Abstract
PurposeRadiotherapy planning based only on positron emission tomography/magnetic resonance imaging (PET/MRI) lacks computed tomography (CT) information required for dose calculations. In this study, a previously developed deep learning model for creating synthetic CT (sCT) from MRI in patients with head and neck cancer was evaluated in 2 scenarios: (1) using an independent external dataset, and (2) using a local dataset after an update of the model related to scanner software-induced changes to the input MRI. Methods and MaterialsSix patients from an external site and 17 patients from a local cohort were analyzed separately. Each patient underwent a CT and a PET/MRI with a Dixon MRI sequence over either one (external) or 2 (local) bed positions. For the external cohort, a previously developed deep learning model for deriving sCT from Dixon MRI was directly applied. For the local cohort, we adapted the model for an upgraded MRI acquisition using transfer learning and evaluated it in a leave-one-out process. The sCT mean absolute error for each patient was assessed. Radiotherapy dose plans based on sCT and CT were compared by assessing relevant absorbed dose differences in target volumes and organs at risk. ResultsThe MAEs were 78 ± 13 HU and 76 ± 12 HU for the external and local cohort, respectively. For the external cohort, absorbed dose differences in target volumes were within ± 2.3% and within ± 1% in 95% of the cases. Differences in organs at risk were <2%. Similar results were obtained for the local cohort. ConclusionsWe have demonstrated a robust performance of a deep learning model for deriving sCT from MRI when applied to an independent external dataset. We updated the model to accommodate a larger axial field of view and software-induced changes to the input MRI. In both scenarios dose calculations based on sCT were similar to those of CT suggesting a robust and reliable method.
Highlights
The use of combined positron emission tomography (PET)/magnetic resonance imaging (MRI) offers new possibilities for individualized radiotherapy planning as it provides spatially and temporally aligned structural and functional information in a single examination.[1]
We evaluated the model performance for radiotherapy when: (1) applied to an independent external dataset from another site, and (2) the model was updated to accommodate input MRI with a larger axial field of view (FOV) and changes in MRI sequence parameters induced by a scanner software upgrade
The mean error (ME) metrics show that soft tissue values are close to the reference computed tomography (CT), but for the entire body synthetic CT (sCT) values are underestimated, which is primarily driven by the underestimation of bone values
Summary
The use of combined positron emission tomography (PET)/magnetic resonance imaging (MRI) offers new possibilities for individualized radiotherapy planning as it provides spatially and temporally aligned structural and functional information in a single examination.[1]. Studies concerning head and neck cancer have demonstrated the feasibility of scanning patients with PET/MRI in the radiotherapy treatment position using dedicated equipment such as flat table overlay and immobilization masks.[9,10]. Information about the electron density of tissue is a prerequisite for dose calculation, which is provided by CT to a very good approximation, but not by MRI. Several studies have reported methods for generating synthetic CT (sCT) from MRI with promising results, especially in the brain and the pelvic region using a variety of different approaches.[11,12,13,14,15,16] The number of studies in head and neck is more limited and while initial methods have used atlasbased approaches[14,17,18] the challenging complex
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
