Classifying Breast Density from Mammogram with Pretrained CNNs and Weighted Average Ensembles
Source abstract
Abstract
We are currently experiencing a revolution in data production and artificial intelligence (AI) applications. Data are produced much faster than they can be consumed. Thus, there is an urgent need to develop AI algorithms for all aspects of modern life. Furthermore, the medical field is a fertile field in which to apply AI techniques. Breast cancer is one of the most common cancers and a leading cause of death around the world. Early detection is critical to treating the disease effectively. Breast density plays a significant role in determining the likelihood and risk of breast cancer. Breast density describes the amount of fibrous and glandular tissue compared with the amount of fatty tissue in the breast. Breast density is categorized using a system called the ACR BI-RADS. The ACR assigns breast density to one of four classes. In class A, breasts are almost entirely fatty. In class B, scattered areas of fibroglandular density appear in the breasts. In class C, the breasts are heterogeneously dense. In class D, the breasts are extremely dense. This paper applies pre-trained Convolutional Neural Network (CNN) on a local mammogram dataset to classify breast density. Several transfer learning models were tested on a dataset consisting of more than 800 mammogram screenings from King Abdulaziz Medical City (KAMC). Inception V3, EfficientNet 2B0, and Xception gave the highest accuracy for both four- and two-class classification. To enhance the accuracy of density classification, we applied weighted average ensembles, and performance was visibly improved. The overall accuracy of ACR classification with weighted average ensembles was 78.11%.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: AI in cancer detection · Digital Radiography and Breast Imaging · Infrared Thermography in Medicine
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 22254, Saudi Arabia
Consultant Radiologist, King Abdulaziz Medical City, Jeddah 11426, Saudi Arabia
References (52)
Show source-provided references
1. History of medical imaging
2. Machine learning in medical imaging10.1016/j.jacr.2017.12.028 ↗
3. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries10.3322/caac.21660 ↗
4. Alotaibi, R.M., Rezk, H.R., Juliana, C.I., and Guure, C. (2018). Breast cancer mortality in Saudi Arabia: Modelling observed and unobserved factors. PLoS ONE, 13.10.1371/journal.pone.0206148 ↗
5. Efficacy of endocrine therapy for the treatment of breast cancer in men: Results from the MALE phase 2 randomized clinical trial10.1001/jamaoncol.2020.7442 ↗
6. Epidemiology of cancer in Saudi Arabia thru 2010–2019: A systematic review with constrained meta-analysis
7. (2021, September 12). Ministry of Health, Cancerous Diseases, Breast Cancer 2018, Available online: https://www.moh.gov.sa/en/HealthAwareness/EducationalContent/Diseases/Cancer/Pages/Cancer-2012-01-18-001.aspx.
8. Discussions of Dense Breasts, Breast Cancer Risk, and Screening Choices in 201910.1001/jama.2019.6247 ↗
9. Saha, D., Vaishnav, N., and Jha, A.K. (2021). Imaging Techniques for Breast Cancer Diagnosis. Biomedical Computing for Breast Cancer Detection and Diagnosis, IGI Global.10.4018/978-1-7998-3456-4.ch009 ↗
10. Breast cancer diagnosis: Imaging techniques and biochemical markers10.1002/jcp.26379 ↗
11. Breast Cancer, R.T. (2021, April 04). Mammography 2019. Available online: https://www.radiologyinfo.org/en/info.cfm?pg=mammo#top/.
12. Komen, S.G. (2016). Imaging Methods Used to Find Breast Cancer, Available online: https://www.komen.org/wp-content/uploads/Imaging-Methods-used-to-Find-Breast-Cancer.pdf.
13. Wolstencroft, R.T.W. (2020, September 30). Digital Database for Screening Mammography 2006, Computer Vision and Pattern Recognition. Available online: http://www.eng.usf.edu/cvprg/Mammography/Database.html.
14. A review of breast boundary and pectoral muscle segmentation methods in computer-aided detection/diagnosis of breast mammography10.1007/s10462-019-09721-8 ↗
15. Center for Disease Control and Prevention (2021, March 16). What Does It Mean to Have Dense Breasts?, Available online: https://www.cdc.gov/cancer/breast/basic_info/dense-breasts.htm.
16. Interval breast cancer: Analysis of occurrence, subtypes and implications for breast cancer screening in a model region10.1016/j.ejrad.2021.109905 ↗
17. Relationship of mammographic densities to breast cancer risk10.1186/s43055-021-00497-y ↗
18. The effect of chemotherapy on the mammographic appearance of breast cancer and correlation with histopathology10.1259/bjr.20150479 ↗
19. Education, I.C. (2022, April 22). Convolutional Neural Networks 2020. Available online: https://www.ibm.com/cloud/learn/convolutional-neural-networks.
20. Sebastian Raschka, V.M. (2017). Python Machine Learning, Packt.
21. Brownlee, J. (2022, January 12). How to Choose Loss Functions When Training Deep Learning Neural Networks. Available online: https://machinelearningmastery.com/how-to-choose-loss-functions-when-training-deep-learning-neural-networks/.
22. Gupta, A. (2021, December 16). A Comprehensive Guide on Deep Learning Optimizers. Available online: https://www.analyticsvidhya.com/blog/2021/10/a-comprehensive-guide-on-deep-learning-optimizers/.
23. A novel deep learning based method for the computational material design of flexoelectric nanostructures with topology optimization10.1016/j.finel.2019.07.001 ↗
24. A comprehensive survey on transfer learning10.1109/JPROC.2020.3004555 ↗
25. Wu, Y., Qin, X., Pan, Y., and Yuan, C. (2018, January 13–15). Convolution neural network based transfer learning for classification of flowers. Proceedings of the 2018 IEEE 3rd international conference on signal and image processing (ICSIP), Shenzhen, China.10.1109/SIPROCESS.2018.8600536 ↗
26. Tsochatzidis, L., Costaridou, L., and Pratikakis, I. (2019). Deep learning for breast cancer diagnosis from mammograms—A comparative study. J. Imaging, 5.10.3390/jimaging5030037 ↗
27. (2021, November 16). Keras Applications 2022. Available online: https://keras.io/api/applications/.
28. Dang, A.T. (2021, May 20). Accuracy and Loss: Things to Know about The Top 1 and Top 5 Accuracy. Available online: https://towardsdatascience.com/accuracy-and-loss-things-to-know-about-the-top-1-and-top-5-accuracy-1d6beb8f6df3.
29. Imagenet large scale visual recognition challenge10.1007/s11263-015-0816-y ↗
30. Ethnic differences in mammographic densities10.1093/ije/30.5.959 ↗
31. Ethnic and geographic differences in mammographic density and their association with breast cancer incidence10.1007/s10549-006-9387-5 ↗
32. Wu, N., Geras, K.J., Shen, Y., Su, J., Kim, S.G., Kim, E., Wolfson, S., Moy, L., and Cho, K. (2018, January 15–20). Breast density classification with deep convolutional neural networks. Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Calgary, AB, Canada.10.1109/ICASSP.2018.8462671 ↗
33. Shi, P., Wu, C., Zhong, J., and Wang, H. (2019, January 23–25). Deep Learning from Small Dataset for BI-RADS Density Classification of Mammography Images. Proceedings of the 2019 10th International Conference on Information Technology in Medicine and Education (ITME), Qingdao, China.10.1109/ITME.2019.00034 ↗
34. Classification of breast density categories based on SE-Attention neural networks10.1016/j.cmpb.2020.105489 ↗
35. BI-RADS density categorization using deep neural networks
36. Tardy, M., Scheffer, B., and Mateus, D. (2019, January 8–11). Breast Density Quantification Using Weakly Annotated Dataset. Proceedings of the 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), Venice, Italy.10.1109/ISBI.2019.8759283 ↗
37. Mammographic breast density assessment using deep learning: Clinical implementation10.1148/radiol.2018180694 ↗
38. Saffari, N., Rashwan, H.A., Abdel-Nasser, M., Kumar Singh, V., Arenas, M., Mangina, E., Herrera, B., and Puig, D. (2020). Fully Automated Breast Density Segmentation and Classification Using Deep Learning. Diagnostics, 10.10.3390/diagnostics10110988 ↗
39. Determination of mammographic breast density using a deep convolutional neural network10.1259/bjr.20180691 ↗
40. Multi-path deep learning model for automated mammographic density categorization
41. Lizzi, F., Laruina, F., Oliva, P., Retico, A., and Fantacci, M.E. (2019). Residual convolutional neural networks to automatically extract significant breast density features. International Conference on Computer Analysis of Images and Patterns, Springer.10.1007/978-3-030-29930-9_3 ↗
42. Lizzi, F., Atzori, S., Aringhieri, G., Bosco, P., Marini, C., Retico, A., Traino, A.C., Caramella, D., and Fantacci, M.E. (2019). Residual Convolutional Neural Networks for Breast Density Classification. BIOINFORMATICS, SciTePress.10.5220/0007522202580263 ↗
43. Cogan, T., and Tamil, L. (2020, January 20–24). Deep Understanding of Breast Density Classification. Proceedings of the 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada.10.1109/EMBC44109.2020.9176628 ↗
44. Multi-institutional assessment and crowdsourcing evaluation of deep learning for automated classification of breast density10.1016/j.jacr.2020.05.015 ↗
45. Roth, H.R., Chang, K., Singh, P., Neumark, N., Li, W., Gupta, V., Gupta, S., Qu, L., Ihsani, A., and Bizzo, B.C. (2020). Federated learning for breast density classification: A real-world implementation. Domain Adaptation and Representation Transfer, and Distributed and Collaborative Learning, Springer.10.1007/978-3-030-60548-3_18 ↗
46. Multi-view mammographic density classification by dilated and attention-guided residual learning10.1109/TCBB.2020.2970713 ↗
47. BASCNet: Bilateral adaptive spatial and channel attention network for breast density classification in the mammogram10.1016/j.bspc.2021.103073 ↗
48. Tlusty, T., Amit, G., and Ben-Ari, R. (2018, January 20–24). Unsupervised clustering of mammograms for outlier detection and breast density estimation. Proceedings of the 2018 24th International Conference on Pattern Recognition (ICPR), Beijing, China.10.1109/ICPR.2018.8545588 ↗
49. Ensemble deep learning in bioinformatics10.1038/s42256-020-0217-y ↗
50. Ganaie, M., and Hu, M. (2021). Ensemble Deep Learning: A Review, Cornell University.10.1016/j.engappai.2022.105151 ↗
51. Liashchynskyi, P., and Liashchynskyi, P. (2019). Grid search, random search, genetic algorithm: A big comparison for NAS. arXiv.
52. Bhattiprolu, S. (2022, January 04). Ensemble Sign Language. Available online: https://github.com/bnsreenu/python_for_microscopists/blob/master/213-ensemble_sign_language.py.