A method for the automated classification of benign and malignant masses on digital breast tomosynthesis images using machine learning and radiomic features
- 44 Downloads
In digital mammography, which is used for the early detection of breast tumors, oversight may occur due to overlap between normal tissues and lesions. However, since digital breast tomosynthesis can acquire three-dimensional images, tissue overlapping is reduced, and, therefore, the shape and distribution of the lesions can be easily identified. However, it is often difficult to distinguish between benign and malignant breast lesions on images, and the diagnostic accuracy can be reduced due to complications from radiological interpretations, owing to acquisition of a higher number of images. In this study, we developed an automated classification method for diagnosing breast lesions on digital breast tomosynthesis images using radiomics to comprehensively analyze the radiological images. We extracted an analysis area centered on the lesion and calculated 70 radiomic features, including the shape of the lesion, existence of spicula, and texture information. The accuracy was compared by inputting the obtained radiomic features to four classifiers (support vector machine, random forest, naïve Bayes, and multi-layer perceptron), and the final classification result was obtained as an output using a classifier with high accuracy. To confirm the effectiveness of the proposed method, we used 24 cases with confirmed pathological diagnosis on biopsy. We also compared the classification results based on the presence or absence of dimension reduction using least absolute shrinkage and a selection operator (LASSO). As a result, when the support vector machine was used as a classifier, the correct identification rate of the benign tumors was 55% and that of malignant tumors was 84%, with best results. These results indicate that the proposed method may help in more accurately diagnosing cases that are difficult to classify on images.
KeywordsBreast cancer Tomosynthesis Image analysis Radiomics
We are grateful to Ms. Tomoko Otsuka of Daido Hospital for annotation of clinical data.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest.
All the procedures in studies involving human participants were performed in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The present article does not contain any studies performed on animals by any of the authors.
Consent from the patients was obtained with a condition that all data were anonymized.
- 1.NIH: Cancer Statistics: Reports on Cancer: Cancer Stat Facts: Female Breast Cancer, 2015. https://seer.cancer.gov/statfacts/html/breast.html. Accessed Feb 2019.
- 11.John M. Core-needle biopsy for breast abnormalities. AHRQ Pub. No.14(16)-EHC040-3-EF; 2016.Google Scholar
- 13.Yamazaki M, Teramoto A, Fujita H. A hybrid detection scheme of architectural distortion in mammograms using iris filter and Gabor filter. Lect Comput Sci. 2016;9699:174–82.Google Scholar
- 18.Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Adv Neurol. 2012;25(6):1106–14.Google Scholar
- 19.Ragab D, Sharkas M, Marshall S, Ren J. Breast cancer detection using deep convolutional neural networks and support vector machines. PeerJ. 2019;7(2601):1–23.Google Scholar
- 20.Arevalo J, González F, Ramos-Pollán R. Convolutional neural networks for mammography mass lesion classification. In: 37th Annual international conference of the IEEE engineering in medicine and biology society (EMBC); 2015, pp 797–800.Google Scholar
- 21.Tian J, Dong D, Liu Z et al. Radiomics in medical imaging-detection, extraction and segmentation. In: Artificial Intelligence in Decision Support Systems for Diagnosis in Medical Imaging. Intelligent Systems Reference Library, vol 140; 2018.Google Scholar
- 23.Ma J, Wang Q, Ren Y, et al. Automatic lung nodule classification with radiomics approach. In: Proceedings of SPIE 9789, medical imaging 2016: PACS and imaging informatics: next generation and innovations, 978906; 2016.Google Scholar
- 25.Hologic's public data. https://www.dclunie.com/pixelmedimagearchive/upmcdigitalmammotomocollection/index.html. Accessed Nov 2018.
- 26.American College of Radiology. Breast imaging reporting and data system (BI-RADS®). 3rd ed. Reston: American College of Radiology; 1998.Google Scholar
- 30.Teramoto A, Tsujimoto M, Inoue T, et al. Automated classification of pulmonary nodules through a retrospective analysis of conventional CT and two-phase PET images in patients undergoing biopsy. Asia Ocean J Nucl Med Biol. 2018;7(1):29–37.Google Scholar
- 31.Yoshikawa R, Teramoto A, Matsubara T, Fujita H. Automated detection of architectural distortion using improved adaptive Gabor filter. International workshop on digital mammography. Cham: Springer; 2014. p. 606–611.Google Scholar
- 32.Selvarajah S, Kodituwakku S. Analysis and comparison of texture features for content based image retrieval. Int J Latest Trends Comput. 2011;2(1):108–13.Google Scholar
- 33.Carlson J. ‘Radiomic' Image Processing Toolbox. https://cran.r-project.org/web/packages/radiomics/radiomics.pdf. Accessed Nov 2018.
- 37.Leijenaar RTH, Carvalho S, Velazquez ER, van Elmpt WJC, Parmar C, Hoekstra OS, Hoekstra CJ, Boellaard R, Dekker ALAJ, Gillies RJ, Aerts HJWL, Lambin P. Stability of FDGPET radiomics features: an integrated analysis of test–retest and inter-observer variability. Acta Oncol. 2013;52(7):1391–7.CrossRefGoogle Scholar
- 39.Onishi Y, Teramoto A, Tsujimoto M, et al. Automated pulmonary nodule classification in computed tomography images using a deep convolutional neural network trained by generative adversarial networks. BioMed Res Int. 2019;2019:1–9 (Article ID 6051939).Google Scholar