Comparison of Patch-Based Conditional Generative Adversarial Neural Net Models with Emphasis on Model Robustness for Use in Head and Neck Cases for MR-Only planning

TL;DR

This study compares two conditional GAN models, Pix2Pix and Cycle GAN, for generating synthetic CT images from MR images in head and neck cancer cases, evaluating their accuracy and robustness for MR-only radiotherapy planning.

Contribution

It provides a comparative analysis of Pix2Pix and Cycle GAN models specifically for synthetic CT generation in challenging clinical scenarios with artifacts and abnormal anatomy.

Findings

Pix2Pix achieved lower MAE than Cycle GAN in test cases.

Both models produced dosimetric errors below 2%.

Generated DRRs were qualitatively similar to real CT-based DRRs.

Abstract

A total of twenty paired CT and MR images were used in this study to investigate two conditional generative adversarial networks, Pix2Pix, and Cycle GAN, for generating synthetic CT images for Headand Neck cancer cases. Ten of the patient cases were used for training and included such common artifacts as dental implants; the remaining ten testing cases were used for testing and included a larger range of image features commonly found in clinical head and neck cases. These features included strong metal artifacts from dental implants, one case with a metal implant, and one case with abnormal anatomy. The original CT images were deformably registered to the mDixon FFE MR images to minimize the effects of processing the MR images. The sCT generation accuracy and robustness were evaluated using Mean Absolute Error (MAE) based on the Hounsfield Units (HU) for three regions (whole body, bone, and air within the body), Mean Error (ME) to observe systematic average offset errors in the sCT generation, and dosimetric evaluation of all clinically relevant structures. For the test set the MAE for the Pix2Pix and Cycle GAN models were 92.4 $\pm$ 13.5 HU, and 100.7 $\pm$ 14.6 HU, respectively, for the body region, 166.3 $\pm$ 31.8 HU, and 184 $\pm$ 31.9 HU, respectively, for the bone region, and 183.7 $\pm$ 41.3 HU and 185.4 $\pm$ 37.9 HU for the air regions. The ME for Pix2Pix and Cycle GAN were 21.0 $\pm$ 11.8 HU and 37.5 $\pm$ 14.9 HU, respectively. Absolute Percent Mean/Max Dose Errors were less than 2% for the PTV and all critical structures for both models, and DRRs generated from these models looked qualitatively similar to CT generated DRRs showing these methods are promising for MR-only planning.