Transfer Learning for automated search for defects on chest X-rays
https://doi.org/10.24835/1607-0763-1243
Abstract
Purpose. To develop and test algorithms for determining the projection and searching for common technical defects on chest -rays using transfer learning with various neural network architectures.
Materials and methods. Algorithms have been created to search for technical remarks such as incorrect choice of study boundaries and errors of patient positioning. Transfer learning of neural network architectures VGG19 and ResNet152V2 was chosen as the basis for creating algorithms. To train and test the algorithms, we used radiographs from open databases (over 230,000 studies in total). To validate the obtained algorithms, a test dataset was prepared from 150 anonymized chest x-rays unloaded from the Unified Radiological Information Service of the Moscow city (URIS) and evaluated by expert doctors and technicians.
Results. All obtained algorithms have high classification quality indicators. The maximum accuracy on the test dataset was obtained for the model that determines the projection, AUC was 1.0, the minimum accuracy: AUC 0.968 was obtained for the model that determines the rotation of the chest on the lateral X-ray. On the validation dataset maximum accuracy was obtained for the model that determines the projection, AUC was 0.996, the minimum accuracy: AUC 0.898 was obtained for the model that determines the rotation of the chest on the lateral x-ray.
Conclusions. All of diagnostic accuracy metrics for each of the models exceeded the threshold value of 0.81 and can be recommended for practical use.
Keywords
chest X-ray,
quality assurance,
machine learning,
artificial neural network,
Deep convolutional neural networks,
transfer learning
Supporting agencies: The article was written as part of a grant with the registration number: 122112400040-1
About the Authors
A. A. Borisov
Pirogov Russian National Research Medical University (Pirogov Medical University);
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department
Russian Federation
Russian Federation
Alexander A. Borisov – software developer; Junior Researcher of Medical Intelligent Technologies
1, Ostrovityanova str., Moscow 117997;
24-1, Petrovka str., Moscow 127051
S. S. Semenov
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department
Russian Federation
Russian Federation
Serafim S. Semenov – Junior Researcher in the Sector of Development of Systems for the Introduction of Medical Intelligent Technologies
24-1, Petrovka str., Moscow 127051
K. M. Arzamasov
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department
Russian Federation
Russian Federation
Kirill M. Arzamasov – Cand. of Sci. (Med.), Head of the Department of Medical Informatics, Radiomics and Radiogenomics
24-1, Petrovka str., Moscow 127051
References
1. Kim T.K., Yi P.H., Wei J. et al. Deep learning method for automated classification of anteroposterior and postero-anterior chest radiographs. J. Digit. Imaging. 2019; 32 (6): 925–930. https://doi.org/10.1007/s10278-019-00208-0
2. Shet N., Chen J., Siegel E. Continuing challenges in defining image quality. Pediatr. Radiol. 2011; 41 (5): 582–589. https://doi.org/10.1007/s00247-011-2028-0
3. McDonald R.J., Schwartz K.M., Eckel L.J. et al. The effects of changes in utilization and technological advancements of cross-sectional imaging on radiologist workload. Acad. Radiol. 2015; 22 (9): 1191–1198. https://doi.org/10.1016/j.acra.2015.05.007
4. Willis C.E., Nishino T.K., Wells J.R. et al. Automated quality control assessment of clinical chest images. Med. Phys. 2018; 45 (10): 4377–4391. https://doi.org/10.1002/mp.13107
5. Rale A.P., Gharpure D.C., Ravindran V.R. et al. Comparison of different ANN techniques for automatic defect detection in X-Ray images. 2009 International Conference on Emerging Trends in Electronic and Photonic Devices & Systems. 2009; 193–197. https://doi.org/10.1109/ELECTRO.2009.5441138
6. Morozov S.P., Kuzmina E.S., Ledikhova N.V., Vladzymyrskyy A.V., Trofimenko I.A., Mokienko O.A., Panina E.V., Andreychenko A.E., Omelyanskaya O.V., Gombolevskiy V.A., Polishchuk N.S., Shulkin I.M., Reshetnikov R.V. Mobilizing the academic and practical potential of diagnostic radiology during the COVID-19 pandemic in Moscow. Digital Diagnostics. 2020; 1 (1): 5–11. https://doi.org/10.17816/DD51043 (In Russian)
7. Order of the Ministry of Health of the Russian Federation of November 30, 2017 N 965n “About approval of the procedure for organizing and providing medical care using telemedicine technologies” base.garant.ru [Electronic resource]. URL: https://base.garant.ru/71851294/ (In Russian)
8. Order of the Department of Health of the Moscow city of April 1, 2020 №323 “About the creation of a remote radiodiagnosis reference center” // tele-med.ai [Electronic resource]. URL: https://tele-med.ai/biblioteka-dokumentov/dzm-prikaz-01042020-323-distanc-referens-centr-ld (In Russian)
9. Order of the Department of Health of the Moscow city of March 15, 2018 № 183 “About approval of the regulations for the organization of the provision of medical care in the profiles “roentgenology” and “radiology” using telemedicine technologies” // consultant.ru [Electronic resource]. URL : http://www.consultant.ru/cons/cgi/online.cgi?req=doc&base=MLAW&n=184819#vjWHo7T82RuFXi2g1 (In Russian)
10. Order of the Department of Health of the Moscow city of December 31, 2019 №1160 “About approval of the Regulations for Data Registration in the Unified Radiological Information Service of the Automated Information System of the Moscow city “Unified Medical Information and Analytical System of the Moscow City””. mos.ru [Electronic resource]. URL: https://www.mos.ru/dzdrav/documents/department-acts/view/233696220/ (In Russian)
11. Morozov S.P., Ledikhova N.V., Panina E.V. et al. Performance quality of X-ray technicians when they interact remotely with the reference center for diagnostic radiology using telemedicine technologies. Natsional'noe Zdravookhranenie. 2021; 2 (2): 36–46.
12. Rajkomar A., Lingam S., Taylor A.G. et al. High-Throughput Classification of Radiographs Using Deep Convolutional Neural Networks. J. Digit. Imaging. 2017; 30 (1): 95–101. https://doi.org/10.1007/s10278-016-9914-9
13. CheXpert Dataset //URL: https://stanfordmlgroup.github.io/competitions/chexpert/ (23.03.2022)
14. Chest X-rays dataset. URL: https://www.kaggle.com/datasets/raddar/chest-xrays-indiana-university (26.03.2022)
15. Chest X-Ray Images (Pneumonia). URL: https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia (20.12.2021)
16. NIH ChestX-ray14. URL: https://nihcc.app.box.com/v/ChestXray-NIHCC (20.12.2021)
17. Order of the Department of Health of the Moscow city of December 25, 2017 № 918 “About the regulation of registration of data in the system “United radiological information service” in medical organizations of the state health system of the Moscow city” // consultant.ru [Electronic resource]. URL: http://www.consultant.ru/cons/cgi/online.cgi?req=doc&base=MLAW&n=183590#zudlo7TsbVSwYBM12 (In Russian)
18. Simonyan K., Zisserman A. Very deep convolutional networks for large-scale image recognition. 3rd Int. Conf. Learn. Represent. ICLR 2015. Conf. Track Proc. 2015: 1–14. https://doi.org/10.48550/arXiv.1409.1556
19. https://roc-analysis.mosmed.ai/ ROC analysis tool of Scientific and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow City Health Department // URL: https://roc-analysis.mosmed.ai/
20. Chan H.P., Samala R.K., Hadjiiski L.M., Zhou C. Deep Learning in Medical Image Analysis. Adv. Exp. Med. Biol. 2020; 1213: 3–21. https://doi.org/10.1007/978-3-030-33128-3_1
21. Litjens G., Kooi T., Bejnordi B.E. et al. A survey on deep learning in medical image analysis. Med. Image Anal. 2017; 42: 60–88. https://doi.org/10.1016/j.media.2017.07.005
22. Sahiner B., Pezeshk A., Hadjiiski L.M. et al. Deep learning in medical imaging and radiation therapy. Med. Phys. 2019; 46 (1): 1–36. https://doi.org/10.1002/mp.13264
23. Mazurowski M.A., Buda M., Saha A., Bashir M.R. Deep learning in radiology: An overview of the concepts and a survey of the state of the art with focus on MRI. J. Magn. Reson. Imaging. 2019; 49 (4) : 939–954. https://doi.org/10.1002/jmri.26534
24. Rahaman M.M., Li C., Yao Y. et al. Identification of COVID-19 samples from chest X-Ray images using deep learning: A comparison of transfer learning approaches. J. Xray Sci. Technol. 2020; 28 (5): 821–839. https://doi.org/10.3233%2FXST-200715
25. Arias-Garzón D., Alzate-Grisales J.A., Orozco-Arias S. et al. COVID-19 detection in X-ray images using convolutional neural networks. Mach. Learn Appl. 2021; 6: 100–138. https://doi.org/10.1016/j.mlwa.2021.100138
Review
For citations:
Borisov A.A., Semenov S.S., Arzamasov K.M. Transfer Learning for automated search for defects on chest X-rays. Medical Visualization. 2023;27(1):158-169. (In Russ.) https://doi.org/10.24835/1607-0763-1243
Views:
702
Email this article 