|
DOI: 10.14489/vkit.2025.04.pp.012-020
Нишчхал Н.
СИММЕТРИЧНАЯ РЕГИСТРАЦИЯ МУЛЬТИМОДАЛЬНЫХ МЕДИЦИНСКИХ ИЗОБРАЖЕНИЙ НА ОСНОВЕ АНАТОМИЧЕСКОЙ ИНФОРМАЦИИ
(с. 12-20)
Аннотация. Представлен метод симметричной регистрации мультимодальных медицинских изображений на основе модели SymReg-GAN, усовершенствованный с использованием модулей анатомического внимания и многоуровневого извлечения признаков. Интеграция предварительно обученных моделей позволяет эффективно совмещать изображения различных модальностей с сохранением анатомической согласованности. Композитная функция потерь обеспечивает точность и реалистичность преобразований. Эксперименты на мультимодальных наборах данных брюшной полости подтверждают превосходство предложенного метода над современными подходами и демонстрируют его потенциал при анализе медицинских изображений.
Ключевые слова: мультимодальная регистрация; симметричная регистрация; глубокое обучение; GAN; анатомическое внимание.
Nishchhal N.
SYMMETRIC REGISTRATION OF MULTI-MODAL MEDICAL IMAGES BASED ON ANATOMICAL INFORMATION
(pp. 12-20)
Abstract. This paper introduces a multi-modal symmetric image registration method leveraging the advanced SymReg-GAN architecture. The framework is specifically designed to address the challenges of aligning medical images from different modalities, such as CT and MRI, which often vary significantly in appearance while representing the same anatomical structures. The proposed method incorporates an Anatomical Attention Module (AAM), which focuses on preserving the anatomical coherence of key structures during the registration process. This ensures that critical regions are accurately aligned without losing important contextual information. Furthermore, the architecture integrates multi-scale feature extraction, enabling the model to capture both fine-grained and large-scale anatomical details across different resolution levels. By utilizing pre-trained models, the system benefits from enhanced feature representation and improved generalization, allowing for more robust performance on various datasets. A composite loss function is employed to optimize the registration process, balancing multiple objectives, including intensity similarity, structural alignment, and transformation regularization. This ensures not only precise alignment but also the generation of realistic and anatomically plausible transformations. The method's efficacy is demonstrated through extensive experiments on multi-modal abdominal datasets, where it consistently outperforms state-of-the-art techniques in terms of accuracy and robustness. The results highlight its potential to significantly advance medical image analysis, facilitating better diagnosis and treatment planning. This work contributes to the field by offering a comprehensive solution for multi-modal image registration, addressing both the complexity of medical imaging and the need for reliable, anatomically coherent alignment methods.
Keywords: Multi-modal registration; Symmetric registration; Deep learning; GAN; Anatomical attention.
Н. Нишчхал (Сибирский государственный университет науки и технологий имени академика М. Ф. Решетнева, Красноярск, Россия) E-mail:
Этот e-mail адрес защищен от спам-ботов, для его просмотра у Вас должен быть включен Javascript
N. Nishchhal (Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, Russia) E-mail:
Этот e-mail адрес защищен от спам-ботов, для его просмотра у Вас должен быть включен Javascript
1. Bharati S., Mondal M.R.H., Podder P., Prasath V.B.S. Deep learning for medical image registration: A comprehensive review // International Journal of Computer Information Systems and Industrial Management Applications. 2022. V. 14. P. 173–190.
2. Darzi F., Bocklitz T. A Review of Medical Image Registration for Different Modalities // Bioengineering. 2024.V.11, No. 8. Art. 786.
3. Learning deep abdominal CT registration through adaptive loss weighting and synthetic data generation / J. Pérez de Frutos, A. Pedersen, E. Pelanis et al. // PLoS ONE. 2023. V. 18, No. 2. P. e0282110.
4. Sinha S., Tiwari R. Abdominal images nonrigid registration using localaffine transformations // Medical Image Computing and Computer-Assisted Intervention. 2020. P. 123–134.
5. Azam M.A., Khan K.B., Ahmad M., Mazzara M. Multimodal medical image registration and fusion for quality enhancement // Computational Materials and Continua. 2021. V. 68, No. 1. P. 821–840.
6. Blendowski M., Hansen L., Heinrich M.P. Weakly-supervised learning of multimodal features for regularised iterative descent in 3D image registration // Medical Image Analysis. 2021. V. 67. Art. 101822.
7. Geometry-consistent adversarial registration model for unsupervised multimodal medical image registration / Y. Liu, W. Wang, Y. Li et al. // IEEE Journal of Biomedical and Health Informatics. 2023. V. 27, No. 7. P. 3455–3466. DOI: 10.1109/JBHI.2023.3270199
8. Deep learning based synthesis of MRI, CT and PET: Review and analysis / S. Dayarathna, K. T. Islam, S. Uribe et al. // Medical Image Analysis. 2024. V. 92. Art. 103046. DOI: 10.1016/j.media.2023.103046
9. SymReg-GAN: Symmetric image registration with generative adversarial networks / Y. Zheng, X. Sui, Y. Jiang et al. // IEEE Transactions on Pattern Analysis and Machine Intelligence. 2022. V. 44, No. 9. P. 5631–5646.
10. Strittmatter A., Zöllner F.G. Multistep networks for deformable multimodal medical image registration // IEEE Access. 2024. V. 12. P. 82676–82692. DOI: 10.1109/ACCESS.2024.3412216
11. Mao-Yang Z., Hao Y., Guang-Hui P. Research progress and challenges of deep learning in medical image registration // Journal of Biomedical Engineering. 2019. V. 36, No. 4. P. 677–683.
12. Zhou S.K., Greenspan H., Davatzikos C., Summers R.M. A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises // Proceedings of the IEEE. 2021. V. 109, No. 5. P. 820–838.
13. Generation of annotated multimodal ground truth datasets for abdominal medical image registration / D. Bauer, T. Russ, B. Waldkirch et al. // International Journal of Computer Assisted Radiology and Surgery. 2021. V. 16, No. 8. P. 1277–1285.
14. CHAOS – combined (CT-MR) healthy abdominal organ segmentation challenge data (Version v1.03) [Data set] / A.E. Kavur, M.A. Selver, O. Dicle et al. // Zenodo. 2019. DOI: 10.5281/zenodo.3362844
1. Bharati S., Mondal M. R. H., Podder P., Prasath V. B. S. (2022). Deep learning for medical image registration: A comprehensive review. International Journal of Computer Information Systems and Industrial Management Applications, 14, 173 – 190.
2. Darzi F., Bocklitz T. (2024). A Review of Medical Image Registration for Different Modalities. Bioengineering, 11(8), Art. no. 786.
3. Pérez de Frutos J., Pedersen A., Pelanis E. et al. (2023). Learning deep abdominal CT registration through adaptive loss weighting and synthetic data generation. PLoS ONE, 18(2).
4. Sinha S., Tiwari R. (2020). Abdominal images nonrigid registration using localaffine transformations. Medical Image Computing and Computer-Assisted Intervention, 123 – 134.
5. Azam M. A., Khan K. B., Ahmad M., Mazzara M. (2021). Multimodal medical image registration and fusion for quality enhancement. Computational Materials and Continua, 68(1), 821 – 840.
6. Blendowski M., Hansen L., Heinrich M. P. (2021). Weakly-supervised learning of multi-modal features for regularised iterative descent in 3D image registration. Medical Image Analysis, 67, Art. 101822.
7. Liu Y., Wang W., Li Y. et al. (2023). Geometry-consistent adversarial registration model for unsupervised multi-modal medical image registration. IEEE Journal of Biomedical and Health Informatics, 27(7), 3455 – 3466. DOI: 10.1109/JBHI.2023.3270199
8. Dayarathna S., Islam K. T., Uribe S. et al. (2024). Deep learning based synthesis of MRI, CT and PET: Review and analysis. Medical Image Analysis, 92, Art. 103046. DOI: 10.1016/j.media.2023.103046
9. Zheng Y., Sui X., Jiang Y. et al. (2022). SymReg-GAN: Symmetric image registration with generative adversarial networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(9), 5631 – 5646.
10. Strittmatter A., Zöllner F. G. (2024). Multistep networks for deformable multimodal medical image registration. IEEE Access, 12, 82676 – 82692. DOI: 10.1109/ACCESS.2024.3412216
11. Mao-Yang Z., Hao Y., Guang-Hui P. (2019). Research progress and challenges of deep learning in medical image registration. Journal of Biomedical Engineering, 36(4), 677 – 683.
12. Zhou S. K., Greenspan H., Davatzikos C., Summers R. M. (2021). A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises. Proceedings of the IEEE, 109(5), 820 – 838.
13. Bauer D., Russ T., Waldkirch B. et al. (2021). Generation of annotated multimodal ground truth datasets for abdominal medical image registration. International Journal of Computer Assisted Radiology and Surgery, 16(8), 1277 – 1285.
14. Kavur A. E., Selver M. A., Dicle O. et al. (2019). CHAOS – combined (CT-MR) healthy abdominal organ segmentation challenge data (Version v1.03). Zenodo. DOI: 10.5281/zenodo.3362844
|
Архив номеров
Разработка концепции и создание сайта - ООО «Издательский дом «СПЕКТР»