ESTRO 2024 - Abstract Book

S4524

Physics - Machine learning models and clinical applications

ESTRO 2024

Figure 1. Digital reconstructed radiographs (DRRs) for the two projections (A, B) of the CyberKnife image guide system calculated using the CT (a, b) and the synthetic CT (d, e) for an indicative case. Axial CT (c) and sCT (f) slices depicting the dose distributions calculated using the CT (solid lines) and sCT (broken line) are also shown.

Conclusion:

In this study, we have provided a compelling proof of concept that an AI-based approach to generate synthetic CTs offers a viable solution for an effective implementation in CyberKnife intracranial radiosurgery. AI based sCTs obviate the need for manual structure segmentation or acquisition of special MR sequences. However, prior to its clinical adoption, it is imperative that the MRI scanning protocol encompasses the entirety of the head & neck region, and further refinements should be made to the AI model to augment the contrast of the generated sCTs, thereby optimizing their utility in this critical medical application.

Keywords: MR only, synthetic CT, CyberKnife

2167

Digital Poster

Streamlining Radiotherapy Structure Nomenclature: A Multimodal Feature Learning Approach

Fahim Irfan Alam 1,2,3 , Matthew Field 1,2,4 , Daniel Al Mouiee 1,2,3 , Phillip Chlap 1,2 , Geoff P. Delaney 5,2,6 , Shalini Vinod 5,2,7 , Shivani Kumar 1,2 , Janet Cui 8 , Ali Haidar 1,2 , Vicky Chin 5,2 , Jonathan Sykes 9,10 , Simon Ashworth 9 , Verity Ahern 9,11 , Kirsty Stuart 9,12,11 , Michael Bailey 13 , Senthilkumar Gandhidasan 13 , Jothy Selvaraj 14 , Prabhakar Ramachandran 15,16 , Lois Holloway 1,2,10 1 Ingham Institute for Applied Medical Research, Medical Physics, Sydney, Australia. 2 University of New South Wales, South Western Sydney Clinical School, Sydney, Australia. 3 Liverpool and Macarthur Cancer Therapy Centres, Medical Physics, Sydney, Australia. 4 AI Medicine & Engineering Pty Ltd, Artificial Intelligence and Machine Learning, Sydney, Australia. 5 Ingham Institute for Applied Medical Research, Radiation Oncology, Sydney, Australia. 6 Collaboration for Cancer Outcomes, Research and Evalulation (CCORE), Radiation Oncology, Sydney, Australia. 7 Liverpool and Macarthur Cancer Therapy Centres, Radiation Oncology, Sydney, Australia. 8 University of New South Wales, Centre for Big Data Research, School of Medicine & Health, Sydney, Australia. 9 Western Sydney Local Health District, Sydney West Radiation Oncology Network, Sydney, Australia. 10 University of Sydney, Institute of Medical Physics, School of Physics, Sydney, Australia. 11 University of Sydney, Westmead Clinical School, Sydney, Australia. 12 Westmead Breast Cancer Institute, Radiation Oncology, Sydney, Australia. 13 ISLHD Illawarra Cancer Care Centre, Radiation Oncology, Sydney, Australia. 14 Canberra Health Services, Medical Physics and Radiation Engineering, Sydney, Australia. 15 Queensland University of Technology, School of Chemistry and Physics, Brisbane, Australia. 16 Princess Alexandra Hospital, Radiation Oncology, Brisbane, Australia

Purpose/Objective: