alam, S.K., Garra, B., Zhang, Y. and Ahmed, T. (2016). Computer-aided diagnosis for breast ultrasound using computerized BI-RADS features and machine learning methods, Ultrasound in Medicine and Biology 42(4): 980–988.
• [38] Shi, X., Cheng, H.-D., Hu, L., Ju, W. and Tian, J. (2010). Detection and classification of masses in breast ultrasound images, Digital Signal Processing 20(3): 824–836.
• [39] Simeone, P., Marrocco, C. and Tortorella, F. (2012). Design of reject rules for ECOC classification systems, Pattern Recognition 45(2): 863–875.
• [40] Singh, B.K., Verma, K., Panigrahi, L. and Thoke, A. (2017a). Integrating radiologist feedback with computer aided diagnostic systems for breast cancer risk prediction in ultrasonic images: An experimental investigation in machine learning paradigm, Expert Systems with Applications 90: 209–223.
• [41] Singh, K., Ranade, S.K. and Singh, C. (2017b). A hybrid algorithm for speckle noise reduction of ultrasound images, Computer Methods and Programs in Biomedicine 148: 55–69.
• [42] Tesfahun, A. and Bhaskari, D.L. (2013). Intrusion detection using random forests classifier with smote and feature reduction, International Conference on Cloud & Ubiquitous Computing & Emerging Technologies (CUBE), Pune, India, pp. 127–132.
• [43] Tortorella, F. (2000). An optimal reject rule for binary classifiers, in F.J. Ferri et al. (Eds), Advances in Pattern Recognition, SSPR/SPR 2000, Lecture Notes in Computer Science, Vol. 1876, Springer, Berlin/Heidelberg, pp. 611–620.
• [44] Tortorella, F. (2004). Reducing the classification cost of support vector classifiers through an ROC-based reject rule, Pattern Analysis and Applications 7(2): 128–143.
• [45] Wang, Z., Wang, Z., He, S., Gu, X. and Yan, Z.F. (2017). Fault detection and diagnosis of chillers using Bayesian network merged distance rejection and multi-source non-sensor information, Applied Energy 188: 200–214.
• [46] Wu, G. and Chang, E.Y. (2005). KBA: Kernel boundary alignment considering imbalanced data distribution, IEEE Transactions on Knowledge and Data Engineering 17(6): 786–795.
• [47] Yassin, N.I., Omran, S., El Houby, E.M. and Allam, H. (2017). Machine learning techniques for breast cancer computer aided diagnosis using different image modalities: A systematic review, Computer Methods and Programs in Biomedicine 156: 25-45.
• [48] Yu, H. and Ni, J. (2014). An improved ensemble learning method for classifying high-dimensional and imbalanced biomedicine data, IEEE/ACM Transactions on Computational Biology and Bioinformatics 11(4): 657–666.
• [49] Yu, X., Hao, X., Wan, J., Wang, Y., Yu, L. and Liu, B. (2018). Correlation between ultrasound appearance of small breast cancer and axillary lymph node metastasis, Ultrasound in Medicine & Biology 44(2): 342–349.
• [50] Zhang, J., Lin, G., Wu, L., Wang, C. and Cheng, Y. (2015). Wavelet and fast bilateral filter based de-speckling method for medical ultrasound images, Biomedical Signal Processing and Control 18: 1–10.
• [51] Zhou, S., Shi, J., Zhu, J., Cai, Y. and Wang, R. (2013). Shearlet-based texture feature extraction for classification of breast tumor in ultrasound image, Biomedical Signal Processing and Control 8(6): 688–696.

Uwagi

PL

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).