Deep learning approaches using 2D and 3D convolutional neural networks for generating male pelvic synthetic CT from MRI.

Researchers

Journal

Modalities

Models

Abstract

The improved soft tissue contrast of magnetic resonance imaging (MRI) compared to computed tomography (CT) makes it a useful imaging modality for radiotherapy treatment planning. Even when MR images are acquired for treatment planning, the standard clinical practice currently also requires a CT for dose calculation and x-ray based patient positioning. This increases workloads, introduces uncertainty due to the required inter-modality image registrations, and involves unnecessary irradiation. While it would be beneficial to use exclusively MR images, a method needs to be employed to estimate a synthetic CT (sCT) for generating electron density maps and patient positioning reference images. We investigated 2D and 3D convolutional neural networks (CNNs) to generate a male pelvic sCT using a T1-weighted MR image and compare their performance.
A retrospective study was performed using CTs and T1-weighted MR images of 20 prostate cancer patients. CTs were deformably registered to MR images to create CT-MR pairs for training networks. The proposed 2D CNN, which contained 27 convolutional layers, was modified from Han’s 2D CNN to save computational memory and prepare for building the 3D model1 . The proposed 2D and 3D models were trained from scratch to map intensities of T1-weighted MR images to CT Hounsfield Unit (HU) values. Each sCT was generated in a five-fold cross-validation framework and compared with the corresponding deformed CT (dCT) using voxel-wise mean absolute error (MAE). The sCT geometric accuracy was evaluated by comparing bone regions, defined by thresholding at 150 HU in the dCTs and the sCTs, using dice similarity coefficient (DSC), recall, and precision. To evaluate sCT patient positioning accuracy, bone regions in dCTs and sCTs were rigidly registered to the corresponding cone-beam CTs. The resulting paired Euler transformation vectors were compared by calculating translation vector distances and absolute differences of Euler angles. Statistical tests were performed to evaluate the differences among the proposed models and Han’s model.
Generating a pelvic sCT required approximately 5.5 s using the proposed models. The average MAEs within the body contour were 40.5±5.4 HU (mean ± SD) and 37.6±5.1 HU for the 2D and 3D CNNs, respectively. The average DSC, recall, and precision for the bone region (thresholding the CT at 150 HU) were 0.81±0.04, 0.85±0.04, and 0.77±0.09 for the 2D CNN, and 0.82±0.04, 0.84±0.04, and 0.80±0.08 for the 3D CNN, respectively. For both models, mean translation vector distances are less than 0.6 mm with mean absolute differences of Euler angles less than 0.5°.
The 2D and 3D CNNs generated accurate pelvic sCTs for the 20 patients using T1-weighted MR images. Statistical tests indicated that the proposed 3D model was able to generate sCTs with smaller MAE and higher bone region precision compared to Han’s and the proposed 2D models. Results of patient alignment tests suggested sCTs generated by the proposed CNNs can provide accurate patient positioning. The accuracy of the dose calculation using generated sCTs will be tested and compared for the proposed models in the future. This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.