Program Information
A Leaning-Based Approach to Generate Patient-Specific Pseudo-CT for MRI-Based Radiation Therapy Treatment Planning
X Yang , Y Lei*, T Liu , P Rossi , H Mao , H Shim , S Tian , W Curran , H Shu , Emory Univ, Atlanta, GA
Presentations
TU-FG-605-5 (Tuesday, August 1, 2017) 1:45 PM - 3:45 PM Room: 605
Purpose: A treatment planning process with MRI as the sole imaging modality could eliminate systematic MRI-CT co-registration errors, reduce medical cost, spare the patient from CT x-ray exposure, and simplify clinical workflow. However, MRI data do not contain electron density information that is necessary for accurate dose calculation. The purpose of this work is to develop a learning-based method generate patient-specific pseudo CT from routine anatomical MRI for MRI-only radiotherapy treatment planning.
Methods: We propose to integrate auto-context model and patch-based anatomical signature into machine learning framework to iteratively predict CT images from MRI. The proposed prediction of CT images consists of two major stages: the training stage and the prediction stage. During the training stage, patch-based anatomical features are extracted from the aligned MRI-CT training images, and the most informative features are identified to train a random forest based on auto-context model. During the prediction stage, we extract the selected features from the new MRI and feed them into the well-trained forests for the CT prediction. We performed leave-one-out cross-validation method to evaluate the proposed CT prediction algorithm. Our predicted CT were compared with the original CT to quantitatively evaluate the prediction accuracy.
Results: This prediction technique was validated with a clinical study of 12 patients with brain MR and CT images. The mean absolute error (MAE) and structural similarity (SSIM) indexes were used to quantify the prediction accuracy. Overall the mean MAE and SSIM were 18.74±1.96 and 0.91±0.02 for 12 patients’ data, which demonstrated the CT prediction accuracy of the proposed learning-based method.
Conclusion: We have investigated a novel learning-based approach to predict CT images from routine MRI and demonstrated its reliability. This CT prediction technique could be a useful tool for MRI-based radiation treatment planning or attenuation correction for quantifying PET images when using a PET/MRI scanner.
Contact Email: