Deep-learning-based generation of synthetic 6-minute MRI from 2-minute MRI for use in head and neck cancer radiotherapy.

Background: Quick magnetic resonance imaging (MRI) scans with low contrast-to-noise ratio are typically acquired for daily MRI-guided radiotherapy setup. However, for patients with head and neck (HN) cancer, these images are often insufficient for discriminating target volumes and organs at risk (OARs). In this study, we investigated a deep learning (DL) approach to generate high-quality synthetic images from low-quality images. Methods: We used 108 unique HN image sets of paired 2-minute T2-weighted scans (2mMRI) and 6-minute T2-weighted scans (6mMRI). 90 image sets (~20,000 slices) were used to train a 2-dimensional generative adversarial DL model that utilized 2mMRI as input and 6mMRI as output. Eighteen image sets were used to test model performance. Similarity metrics, including the mean squared error (MSE), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) were calculated between normalized synthetic 6mMRI and ground-truth 6mMRI for all test cases. In addition, a previously trained OAR DL auto-segmentation model was used to segment the right parotid gland, left parotid gland, and mandible on all test case images. Dice similarity coefficients (DSC) were calculated between 2mMRI and either ground-truth 6mMRI or synthetic 6mMRI for each OAR; two one-sided t-tests were applied between the ground-truth and synthetic 6mMRI to determine equivalence. Finally, a visual Turing test using paired ground-truth and synthetic 6mMRI was performed using three clinician observers; the percentage of images that were correctly identified was compared to random chance using proportion equivalence tests. Results: The median similarity metrics across the whole images were 0.19, 0.93, and 33.14 for MSE, SSIM, and PSNR, respectively. The median of DSCs comparing ground-truth vs. synthetic 6mMRI auto-segmented OARs were 0.86 vs. 0.85, 0.84 vs. 0.84, and 0.82 vs. 0.85 for the right parotid gland, left parotid gland, and mandible, respectively (equivalence p<0.05 for all OARs). The percent of images correctly identified was equivalent to chance (p<0.05 for all observers). Conclusions: Using 2mMRI inputs, we demonstrate that DL-generated synthetic 6mMRI outputs have high similarity to ground-truth 6mMRI, but further improvements can be made. Our study facilitates the clinical incorporation of synthetic MRI in MRI-guided radiotherapy.

Intensity standardization methods in magnetic resonance imaging of head and neck cancer.

BACKGROUND AND PURPOSE: Conventional magnetic resonance imaging (MRI) poses challenges in quantitative analysis because voxel intensity values lack physical meaning. While intensity standardization methods exist, their effects on head and neck MRI have not been investigated. We developed a workflow based on healthy tissue region of interest (ROI) analysis to determine intensity consistency within a patient cohort. Through this workflow, we systematically evaluated intensity standardization methods for MRI of head and neck cancer (HNC) patients. MATERIALS AND METHODS: Two HNC cohorts (30 patients total) were retrospectively analyzed. One cohort was imaged with heterogenous acquisition parameters (HET cohort), whereas the other was imaged with homogenous acquisition parameters (HOM cohort). The standard deviation of cohort-level normalized mean intensity (SD NMIc), a metric of intensity consistency, was calculated across ROIs to determine the effect of five intensity standardization methods on T2-weighted images. For each cohort, a Friedman test followed by a post-hoc Bonferroni-corrected Wilcoxon signed-rank test was conducted to compare SD NMIc among methods. RESULTS: Consistency (SD NMIc across ROIs) between unstandardized images was substantially more impaired in the HET cohort (0.29 ± 0.08) than in the HOM cohort (0.15 ± 0.03). Consequently, corrected p-values for intensity standardization methods with lower SD NMIc compared to unstandardized images were significant in the HET cohort (p < 0.05) but not significant in the HOM cohort (p > 0.05). In both cohorts, differences between methods were often minimal and nonsignificant. CONCLUSIONS: Our findings stress the importance of intensity standardization, either through the utilization of uniform acquisition parameters or specific intensity standardization methods, and the need for testing intensity consistency before performing quantitative analysis of HNC MRI.