A Comprehensive Framework for Multi-Class Breast Lesion Classification Using a Gaussian Process-Optimized VGG16-Capsule Hybrid Model on Ultrasound Radio Frequency B-mode Images

Document Type : Research Article

Authors

1 Department of Biomedical Engineering, Amirkabir University of Technology, Tehran, Iran

2 Department of Medical Sciences and Technologies, Islamic Azad University, Science and Research Branch, Tehran, Iran

3 Radiology-Medical Imaging Center, Cancer Research Institute, Imam Khomeini Hospital Advanced Diagnostic and Interventional Radiology Research Center, Tehran University of Medical Sciences, Tehran, Iran

Abstract

Effective breast cancer screening is essential for early detection and treatment. Ultrasound (US) radio frequency (RF) data offers a novel, equipment-independent approach. However, class imbalance and limited interpretability hinder its application in clinical practice. This study proposes a hybrid deep learning model combining a pre-trained convolutional neural network (CNN) based on VGG16 and capsule neural networks (CapsNets) to classify breast lesions. The model was evaluated using an RFTSBU dataset, comprising 220 data points from 118 patients, acquired on the SuperSonic Imagine Aixplorer® system with a linear transducer. To address data imbalance, the synthetic minority over-sampling technique (SMOTE) was employed to generate synthetic samples while preserving data distribution. Furthermore, Gaussian process (GP) was applied to fine-tune CapsNet hyperparameters, improving classification performance. Three experiments were conducted to classify breast lesions into two, three, and four classes: (I) CapsNet with balanced datasets based on class weight, (II) CapsNet with balanced datasets using SMOTE, and (III) CapsNet with hyperparameters optimized using GP on SMOTE-balanced datasets. The proposed model achieved average accuracies of 98.81%, 97.89%, and 95.94% for two-, three-, and four-class classifications, respectively. The hybrid VGG16-CapsNet model effectively addresses class imbalance and captures critical lesion attributes such as size, perspective, and orientation. Integrating GP optimization achieves superior accuracy in multi-class breast lesion classification. The proposed approach can serve as a valuable aid in breast tumor classification using US RF B-mode images. Its enhanced interpretability and efficiency enable clinicians to move beyond binary classification, facilitating the identification and differentiating a broader spectrum of breast lesions.

Keywords

Main Subjects

References

[1] P. Jarosik, Z. Klimonda, M. Lewandowski, M. Byra, Breast lesion classification based on ultrasonic radio-frequency signals using convolutional neural networks, Biocybernetics and Biomedical Engineering, 40(3) (2020) 977-986.
[2] M. Arnold, E. Morgan, H. Rumgay, A. Mafra, D. Singh, M. Laversanne, J. Vignat, J.R. Gralow, F. Cardoso, S. Siesling, I. Soerjomataram, Current and future burden of breast cancer: Global statistics for 2020 and 2040, The Breast, 66 (2022) 15-23.
[3] G.C. Observatory, Population Fact Sheets of Iran 2020, in:  Population Fact Sheets, International Agency for Research on Cancer, France 2021, pp. 1-2.
[4] M.L. Brown, F. Houn, E.A. Sickles, L.G. Kessler, Screening mammography in community practice: Positive predictive value of abnormal findings and yield of follow-up diagnostic procedures, AJR Am. J. Roentgenol., 165(6) (1995) 1373-1377.
[5] S.P. Poplack, A.N. Tosteson, M.R. Grove, W.A. Wells, P.A. Carney, Mammography in 53,803 women from the New Hampshire mammography network, Radiology, 217(3) (2000) 832-840.
[6] S.P. Poplack, P.A. Carney, J.E. Weiss, L. Titus-Ernstoff, M.E. Goodrich, A.N. Tosteson, Screening mammography: Costs and use of screening-related services, Radiology, 234(1) (2005) 79-85.
[7] J. Haas, C. Kaplan, A. McMillan, L.J. Esserman, Does timely assessment affect the anxiety associated with an abnormal mammogram result?, J. Womens Health Gend. Based Med., 10(6) (2001) 599-605.
[8] N.M. Lindberg, D. Wellisch, Anxiety and compliance among women at high risk for breast cancer, Ann. Behav. Med., 23(4) (2001) 298-303.
[9] R.C. Burack, M.S. Simon, M. Stano, J. George, J. Coombs, Follow-up among women with an abnormal mammogram in an HMO: is it complete, timely, and efficient?, Am. J. Manag. Care, 6(10) (2000) 1102-1113.
[10] K. Drukker, M.L. Giger, K. Horsch, M.A. Kupinski, C.J. Vyborny, E.B. Mendelson, Computerized lesion detection on breast ultrasound, Med. Phys., 29(7) (2002) 1438-1446.
[11] H. Yu-Len, C. Dar-Ren, L. Ya-Kuang, Breast cancer diagnosis using image retrieval for different ultrasonic systems, in:  2004 International Conference on Image Processing, 2004. ICIP '04., 2004, pp. 2957-2960 Vol. 2955.
[12] B.O. Anderson, R. Shyyan, A. Eniu, R.A. Smith, C.H. Yip, N.S. Bese, L.W. Chow, S. Masood, S.D. Ramsey, R.W. Carlson, Breast cancer in limited-resource countries: An overview of the Breast Health Global Initiative 2005 guidelines, The breast journal, 12 (2006) S3-15.
[13] M. Costantini, P. Belli, R. Lombardi, G. Franceschini, A. Mulè, L. Bonomo, Characterization of solid breast masses: Use of the sonographic breast imaging reporting and data system lexicon, J. Ultrasound Med., 25(5) (2006) 649-659;.
[14] C.M. Chen, Y.H. Chou, K.C. Han, G.S. Hung, C.M. Tiu, H.J. Chiou, S.Y. Chiou, Breast lesions on sonograms: Computer-aided diagnosis with nearly setting-independent features and artificial neural networks, Radiology, 226(2) (2003) 504-514.
[15] B. Sahiner, H.P. Chan, M.A. Roubidoux, L.M. Hadjiiski, M.A. Helvie, C. Paramagul, J. Bailey, A.V. Nees, C. Blane, Malignant and benign breast masses on 3D US volumetric images: effect of computer-aided diagnosis on radiologist accuracy, Radiology, 242(3) (2007) 716-724.
[16] K. Wei, B. Wang, J. Saniie, Faster region convolutional neural networks applied to ultrasonic images for breast lesion detection and classification, in:  2020 IEEE International Conference on Electro Information Technology (EIT), IEEE, Chicago, IL, USA, 2020, pp. 171-174.
[17] B. Behboodi, H. Rasaee, A. Tehrani, H. Rivaz, Deep classification of breast cancer in ultrasound images: more classes, better results with multi-task learning, in:  2021 Ultrasonic Imaging and Tomography Conference, 2021.
[18] M. Qiao, Z. Fang, Y. Guo, S. Zhou, C. Chang, Y. Wang, Breast calcification detection based on multichannel radiofrequency signals via a unified deep learning framework, Expert Systems with Applications, 168(114218) (2021) 1-11.
[19] S. Kim, J. Park, J. Yi, H. Kim, End-to-end convolutional neural network framework for breast ultrasound analysis using multiple parametric images generated from radiofrequency signals, Applied Sciences, 12(10) (2022) 1-17.
[20] M. Byra, P. Jarosik, K. Dobruch-Sobczak, Z. Klimonda, H. Piotrzkowska-Wroblewska, J. Litniewski, A. Nowicki, Joint segmentation and classification of breast masses based on ultrasound radio-frequency data and convolutional neural networks, Ultrasonics, 121(6) (2022) 1-9.
[21] X. Li, Y. Sang, X. Ma, Y. Cai, Quantitative feature classification for breast ultrasound images using improved naive bayes, IET Image Processing, 17(5) (2023) 1417-1426.
[22] G.R. Gare, J. Li, R. Joshi, R. Magar, M.P. Vaze, M. Yousefpour, R.L. Rodriguez, J.M. Galeotti, W-Net: Dense and diagnostic semantic segmentation of subcutaneous and breast tissue in ultrasound images by incorporating ultrasound RF waveform data, Med. Image Anal., 76 (2022) 1-14.
[23] H. Taleghamar, S.A. Jalalifar, G.J. Czarnota, A. Sadeghi-Naini, Deep learning of quantitative ultrasound multi-parametric images at pre-treatment to predict breast cancer response to chemotherapy, Sci. Rep., 12(1) (2022) 1-13.
[24] L.M. AbouEl-Magd, A. Darwish, V. Snasel, A.E. Hassanien, A pre-trained convolutional neural network with optimized capsule networks for chest X-rays COVID-19 diagnosis, Cluster Computing, 26(2) (2023) 1389-1403.
[25] S. Tiwari, A. Jain, Convolutional capsule network for COVID-19 detection using radiography images, International journal of imaging systems and technology, 31(2) (2021) 525-539.
[26] A.R. Bushara, R.S. Vinod Kumar, S.S. Kumar, An ensemble method for the detection and classification of lung cancer using Computed Tomography images utilizing a capsule network with Visual Geometry Group, Biomedical signal processing and control, 85 (2023) 104930.
[27] M. Kwabena Patrick, A. Felix Adekoya, A. Abra Mighty, B.Y. Edward, Capsule networks – A survey, Journal of King Saud University - Computer and Information Sciences, 34(1) (2022) 1295-1310.
[28] M.U. Haq, M.A. Sethi, A.U. Rehman, Capsule Network with Its Limitation, Modification, and Applications—A Survey, in:  Machine Learning and Knowledge Extraction, 2023, pp. 891-921.
[29] G.E. Hinton, S. Sabour, N. Frosst, Matrix capsules with EM routing, in:  International Conference on Learning Representations (ICLR18), 2018, pp. 1-15.
[30] Y. Wang, D. Ning, S. Feng, A Novel Capsule Network Based on Wide Convolution and Multi-Scale Convolution for Fault Diagnosis, in:  Appl. Sci., 2020, pp. 1-16.
[31] C.E. Rasmussen;, C. Williams;, Gaussian processes for machine learning, MIT Press, 2006.
[32] I. Guyon, A. Statnikov, B.B. Batu, Automated machine learning: Methods, systems, challenges, Springer, 2019.
[33] D.R. Jones, M. Schonlau, W.J. Welch, Efficient Global Optimization of Expensive Black-Box Functions, Journal of Global Optimization, 13(4) (1998) 455-492.
[34] N. Chawla, K. Bowyer, L. Hall, W. Kegelmeyer, SMOTE: Synthetic minority over-sampling technique, Journal of Artificial Intelligence Research, 16 (2002) 321-357.
[35] D. Elreedy, A.F. Atiya, A comprehensive analysis of synthetic minority oversampling technique (SMOTE) for handling class imbalance, Journal of information science, 505 (2019) 32-64.
[36] Y. Lecun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to document recognition, Proceedings of the IEEE, 86(11) (1998) 2278-2324.
[37] A. Kaya, A.S. Keceli, C. Catal, H.Y. Yalic, H. Temucin, B. Tekinerdogan, Analysis of transfer learning for deep neural network based plant classification models, Computers and Electronics in Agriculture, 158 (2019) 20-29.
[38] Sara Sabour, Nicholas Frosst, G.E. Hinton, Dynamic routing between capsules, Adv. Neural Inf. Process. Syst.,  (2017).
[39] G.E. Hinton, A. Krizhevsky, S.D. Wang, Transforming auto-encoders, in: T. Honkela, W. Duch, M. Girolami, S. Kaski (Eds.) Artificial Neural Networks and Machine Learning – ICANN 2011, Springer Berlin Heidelberg, Berlin, Heidelberg, 2011, pp. 44-51.

Files

Share

How to cite

Statistics