Active learning of the ground truth for retinal image segmentation

Nedoshivina, L.S., Lensu Lasse

Full text «Opticheskii Zhurnal»

Full text on elibrary.ru

Publication in Journal of Optical Technology

For Russian citation (Opticheskii Zhurnal):

L. Nedoshivina, L. Lensu Active learning of the ground truth for retinal image segmentation (Активное машинное обучение в задаче формирования обучающей выборки для автоматического анализа изображений сетчатки глаза) [на англ. яз.] // Оптический журнал. 2019. Т. 86. № 11. С. 29–36. http://doi.org/10.17586/1023-5086-2019-86-11-29-36

L. Nedoshivina, L. Lensu Active learning of the ground truth for retinal image segmentation (Активное машинное обучение в задаче формирования обучающей выборки для автоматического анализа изображений сетчатки глаза) [in English] // Opticheskii Zhurnal. 2019. V. 86. № 11. P. 29–36. http://doi.org/10.17586/1023-5086-2019-86-11-29-36

For citation (Journal of Optical Technology):

L. Nedoshivina and L. Lensu, "Active learning of the ground truth for retinal image segmentation," Journal of Optical Technology. 86(11), 697-703 (2019). https://doi.org/10.1364/JOT.86.000697

Abstract:

Different diseases can be diagnosed from eye fundus images by medical experts. Automated diagnosis methods can help medical doctors to increase the diagnosis accuracy and decrease the time needed. In order to have a proper dataset for training and evaluating the methods, a large set of images should be annotated by several experts to form the ground truth. To enable efficient utilization of expert’s time, active learning is studied to accelerate the collection of the ground truth. Since one of the important steps in the retinal image diagnosis is the blood vessel segmentation, the corresponding approaches were studied. Two approaches were implemented and extended by proposed active learning methods for selecting the next image to be annotated. The performance of the methods in the case of standard implementation and active learning application was compared for several retinal images databases.

Keywords:

computer aided diagnosis, active learning, retinal image segmentation

OCIS codes: 100.4996, 100.4993, 100.7410, 100.2960, 110.7410, 110.3080, 170.5755

References:

1. Faust O., Acharya R., Ng E.Y.K., et al. Algorithms for the automated detection of diabetic retinopathy using digital fundus images: A review // J. Medical Systems. 2012. V. 36. № 1. P. 145–157.
2. Gulshan V., Peng L., Coram M., et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs // JAMA. 2016. V. 316. № 22. P. 2402–2410.
3. Shalev-Shwartz S. Online learning and online convex optimization // Foundations and Trends in Machine Learning. 2012. V. 4. № 2. P. 107–194.
4. Settles B. Active learning // Synthesis Lectures on Artificial Intelligence and Machine Learning. 2012. V. 6. № 1. P. 1–114.
5. Lewis D.D., Gale W.A. A sequential algorithm for training text classifiers // Proc. 17th Annual Internat. ACM SIGIR Conf. Research and Development in Information Retrieval. N.Y.: Springer-Verlag, Inc, 1994. P. 3–12.
6. Sánchez C.I., Niemeijer M., Kockelkorn T., et al. Active learning approach for detection of hard exudates, cotton wool spots, and drusen in retinal images // Medical Imaging 2009: Computer-Aided Diagnosis. 2009. V. 7260. P. 72601I.
7. Sánchez C.I., Niemeijer M., Abràmoff M.D., et al. Active learning for an efficient training strategy of computer-aided diagnosis systems: Application to diabetic retinopathy screening // Internat. Conf. Medical Image Computing and Computer-Assisted Intervention. 2010. P. 603–610.
8. Hoover A., Goldbaum M. Locating the optic nerve in a retinal image using the fuzzy convergence of the blood vessels // IEEE Trans. Medical Imaging. 2003. V. 22. № 8. P. 951–958.
9. Staal J., Abràmoff M.D., Niemeijer M., et al. Ridge-based vessel segmentation in color images of the retina // IEEE Trans. Medical Imaging. 2004. V. 23. № 4. P. 501–509.

10. Owen C.G., Rudnicka A.R., Mullen R., et al. Measuring retinal vessel tortuosity in 10-year-old children: Validation of the computer-assisted image analysis of the retina (CAIAR) program // Investigative Ophthalmology & Visual Science. 2009. V. 50. № 5. P. 2004–2010.
11. Farnell D.J., Hatfield F., Knox P., et al. Enhancement of blood vessels in digital fundus photographs via the application of multiscale line operators // J. Franklin Institute. 2008. V. 345. № 7. P. 748–765.
12. Soares J.V., Leandro J.J., Cesar R.M., et al. Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification // IEEE Trans. Medical Imaging. 2006. V. 25. № 9. P. 1214–1222.
13. Ronneberger O., Fischer P., Brox T. U-net: Convolutional networks for biomedical image segmentation // Internat. Conf. Medical Image Computing and Computer-Assisted Intervention. Springer, 2015. P. 234–241.
14. Badrinarayanan V., Kendall A., Cipolla R. Segnet: A deep convolutional encoder-decoder architecture for image segmentation // IEEE Trans. Pattern Analysis and Machine Intelligence. 2017. V. 39. № 12. P. 2481–2495.
15. Kendall A., Badrinarayanan V., Cipolla R. Bayesian segnet: Model uncertainty in deep convolutional encoder-decoder architectures for scene understanding // 28th British Machine Vision Conf., 2017.
16. Sorenson T. A method of establishing groups of equal amplitude in plant sociology based on similarity of species content // K Dan Vidensk Selsk Biol Skr. 1948. V. 5. P. 1–34.
17. Vostatek P., Claridge E., Uusitalo H., et al. Performance comparison of publicly available retinal blood vessel segmentation methods // Computerized Medical Imaging and Graphics. 2017. V. 55. P. 2–12.
18. Gorriz M. Cost-effective active learning for melanoma segmentation // arXiv preprint arXiv:1711.09168. 2017.