ESTRO 2024 - Abstract Book

S4485

Physics - Machine learning models and clinical applications

ESTRO 2024

Conclusion:

For the purpose of detecting subtle changes in irradiated animals occurring early after radiotherapy, a transfer learning strategy with a multi head attention model with twelve attention heads was developed and tested on T2 weighted MR images. With a classification accuracy of 81%, the model shows a great potential in identifying early radiotherapy-induced changes. Further work is needed to evaluate model explainability and clinical potential.

Keywords: Preclinical research

References:

[1] F. Siddiqui and B. Movsas, "Management of Radiation Toxicity in Head and Neck Cancers," (in eng), Semin Radiat Oncol, vol. 27, no. 4, pp. 340-349, Oct 2017, doi: 10.1016/j.semradonc.2017.04.008.

[2] I. S. Juvkam et al., "A preclinical model to investigate normal tissue damage following fractionated radiotherapy to the head and neck," Journal of Radiation Research, vol. 64, no. 1, pp. 44-52, 2022, doi: 10.1093/jrr/rrac066.

[3] S. M. Pizer, R. E. Johnston, J. P. Ericksen, B. C. Yankaskas, and K. E. Muller, "Contrast-limited adaptive histogram equalization: speed and effectiveness," in [1990] Proceedings of the First Conference on Visualization in Biomedical Computing, 22-25 May 1990, pp. 337-345, doi: 10.1109/VBC.1990.109340. [4] J. Varghese, S. Subhash, K. Subramaniam, and K. P. Sridhar, "Adaptive Gaussian notch filter for removing periodic noise from digital images," IET Image Processing, vol. 14, no. 8, pp. 1529-1538, 2020. doi: https://doi.org/10.1049/iet-ipr.2018.5707

[5] A. Dosovitskiy et al., "An image is worth 16x16 words: Transformers for image recognition at scale," arXiv preprint arXiv:2010.11929, 2020. doi:https://doi.org/10.48550/arXiv.2010.11929

[6] F. Zhuang et al., "A comprehensive survey on transfer learning," Proceedings of the IEEE, vol. 109, no. 1, pp. 43 76, 2020. doi: 10.1109/JPROC.2020.3004555

1545

Digital Poster

Introducing a co-creation methodology for the design of an AI-based error detection in radiotherapy

Luca M Heising 1,2 , Frank Verhaegen 1 , Viola Mottarella 3 , Yin-Ho Chong 3 , Bas Nijsten 1 , Ans Swinnen 1 , Michel Öllers 1 , Maria J.G. Jacobs 1,2 , Carol X.J. Ou 2 , Cecile J.A. Wolfs 1 1 GROW, Radiation Oncology (Maastro), Maastricht, Netherlands. 2 Tilburg University, Department of Management, Tilburg, Netherlands. 3 Varian, a Siemens Healthineers Company, Medical Systems Imaging Laboratory, Baden, Switzerland