publication

Synthetic CT generation for Head and Neck radiotherapy by a 3D convolutional neural network

Dinkla, A., Florkow, M., Maspero, M., Savenije, M., Zijlstra, F., Doornaert, P., Van Stralen, M., Philippens, M., Seevinck, P., Van den Berg, N.

Radiotherapy & Oncology 133 p. S268-S269

Abstract

Purpose or Objective Synthetic CT generation is required for MR-only simulation workflow where CT-MRI registration can be avoided. More recently, also sCTs are becoming desirable for MRI-guided Radiotherapy where planning is performed based on daily MR images. In the Head and Neck region (H&N), atlas-based methods have been adopted. However, their robustness is limited as in H&N abnormal patient anatomies can occur due to large tumors or surgical excision. Therefore, here a patch-based deep learning method was chosen to improve robustness. In particular, we used a 3D patch-based convolutional neural network (CNN) to generate sCTs based on T2-weighted Turbo Spin echo (TSE) images and evaluated its image and dosimetric performance. Material and Methods We conducted a retrospective study on 34 patients with Head and Neck cancer who underwent CT (Philips Brilliance Big Bore) and MR imaging (3T Philips Ingenia) for radiotherapy simulation. To generate the sCTs, a large field-of-view (FOV) transverse T2-w TSE mDixon MRI, originally used for tumor/OAR contouring, was selected from the clinical protocol. 83 transverse slices with 3 mm thickness were acquired with a FOV of 45x45 cm2 and 0.94x0.94 mm2 resolution in 5min24s and readout bandwidth of 876 Hz/px. Cases with severe image artefacts from dental implants (CT) or motion (MRI) were excluded from the training. To align images for training and evaluation, CT scans were non-rigidly registered (CTreg) to the in-phase MR images (Elastix 4.7) and all images were resampled to 1x1x1 mm3 isotropic resolution. The CNN was based on a U-net architecture and consisted of 14 layers with 3x3x3 filters. Patches of 48x48x48 were randomly extracted and fed into the training. sCTs were created for all patients using three-fold cross validation. The CT-based treatment plan was recalculated on sCT using Monaco TPS (Elekta). Results sCT generation took 4 min. on a single GPU. The patchbased approach allowed proper sCT generation for nonstandard anatomies (fig. 1). Mean absolute error (MAE) over the patient population of the sCT within the intersection of body contours was 75±9 HU, and the mean error (ME) was 9±11 HU. Dice scores of the air (<-200HU) and bone (>250HU) masks (CTreg vs sCT) were 0.79±0.08 and 0.70±0.07 respectively. Dosimetric analysis showed mean differences of -0.03±0.05% for dose within the body contours and -0.07±0.22% inside the high dose region (dose >90%). Dental artefacts obscuring the CT, could be circumvented in the sCT by the CNN-based approach in combination with TSE MRI sequence that typically is less prone to susceptibility artefacts (fig. 2). Conclusion The 3D patch-based CNN generated sCTs of the H&N region that were dosimetrically accurate. The sCT were generated based on T2W TSE images already used for tumor/OAR contouring and thus no extra scan time was added. Moreover, for H&N cancers, the use of TSE as input for sCT generation has as particular advantage that it is less affected by dental artefacts compared to commonly used gradient echo sequences.