Nowadays, intelligent systems are widely used in the field of medicine. Especially relevant is the problem of developing intelligent computer-aided diagnostics (CAD) systems which can be used as an auxiliary tool to improve specialist’s efficiency in the context of the growing volume of medical data requiring analysis and processing. One of the important components of modern CAD systems is the module for recognizing pathological changes in medical images. The paper considers the problem of training a convolutional neural network to recognize cerebral vascular aneurysms. The architecture of a fully convolutional neural network based on the UNet architecture, a data preprocessing technique, a technique for constructing a seamless prediction based on the separation of the original image into a set of intersecting fragments are proposed. The influence of the size of image fragments used for training on the effectiveness of neural network training was investigated. Drawing on the statistical analysis of the results of the conducted computational experiments, it was concluded that the size of the fragment is not a determining parameter since no increase in recognition accuracy is observed with its increase. At the same time, experiments have shown that increasing the batch size while fixing the remaining parameters at the same level can significantly improve the recognition accuracy.
1. Doronicheva A. V., Savin S. Z. Methods of medical image recognition for computer-aided diagnostics. Sovremennye problemy nauki i obrazovanija = Modern problems of science and education. 2014;(4). (In Russ.).
2. Berestov V.V. Embolization of cerebral aneurysms in case of an acute hemorrhagic stroke. Candidate’s dissertation. Novosibirsk; 2021. 138 p.
3. Park A., Chute C., Rajpurkar P., Lou J., Ball R.L., Shpanskaya K., et al. Deep Learning-Assisted Diagnosis of Cerebral Aneurysms Using the HeadXNet Model. JAMA network open. 2019;2(6):e195600. Available from: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2735471.
4. Shi Z., Miao C., Schoepf U.J., Savage R.H., Dargis D.M., Pan C, et al. A clinically applicable deep-learning model for detecting intracranial aneurysm in computed tomography angiography images. Nature Communications. 2020;11(1):1–11. DOI: 10.1038/s41467-020-19527-w.
5. Zhu G., Luo X., Yang T., Cai L., Yeo J.H., Yan G., et al. Deep learning-based recognition and segmentation of intracranial aneurysms under small sample size. Frontiers in Physiology. 2022;13:2580.
6. Cicek O., Abdulkadir A., Lienkamp S.S., Brox T., Ronneberger O. 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation. Medical Image Computing and Computer-Assisted Intervention – MICCAI. 2016. Available from: http://arxiv.org/abs/1606.06650.
7. Milletari F., Navab N., Ahmadi S.A. V-net: Fully convolutional neural networks for volumetric medical image segmentation. 2016 fourth international conference on 3D vision (3DV). 2016:565–571.
8. Kerfoot E., Clough J., Oksuz I., Lee J., King A.P., Schnabel J.A. Left-ventricle quantification using residual U-Net. Statistical Atlases and Computational Models of the Heart. Atrial Segmentation and LV Quantification Challenges: 9th International Workshop, STACOM 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16, 2018, Revised Selected Papers 9. 2019:371–380.
9. Brutzkus A., Globerson A., Malach E., Netser A.R., Shalev-Schwartz S. Efficient Learning of CNNs using Patch Based Features. International Conference on Machine Learning. 2022:2336–2356.
10. Long J., Shelhamer E., Darrell T. Fully convolutional networks for semantic segmentation. Proceedings of the IEEE conference on computer vision and pattern recognition. 2015:3431–3440.
11. CADA-Cerebral Aneurysm Detection. Grand Challenge; Available from: https://cada.grand-challenge.org/ (accessed on 20.09.2022).
12. Ronneberger O., Fischer P., Brox T. U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention (MICCAI). 2015;9351:234–241.
13. Oktay O., Schlemper J., Folgoc L.L., Lee M., Heinrich M., Misawa K., et al. Attention U-Net: Learning Where to Look for the Pancreas; 2018. Available from: https://arxiv.org/abs/1804.03999v3.
14. Su Z., Jia Y., Liao W., Lv Y., Dou J., Sun Z., et al. 3D attention U-Net with pretraining: a solution to CADA-Aneurysm segmentation challenge. Cerebral Aneurysm Detection and Analysis: First Challenge, CADA 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 8, 2020, Proceedings 1. 2021:58–67.
15. Kruzhalov A., Philippovich A. Analysis of Thresholding Methods for the Segmentation of Brain Vessels. In: Recent Trends in Analysis of Images, Social Networks and Texts: 10th International Conference, AIST 2021, Tbilisi, Georgia, December 16–18, 2021, Revised Selected Papers. 2022:85–95.
16. Shahzad R., Pennig L., Goertz L., Thiele F., Kabbasch C., Schlamann M., et al. Fully automated detection and segmentation of intracranial aneurysms in subarachnoid hemorrhage on CTA using deep learning. Scientific Reports. 2020;10(1):1–12.
