Material Decomposition in Spectral CT Using Deep Learning: A Sim2Real Transfer Approach
| Autor: | Abascal, Juan F P J, Abascal, Juan Fpj, Ducros, Nicolas, Pronina, Valeriya, Rit, Simon, Rodesch, Pierre-Antoine, Broussaud, Thomas, Bussod, Suzanne, Douek, Philippe, Hauptmann, Andreas, Arridge, Simon, PEYRIN, Françoise |
|---|---|
| Přispěvatelé: | Imagerie Tomographique et Radiothérapie, Centre de Recherche en Acquisition et Traitement de l'Image pour la Santé (CREATIS), Université Jean Monnet [Saint-Étienne] (UJM)-Hospices Civils de Lyon (HCL)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Jean Monnet [Saint-Étienne] (UJM)-Hospices Civils de Lyon (HCL)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National de la Santé et de la Recherche Médicale (INSERM) |
| Rok vydání: | 2021 |
| Předmět: |
General Computer Science
Computer science Computed tomography 02 engineering and technology Iterative reconstruction transfer learning Regularization (mathematics) Convolutional neural network 030218 nuclear medicine & medical imaging Spectral CT 03 medical and health sciences 0302 clinical medicine [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] [INFO.INFO-IM]Computer Science [cs]/Medical Imaging 0202 electrical engineering electronic engineering information engineering medicine General Materials Science Electrical and Electronic Engineering medicine.diagnostic_test business.industry Deep learning General Engineering deep learning Experimental data Decomposition Nonlinear system inverse problem 020201 artificial intelligence & image processing lcsh:Electrical engineering. Electronics. Nuclear engineering Artificial intelligence business lcsh:TK1-9971 [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing Algorithm |
| Zdroj: | IEEE Access IEEE Access, IEEE, In press, ⟨10.1109/ACCESS.2021.3056150⟩ IEEE Access, IEEE, 2021, 9, pp.25632-25647. ⟨10.1109/ACCESS.2021.3056150⟩ IEEE Access, Vol 9, Pp 25632-25647 (2021) |
| ISSN: | 2169-3536 |
| DOI: | 10.1109/access.2021.3056150 |
| Popis: | International audience; The state-of-the art for solving the nonlinear material decomposition problem in spectral computed tomogra-phy is based on variational methods, but these are computationally slow and critically depend on the particular choice of the regularization functional. Convolutional neural networks have been proposed for addressing these issues. However, learning algorithms require large amounts of experimental data sets. We propose a deep learning strategy for solving the material decomposition problem based on a U-Net architecture and a Sim2Real transfer learning approach where the knowledge that we learn from synthetic data is transferred to a real-world scenario. In order for this approach to work, synthetic data must be realistic and representative of the experimental data. For this purpose, numerical phantoms are generated from human CT volumes of the KiTS19 Challenge dataset, segmented into specic materials (soft tissue and bone). These volumes are projected into sinogram space in order to simulate photon counting data, taking into account the energy response of the scanner. The network is trained to decompose the materials in the projection domain after which we apply any conventional tomographic method to reconstruct the dierent material volumes. The proposed decomposition method is compared to a regularized Gauss-Newton (RGN) method on synthetic data, experimental phantom data and human thorax data. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|