Tomosentez Görüntüleri ile Yapılan Derin Öğrenme Çalışmalarında Kullanılan Görüntü Ön İşleme Yöntemleri Üzerine Bir Literatür Araştırması

Öz

Bu makale, tomosentez görüntülerinin derin öğrenme çalışmalarında kullanılmasına odaklanarak, görüntü ön işleme yöntemleri üzerine bir literatür araştırması sunmaktadır. Tomosentez, meme dokusunun 3 boyutlu, kesitsel olarak taranmasını sağlayan gelişmiş bir tıbbi görüntüleme tekniğidir. Bu teknikle elde edilen görüntüler 2 boyutlu mamografilere oranla daha yüksek boyutlu olduğu gibi daha gürültülü de olabilirler. Bu nedenle bu görüntülerin derin öğrenme modellerine uygun hale getirilmesi için ön işleme yapılması gerekmektedir. Bu literatür araştırması, tomosentez görüntülerinde kullanılan farklı ön işleme yöntemlerini ele almaktadır. Öncelikle Tomosentez görüntülerinin özellikleri ve derin öğrenme yöntemleri hakkında bir giriş yapılacaktır. Daha sonra, kullanılan ön işleme yöntemleri arasında yer alan filtreleme, normalizasyon, segmentasyon ve artırma gibi teknikler hakkında yapılan literatür araştırmasına ait bilgi verilecektir. Ayrıca, bu yöntemlerin bir arada kullanıldığı örnekler de incelenecektir. Sonuç olarak, bu makale ile Tomosentez görüntüleri üzerinde derin öğrenme çalışmaları yapmak isteyen araştırmacılara faydalı bir Türkçe kaynak sunmak hedeflenmektedir. Yapılan araştırma, görüntü ön işleme yöntemlerinin doğru seçiminin, derin öğrenme modellerinin performansını önemli ölçüde artırabileceğini göstermektedir.

Anahtar Kelimeler

• Derin Öğrenme
• Görüntü Ön İşleme
• Tomosentez
• Meme Kanseri

A Literature Review on Image Preprocessing Methods Used in Deep Learning Studies Using Tomosynthesis Images

Abstract

This article presents a literature review on image preprocessing methods, focusing on the use of tomosynthesis images in deep learning studies. Tomosynthesis is an advanced medical imaging technique that provides 3-dimensional, cross-sectional scanning of breast tissue. The images obtained with this technique can be higher dimensional and noisier than 2D mammograms. Therefore, preprocessing is required to make these images suitable for deep learning models. This literature review addresses the different preprocessing methods used in tomosynthesis images. First of all, an introduction will be made about the properties of Tomosynthesis images and deep learning methods. Then, information about the techniques such as filtering, normalization, segmentation and augmentation, which are among the preprocessing methods used, will be given from the literature search. In addition, examples where these methods are used together will also be examined. In conclusion, with this article, it is aimed to present a useful Turkish resource to researchers who want to do deep learning studies on Tomosynthesis images. The research shows that the right choice of image preprocessing methods can significantly improve the performance of deep learning models.

Keywords

• Deep Learning
• Image Pre-Processing
• Tomosynthesis
• Breast Cancer

References

1. Ahmed, L., Iqbal, M. M., Aldabbas, H., Khalid, S., Saleem, Y., & Saeed, S. (2020). Images data practices for semantic segmentation of breast cancer using Deep Neural Network. Journal of Ambient Intelligence and Humanized Computing. https://doi.org/10.1007/s12652-020-01680-1
2. Alguliyev, R. M., Aliguliyev, R. M., & Abdullayeva, F. J. (2019). The improved LSTM and CNN models for ddos attacks prediction in social media. International Journal of Cyber Warfare and Terrorism, 9(1), 1–18. https://doi.org/10.4018/ijcwt.2019010101
3. Amit, G., Ben-Ari, R., Hadad, O., Monovich, E., Granot, N., & Hashoul, S. (2017). Classification of breast MRI lesions using small-size training sets: Comparison of Deep Learning Approaches. SPIE Proceedings. https://doi.org/10.1117/12.2249981
4. Bengio, Y. (2009). Learning deep architectures for AI. Foundations and Trends® in Machine Learning, 2(1), 1–127. https://doi.org/10.1561/2200000006
5. Bevilacqua, V., Brunetti, A., Guerriero, A., Trotta, G. F., Telegrafo, M., & Moschetta, M. (2019). A performance comparison between shallow and deeper neural networks supervised classification of tomosynthesis breast lesions images. Cognitive Systems Research, 53, 3–19. https://doi.org/10.1016/j.cogsys.2018.04.011
6. Boser, B., LeCun, Y., Denker, J. S. (1989 ). Handwritten Digit Recognition with a Back-Propagation Network.
7. Buda, M., Saha, A., Walsh, R., Ghate, S., Li, N., Swiecicki, A., Lo, J. Y., & Mazurowski, M. A. (2021). A data set and deep learning algorithm for the detection of masses and architectural distortions in digital breast tomosynthesis images. JAMA Network Open, 4(8). https://doi.org/10.1001/jamanetworkopen.2021.19100
8. Ciresan, D. C., Meier, U., Gambardella, L. M., & Schmidhuber, J. (2011). Convolutional neural network committees for handwritten character classification. 2011 International Conference on Document Analysis and Recognition. https://doi.org/10.1109/icdar.2011.229

9. El-Shazli, A. M., Youssef, S. M., & Soliman, A. H. (2022). Intelligent Computer-aided model for efficient diagnosis of digital breast tomosynthesis 3D imaging using Deep Learning. Applied Sciences, 12(11), 5736. https://doi.org/10.3390/app12115736
10. Elman, J. L. (1990). Finding structure in time. Cognitive Science, 14(2), 179–211. https://doi.org/10.1207/s15516709cog1402_1
11. Graves, A., & Schmidhuber, J. (2005). Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural Networks, 18(5–6), 602–610. https://doi.org/10.1016/j.neunet.2005.06.042
12. Harron, N. A., Osman, N. F., Sulaiman, S. N., Karim, N. K., Ismail, A. P., & Soh, Z. H. (2022). An image denoising model using deep learning for Digital Breast Tomosynthesis Images. 2022 IEEE 13th Control and System Graduate Research Colloquium (ICSGRC). https://doi.org/10.1109/icsgrc55096.2022.9845152
13. Helvie, M. A. (2010). Digital Mammography Imaging: Breast Tomosynthesis and Advanced Applications. Radiologic Clinics of North America, 48(5), 917–929. https://doi.org/10.1016/j.rcl.2010.06.009
14. Hinton, G. E., Osindero, S., & The, Y. (2006). A fast learning algorithm for deep belief nets. Neural Computation, 18(7), 1527–1554. https://doi.org/10.1162/ neco.2006.18.7.1527
15. Hinton, G. E., Srivastava, N., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. R. (2012). Improving neural networks by preventing co-adaptation of feature detectors. CoRR, abs/1207.0580.
16. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
17. Hooley, R. J., Durand, M. A., & Philpotts, L. E. (2017). Advances in Digital Breast Tomosynthesis. American Journal of Roentgenology,208(2),256–266. https://doi.org/10.2214/ajr.16.17127
18. Liu, W., Wang, Z., Liu, X., Zeng, N., Liu, Y., & Alsaadi, F. E. (2017). A survey of deep neural network architectures and their applications. Neurocomputing, 234, 11–26. https://doi.org/10.1016/j.neucom.2016.12.038
19. Lu, M. T., Ivanov, A., Mayrhofer, T., Hosny, A., Aerts, H. J., & Hoffmann, U. (2019). Deep learning to assess long-term mortality from chest radiographs. JAMA Network Open, 2(7). https://doi.org/10.1001/jamanetworkopen.2019.7416
20. Memisevic, R., & Hinton, G. E. (2010). Learning to represent spatial transformations with factored higher-order Boltzmann machines. Neural Computation, 22(6), 1473–1492. https://doi.org/10.1162/neco.2010.01-09-953
21. Ren, J., Green, M., & Huang, X. (2021). From traditional to deep learning: Fault diagnosis for Autonomous Vehicles. Learning Control, 205–219. https://doi.org/10.1016/b978-0-12-822314-7.00013-4
22. Ricciardi, R., Mettivier, G., Staffa, M., Sarno, A., Acampora, G., Minelli, S., Santoro, A., Antignani, E., Orientale, A., Pilotti, I. A. M., Santangelo, V., D’Andria, P., & Russo, P. (2021). A deep learning classifier for digital breast tomosynthesis. Physica Medica, 83, 184–193. https://doi.org/10.1016/j.ejmp.2021.03.021
23. Salakhutdinov, R. & Larochelle, H. (2009). Efficient learning of deep Boltzmann machines, in Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, 16‐18 April., Florida, USA, 693‐700.
24. Salakhutdinov, R., & Hinton, G. (2009). Deep Boltzmann machines, in Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, 16–18 April, Florida, USA, 448–455.
25. Salakhutdinov, R., & Murray, I. (2008). On the quantitative analysis of deep belief networks, Proceedings of the 25th international conference on Machine learning ‐ ICML '08, USA, 10–20.
26. Samala, R. K., Chan, H.-P., Hadjiiski, L., Helvie, M. A., Richter, C. D., & Cha, K. H. (2019). Breast cancer diagnosis in digital breast tomosynthesis: Effects of training sample size on multi-stage transfer learning using deep neural nets. IEEE Transactions on Medical Imaging, 38(3), 686–696. https://doi.org/10.1109/tmi.2018.2870343
27. Sarker, I. H. (2021). Deep learning: A comprehensive overview on techniques, taxonomy, applications and Research Directions. SN Computer Science, 2(6). https://doi.org/10.1007/s42979-021-00815-1
28. Schwenzow, J., Hartmann, J., Schikowsky, A. and Heitmann, M. (2021), “Understanding videos at scale: how to extract insights for business research”, Journal of Business Research, Vol. 123, pp. 367-379, doi: 10.1016/j.jbusres.2020.09.059.
29. Sechopoulos, I., Teuwen, J., & Mann, R. (2021). Artificial Intelligence for Breast Cancer Detection in mammography and Digital Breast Tomosynthesis: State of the art. Seminars in Cancer Biology, 72, 214–225. https://doi.org/10.1016/j.semcancer.2020.06.002
30. Shimokawa, D., Takahashi, K., Kurosawa, D., Takaya, E., Oba, K., Yagishita, K., Fukuda, T., Tsunoda, H., & Ueda, T. (2022). Deep learning model for breast cancer diagnosis based on bilateral asymmetrical detection (Bilad) in digital breast tomosynthesis images. Radiological Physics and Technology, 16(1), 20–27. https://doi.org/10.1007/s12194-022-00686-y
31. Shimokawa, D., Takahashi, K., Oba, K., Takaya, E., Usuzaki, T., Kadowaki, M., Kawaguchi, K., Adachi, M., Kaneno, T., Fukuda, T., Yagishita, K., Tsunoda, H., & Ueda, T. (2022). Deep Learning Model for Predicting the Presence of Stromal Invasion of Breast Cancer on Digital Breast Tomosynthesis. https://doi.org/10.21203/rs.3.rs-1807556/v1 Singh, S., Matthews, T. P., Shah, M., Mombourquette, B., Tsue, T., Long, A., Almohsen, R., Pedemonte, S., & Su, J. (2020). Adaptation of a deep learning malignancy model from full-field digital mammography to Digital Breast Tomosynthesis. Medical Imaging 2020: Computer-Aided Diagnosis. https://doi.org/10.1117/12.2549923
32. Skaane, P., Bandos, A. I., Gullien, R., Eben, E. B., Ekseth, U., Haakenaasen, U., Izadi, M., Jebsen, I. N., Jahr, G., Krager, M., Niklason, L. T., Hofvind, S., & Gur, D. (2013). Comparison of Digital Mammography alone and Digital Mammography Plus Tomosynthesis in a population-based screening program. Radiology, 267(1), 47–56. https://doi.org/10.1148/radiol.12121373
33. Usuga Cadavid, J.P., Grabot, B., Lamouri, S., Pellerin, R. and Fortin, A. (2022), “Valuing free-form text data from maintenance logs through transfer learning with CamemBERT”, Enterprise Information Systems, Vol. 16 No. 6, pp. 1-29, 1790043, doi: 10.1080/17517575.2020.1790043.
34. Usuga-Cadavid, J.P., Lamouri, S., Grabot, B. and Fortin, A. (2022), “Using deep learning to value freeform text data for predictive maintenance”, International Journal of Production Research, Vol. 60 No. 14, pp. 4548-4575, doi: 10.1080/00207543.2021.1951868.
35. Vedantham, S., Karellas, A., Vijayaraghavan, G. R., & Kopans, D. B. (2015). Digital Breast Tomosynthesis: State of the art. Radiology, 277(3), 663–684. https://doi.org/10.1148/radiol.2015141303
36. Yousefi, M., Krzyżak, A., & Suen, C. Y. (2018). Mass detection in digital breast tomosynthesis data using convolutional neural networks and multiple instance learning. Computers in Biology and Medicine, 96, 283–293. https://doi.org/10.1016/j.compbiomed.2018.04.004
37. Zhang, X., Zhang, Y., Han, E. Y., Jacobs, N., Han, Q., Wang, X., & Liu, J. (2018). Classification of whole mammogram and tomosynthesis images using deep convolutional neural networks. IEEE Transactions on NanoBioscience, 17(3), 237–242. https://doi.org/10.1109/tnb.2018.2845103
38. Zhang, Y., Wang, X., Blanton, H., Liang, G., Xing, X., & Jacobs, N. (2019). 2d Convolutional Neural Networks for 3D Digital Breast Tomosynthesis Classification. 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). https://doi.org/10.1109/bibm47256.2019.8983097
39. Zhao, B., Zhang, X., Li, H. and Yang, Z. (2020), “Intelligent fault diagnosis of rolling bearings based on normalized CNN considering data imbalance and variable working conditions”, KnowledgeBased Systems, Vol. 199, 105971, doi: 10.1016/j.knosys.2020.105971.
40. Zorzi, M., Testolin, A., & Stoianov, I. P. (2013). Modeling language and cognition with deep unsupervised learning: A tutorial overview. Frontiers in Psychology, 4(1), 515–527.