Meeting Banner
Abstract #3792

CT synthesis for MR-only brain radiotherapy treatment planning using convolutional neural networks

Anna M. Dinkla1, Jelmer M. Wolterink2, Matteo Maspero1, Mark H.F. Savenije1, Joost J.C. Verhoeff1, Ivana Isgum2, Peter R. Seevinck2, Jan J.W. Lagendijk1, and Cornelis A.T. van den Berg1

1Radiation Oncology, UMC Utrecht, Utrecht, Netherlands, 2Image Sciences Institute, UMC Utrecht, Utrecht, Netherlands

In MR-only radiotherapy, a synthetic CT (sCT) needs to be generated from MR to allow radiation dose planning without CT acquisition. We aim for such an MR-only workflow to decrease radiotherapy preparation time. If we can plan and deliver the dose in a single day, in a ‘one stop shop’ procedure, we prevent treatment delay which is of high importance for patients with fast growing brain metastases. In this study sCTs of the head were generated in a short amount of time, that highly resemble the original CTs and that enable accurate dose calculation.

This abstract and the presentation materials are available to members only; a login is required.

Join Here