Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition
9 May 2026 – 14 May 2026 · Cape Town, South Africa
660-04-009 ISMRM Abstract

From YouTube to MRI Reconstruction: Overcoming Data Scarcity with Physics-Informed Video Pre-Training

Primary: Acquisition & Reconstruction - Image Reconstruction: AI
Secondary: Acquisition & Reconstruction - AI methods
660-04-009 · AI: Anything Synthetic or Correcting Artifacts · Thursday, 14 May, 2:35 PM–3:30 PM · Digital Posters Row A
Keywords: AI/ML Image Reconstruction Domain adaptation Data Scarcity Physics-informed Deep Learning Video-based Pre-training
Accepted
Simon Graf 1,2, Walter A Wohlgemuth1,2, Andreas Deistung1,2
1Medical Physics Group, University Clinic and Outpatient Clinic for Radiology, University Hospital Halle (Saale), Germany
2Halle MR Imaging Core Facility, Medical Faculty, Martin-Luther-University Halle-Wittenberg, Germany
Presenting Author: Simon Graf

Synopsis

Motivation:
Goals:
Approach:
Results:
Full abstract & presentation

The full text, figures, and any recorded presentation for this abstract are not shown here. Log in if you are a member or registered attendee with access.

Full abstracts, figures, and presentations for Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition are available to registered attendees. This content becomes freely available to the public roughly two years after the meeting.

To request or purchase access, contact the ISMRM Central Office at info@ismrm.org.

Log in

References

1. Deistung A, Schweser F, Reichenbach JR. Overview of quantitative susceptibility mapping. NMR Biomed. 2017;30(4). doi: 10.1002/nbm.3569. [doi]
2. Kay W, Carreira J, Simonyan K, et al. The Kinetics Human Action Video Dataset; 2017. Available at: https://arxiv.org/abs/1705.069501705.06950v11705.06950v1.
3. Graf S, Wohlgemuth WA, Deistung A. Incorporating a-priori information in deep learning models for quantitative susceptibility mapping via adaptive convolution. Front Neurosci. 2024;18:1366165. doi: 10.3389/fnins.2024.1366165. [doi]
4. Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O. 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation; 2016. Available at: http://arxiv.org/pdf/1606.06650v11606.06650v11606.06650v1.
5. Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov. Dropout: A Simple Way to Prevent Neural Networks from Overfitting. Journal of Machine Learning Research. 2014(15):1929 -1958. http://jmlr.org/papers/v15/srivastava14a.htmldoi:
6. Loshchilov I, Hutter F. Decoupled Weight Decay Regularization; 2017. Available at: http://arxiv.org/pdf/1711.05101v31711.05101v31711.05101v3.
7. Graf S, Wohlgemuth W, Deistung A. The Role of Optimizing Loss Functions in Transfer Learning to Address Data Scarcity. In: 2025 ISMRM & ISMRT Annual Meeting; 2025. ISMRM Annual Meeting.
8. Shi Y, Feng R, Li Z, Zhuang J, Zhang Y, Wei H. Towards in vivo ground truth susceptibility for single-orientation deep learning QSM: A multi-orientation gradient-echo MRI dataset. NeuroImage. 2022;261:119522. doi: 10.1016/j.neuroimage.2022.119522. [doi]
9. Xiong Z, Gao Y, Liu F, Sun H. Affine transformation edited and refined deep neural network for quantitative susceptibility mapping. NeuroImage. 2023;267:119842. doi: 10.1016/j.neuroimage.2022.119842. [doi]
10. Schweser F, Sommer K, Deistung A, Reichenbach JR. Quantitative susceptibility mapping for investigating subtle susceptibility variations in the human brain. NeuroImage. 2012;62(3):2083-2100. doi: 10.1016/j.neuroimage.2012.05.067. [doi]
11. Zhou Z, Sodha V, Pang J, Gotway MB, Liang J. Models Genesis. Medical image analysis. 2021;67:101840. doi: 10.1016/j.media.2020.101840. [doi]
12. Azizi S, Mustafa B, Ryan F, et al. Big Self-Supervised Models Advance Medical Image Classification; 2021. Available at: http://arxiv.org/pdf/2101.05224v22101.05224v22101.05224v2.
13. Malik N, Bzdok D. From YouTube to the brain: Transfer learning can improve brain-imaging predictions with deep learning. Neural networks : the official journal of the International Neural Network Society. 2022;153:325-338. doi: 10.1016/j.neunet.2022.06.014. [doi]
14. Ke A, Huang S-C, O’Connell CP, Klimont M, Yeung S, Rajpurkar P. Video Pretraining Advances 3D Deep Learning on Chest CT Tasks; 2023. Available at: http://arxiv.org/pdf/2304.00546v12304.00546v12304.00546v1.
15. Dhinagar NJ, Singh A, Ozarkar S, et al. Video and Synthetic MRI Pre-training of 3D Vision Architectures for Neuroimage Analysis; 2023. Available at: http://arxiv.org/pdf/2309.04651v12309.04651v12309.04651v1.
16. Tayebi Arasteh S, Misera L, Kather JN, Truhn D, Nebelung S. Enhancing diagnostic deep learning via self-supervised pretraining on large-scale, unlabeled non-medical images. European radiology experimental. 2024;8(1):10. doi: 10.1186/s41747-023-00411-3. [doi]

Cite this abstract

http://echo.ismrm.org/p/ISMRM2026/660-04-009