Abstract
The objective of the study was to determine if the pathology depicted on a mammogram is either benign or malignant (ductal or non-ductal carcinoma) using deep learning and artificial intelligence techniques. A total of 559 patients underwent breast ultrasound, mammography, and ultrasound-guided breast biopsy. Based on the histopathological results, the patients were divided into three categories: benign, ductal carcinomas, and non-ductal carcinomas. The mammograms in the cranio-caudal view underwent pre-processing and segmentation. Given the large variability of the areola, an algorithm was used to remove it and the adjacent skin. Therefore, patients with breast lesions close to the skin were removed. The remaining breast image was resized on the Y axis to a square image and then resized to 512 × 512 pixels. A variable square of 322,622 pixels was searched inside every image to identify the lesion. Each image was rotated with no information loss. For data augmentation, each image was rotated 360 times and a crop of 227 × 227 pixels was saved, resulting in a total of 201,240 images. The reason why our images were cropped at this size is because the deep learning algorithm transfer learning used from AlexNet network has an input image size of 227 × 227. The mean accuracy was 95.8344% ± 6.3720% and mean AUC 0.9910% ± 0.0366%, computed on 100 runs of the algorithm. Based on the results, the proposed solution can be used as a non-invasive and highly accurate computer-aided system based on deep learning that can classify breast lesions based on changes identified on mammograms in the cranio-caudal view.
Similar content being viewed by others
Availability of Data and Material
Due to the nature of this research, participants of this study did not agree for their data to be shared publicly.
Code Availability
The code used in this study is available from the corresponding author, MSS, upon reasonable request.
Abbreviations
- ACC:
-
Accuracy
- DL:
-
Deep learning
- ML:
-
Machine learning
- AI:
-
Artificial intelligence
- CNN:
-
Convolutional neural network
- MM:
-
Mammography
- MLO:
-
Medio-lateral oblique
- CC:
-
Cranio-caudal
- BI-RADS:
-
Breast Imaging-Reporting and Data System
References
Iacoviello L, Bonaccio M, de Gaetano G, Donati MB: Epidemiology of breast cancer, a paradigm of the “common soil” hypothesis. Semin Cancer Biol, https://doi.org/10.1016/j.semcancer, February 20, 2020
Gheonea IA, Donoiu L, Camen D, Popescu FC, Bondari S: Sonoelastography of breast lesions: A prospective study of 215 cases with histopathological correlation. Rom J Morphol Embryol 52:1209-1214,2011
Gheonea IA, Stoica Z, Bondari S: Differential diagnosis of breast lesions using ultrasound elastography. Indian J Radiol Imaging 21:301–305,2011
Donoiu L, Camen D, Camen G, Calota F: A comparison of echography and elastography in the differentiation of breast tumors. Ultraschall Med – Eur J Ultrasound 29:OP_2_5,2008
Shen L, Margolies LR, Rothstein JH, Fluder E, McBride R, Sieh W: Deep Learning to Improve Breast Cancer Detection on Screening Mammography. Sci Rep 9:1–2,2019
Abdelhafiz D, Yang C, Ammar R, Nabavi S: Deep convolutional neural networks for mammography: Advances, challenges and applications. BMC Bioinformatics 20:281,2019
Kooi T, Litjens G, Van Ginneken B, Gubern-Mérida A, Sánchez CI, Mann R, den Heeten A, Karssemeijer N: Large scale deep learning for computer aided detection of mammographic lesions. Med Image Anal 35:303–312,2017
Hayes Balmadrid MA, Shelby RA, Wren AA, Miller LS, Yoon SC, Baker JA, Wildermann LA, Soo MS: Anxiety prior to breast biopsy: Relationships with length of time from breast biopsy recommendation to biopsy procedure and psychosocial factors. J Health Psychol 22(5):561-571,2017
Thatcher effect. Available at https://en.wikipedia.org/wiki/Thatcher_effect. Accessed 12 June 2020
AlexNet convolutional neural network – MATLAB alexnet. Available at https://uk.mathworks.com/help/deeplearning/ref/alexnet.html. Accessed 10 June 2020
Belciug S: Artificial Intelligence in cancer, diagnostic to tailored treatment, 1st Edition, Cambridge Massachusetts, Academic Press, 2020
Venkatesan A, Chu P, Kerlikowske K, Sickles E, Smith-Bindman R: Positive predictive value of specific mammographic findings according to reader and patient variables. Radiology 250:648–657,2009
Invasive Ductal Carcinoma (IDC). Available at https://www.hopkinsmedicine.org/breast_center/breast_cancers_other_conditions/invasive_ductal_carcinoma.html#:~:text=Invasive%20ductal%20carcinoma%20(IDC)%2C,of%20all%20breast%20cancer%20diagnoses. Accessed 12 June 2020
Artificial Intelligence, Machine Learning & Deep learning. Available at https://becominghuman.ai/artificial-intilligence-machine-learning-deep-learning-df6dd0af500e. Accessed 12 June 2020
Rodriguez-Ruiz A, Lång K, Gubern-Merida A, Broeders M, Gennaro G, Clauser P, Helbich TH, Chevalier M, Tan T, Mertelmeier T, Wallis MG: Stand-Alone Artificial Intelligence for Breast Cancer Detection in Mammography: Comparison With 101 Radiologists. JNCI J Natl Cancer Inst 111:916–922,2019
Dhungel N, Carneiro G, Bradley AP: A deep learning approach for the analysis of masses in mammograms with minimal user intervention. Med Image Anal 37:114–128,2017
Institute of Electrical and Electronics Engineers, International Association for Pattern Recognition, Australian Pattern Recognition Society: Automated Mass Detection in Mammograms using Cascaded Deep Learning and Random Forests. Available at https://cs.adelaide.edu.au/~carneiro/publications/mass_detection_dicta.pdf. Accessed 10 June 2020
Breast Imaging-Reporting and Data System (BI-RADS). Available at https://radiopaedia.org/articles/breast-imaging-reporting-and-data-system-bi-rads. Accessed 21 July 2021
Boyd NF, Martin LJ, Bronskill M, Yaffe MJ, Duric N, Minkin S: Breast tissue composition and susceptibility to breast cancer. J Natl Cancer Inst 102:1224–1237,2010
Kerlikowske K, Zhu W, Tosteson AN, Sprague BL, Tice JA, Lehman CD, Miglioretti DL: Identifying Women With Dense Breasts at High Risk for Interval Cancer: A Cohort Study. Ann Intern Med 162:673–681,2015
Merino Bonilla JA, Torres Tabanera M, Ros Mendoza LH: Breast Cancer in the 21st Century: From Early Detection to New Therapies. Radiologia 59:368–379,2017
Mammography views. Available at https://radiopaedia.org/articles/mammography-views. Accessed 21 July 2021
Cogan T, Tamil L: Deep Understanding of Breast Density Classification. Annu Int Conf IEEE Eng Med Biol Soc. 2020:1140-1143,2020
Mohamed AA, Luo Y, Peng H, Jankowitz RC, Wu S: Understanding Clinical Mammographic Breast Density Assessment: a Deep Learning Perspective. J Digit Imaging 31(4):387-392,2018
Author information
Authors and Affiliations
Contributions
REN, MSS, and IAG share main authorship due to conceiving the main conceptual ideas and designing the study. GCC and REN performed all the imaging techniques and ultrasound-guided breast biopsies. MSS designed the model and the computational framework and with support from CTS and LMF carried out the implementation and worked out all the technical details. IAG supervised the study and together with the rest of the authors provided critical feedback and helped shape the research, analysis, and manuscript.
Corresponding author
Ethics declarations
Ethics Approval
Institutional review board approval was obtained (University of Medicine and Pharmacy of Craiova, Committee of Ethics and Academic and Scientific Deontology approval 45/17.06.2020).
Consent to Participate
Written informed consent was obtained from all subjects (patients) in this study.
Conflict of Interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Key Points
1. Deep learning computer-aided diagnosis of breast pathology on a digital mammogram helps clinicians classify lesions either benign or malignant.
2. Deep learning algorithms may also be used as an objective first or second reader and as a support tool to accelerate radiologists’ time to process examinations.
3. The patients can benefit from a more appropriate and a less invasive management and treatment of the breast lesions.
Rights and permissions
About this article
Cite this article
Nica, RE., Șerbănescu, MS., Florescu, LM. et al. Deep Learning: a Promising Method for Histological Class Prediction of Breast Tumors in Mammography. J Digit Imaging 34, 1190–1198 (2021). https://doi.org/10.1007/s10278-021-00508-4
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10278-021-00508-4