Virtual Contrast-Enhanced Magnetic Resonance Images Synthesis for Patients With Nasopharyngeal Carcinoma Using Multimodality-Guided Synergistic Neural Network

Li, Wen; Xiao, Haonan; Li, Tian; Ren, Ge; Lam, Saikit; Teng, Xinzhi; Liu, Chenyang; Zhang, Jiang; Lee, Francis Kar-ho; Au, Kwok‐Hung; Lee, Victor Ho-fun; Chang, Artemis; Cai, Jing

doi:10.1016/j.ijrobp.2021.11.007

Cited by 34 publications

(59 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recent technique that virtually enhances the MRI contrast demonstrates a noninvasive alternative to DCE assessment. 62 However, DWI is inferior to DCE in depicting intratumoral heterogeneity owing to the lower resolution in the ADC map than the DCE kinetic map. 63 On the other hand, MRS also has several limitations, making it less suitable for applying in the treatment response assessment in NPC.…”

Section: Discussionmentioning

confidence: 99%

“…This may raise secondary harms to NPC patients and thus limit the frequency of DCE application before, during, and after treatment. A recent technique that virtually enhances the MRI contrast demonstrates a noninvasive alternative to DCE assessment 62 . However, DWI is inferior to DCE in depicting intratumoral heterogeneity owing to the lower resolution in the ADC map than the DCE kinetic map 63 …”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Review of functional magnetic resonance imaging in the assessment of nasopharyngeal carcinoma treatment response

Wong

Cheng

Lam

et al. 2022

Precision Radiation Oncology

Self Cite

View full text Add to dashboard Cite

Nasopharyngeal carcinoma (NPC) is a common malignancy endemic in South‐East Asia. Functional magnetic resonance imaging (fMRI) has been used for prompt detection of treatment response before visible morphological changes in NPC treatment. Among different fMRI techniques, diffusion‐weighted imaging (DWI) and dynamic contrast enhancement (DCE) were proved to be more successful in NPC treatment response assessment whilst the application of magnetic resonance spectroscopy (MRS) remains questionable. Apart from discussing the imaging technique, time points for post‐treatment response assessments are recommended. Four instead of three months is recommended for prompt identification of non‐ or partial responders and 6–9 months post‐treatment multiparametric MRI is also recommended for effective confirmation of complete responding individuals to avoid residual disease. For future advancement, in addition to the post‐treatment response assessment, continuous or longitudinal assessment on the treatment response with the use of magnetic resonance simulator (MR‐simulator) or magnetic resonace imaging guided linear accelerator (MR‐Linac) tailor‐made for radiotherapy (RT) maybe feasible. Longitudinal assessments or predictive radiomics modeling allow spotting out the possibility of treatment failure before the completion or even before the start of treatments. Adaptive instead of salvage treatments can then be prepared, thus reducing the damage and side effects of consecutive cytotoxic treatments.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Review of functional magnetic resonance imaging in the assessment of nasopharyngeal carcinoma treatment response

Wong

Cheng

Lam

et al. 2022

Precision Radiation Oncology

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, SSIM provides an intensity scale-invariant method of calculating similarity, which allows for a surface-level comparison with other investigations. To our knowledge, the only investigation of DL-generated synthetic MRI from input MRI sequences in HNC was performed by Li et al (31).…”

Section: Discussionmentioning

confidence: 99%

“…However, since model training occurred on a slice-by-slice basis, we utilized on the order of ~20,000 training data points, which allowed us to leverage DL approaches effectively. Moreover, we only tested one DL approach; several architectural modifications have been proposed that could improve our models in terms of similarity metric performance (31). However, since the end goal of our study is clinician acceptability in an RT workflow and our model passed the Turing test, additional improvement of model performance may be moot.…”

Section: Discussionmentioning

confidence: 99%

“…One less routinely studied domain is synthetic image generation, i.e., mapping an input image to an output image. Recent work has highlighted the utility of DL for synthetically generating CT images from MRI sequences (14)(15)(16)(17)(18)(19)(20)(21), MRI sequences from CT images (22)(23)(24)(25)(26), and MRI sequences from other MRI sequences (27)(28)(29)(30)(31). However, to date, no studies have investigated the feasibility of using DL to generate high-resolution synthetic MRI sequences from low-resolution MRI sequences to decrease the required scan time for HNC-related imaging.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep-Learning-Based Generation of Synthetic High-Resolution MRI from Low-Resolution MRI for Use in Head and Neck Cancer Adaptive Radiotherapy

Wahid

El-Habashy

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Quick, low contrast resolution magnetic resonance imaging (MRI) scans 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-resolution synthetic images from low-resolution 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 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 0.84 vs. 0.83, 0.82 vs. 0.82, and 0.80 vs. 0.83 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. Our study facilitates the clinical incorporation of synthetic MRI in MRI-guided radiotherapy.

show abstract

Contrast‐enhanced MRI synthesis using dense‐dilated residual convolutions based 3D network toward elimination of gadolinium in neuro‐oncology

Osman

Tamam

2023

J Applied Clin Med Phys

View full text Add to dashboard Cite

Recent studies have raised broad safety and health concerns about using of gadolinium contrast agents during magnetic resonance imaging (MRI) to enhance identification of active tumors. In this paper, we developed a deep learning‐based method for three‐dimensional (3D) contrast‐enhanced T1‐weighted (T1) image synthesis from contrast‐free image(s). The MR images of 1251 patients with glioma from the RSNA‐ASNR‐MICCAI BraTS Challenge 2021 dataset were used in this study. A 3D dense‐dilated residual U‐Net (DD‐Res U‐Net) was developed for contrast‐enhanced T1 image synthesis from contrast‐free image(s). The model was trained on a randomly split training set (n = 800) using a customized loss function and validated on a validation set (n = 200) to improve its generalizability. The generated images were quantitatively assessed against the ground‐truth on a test set (n = 251) using the mean absolute error (MAE), mean‐squared error (MSE), peak signal‐to‐noise ratio (PSNR), structural similarity (SSIM), normalized mutual information (NMI), and Hausdorff distance (HDD) metrics. We also performed a qualitative visual similarity assessment between the synthetic and ground‐truth images. The effectiveness of the proposed model was compared with a 3D U‐Net baseline model and existing deep learning‐based methods in the literature. Our proposed DD‐Res U‐Net model achieved promising performance for contrast‐enhanced T1 synthesis in both quantitative metrics and perceptual evaluation on the test set (n = 251). Analysis of results on the whole brain region showed a PSNR (in dB) of 29.882 ± 5.924, a SSIM of 0.901 ± 0.071, a MAE of 0.018 ± 0.013, a MSE of 0.002 ± 0.002, a HDD of 2.329 ± 9.623, and a NMI of 1.352 ± 0.091 when using only T1 as input; and a PSNR (in dB) of 30.284 ± 4.934, a SSIM of 0.915 ± 0.063, a MAE of 0.017 ± 0.013, a MSE of 0.001 ± 0.002, a HDD of 1.323 ± 3.551, and a NMI of 1.364 ± 0.089 when combining T1 with other MRI sequences. Compared to the U‐Net baseline model, our model revealed superior performance. Our model demonstrated excellent capability in generating synthetic contrast‐enhanced T1 images from contrast‐free MR image(s) of the whole brain region when using multiple contrast‐free images as input. Without incorporating tumor mask information during network training, its performance was inferior in the tumor regions compared to the whole brain which requires further improvements to replace the gadolinium administration in neuro‐oncology.

show abstract

Virtual Contrast-Enhanced Magnetic Resonance Images Synthesis for Patients With Nasopharyngeal Carcinoma Using Multimodality-Guided Synergistic Neural Network

Cited by 34 publications

References 26 publications

Review of functional magnetic resonance imaging in the assessment of nasopharyngeal carcinoma treatment response

Review of functional magnetic resonance imaging in the assessment of nasopharyngeal carcinoma treatment response

Deep-Learning-Based Generation of Synthetic High-Resolution MRI from Low-Resolution MRI for Use in Head and Neck Cancer Adaptive Radiotherapy

Contrast‐enhanced MRI synthesis using dense‐dilated residual convolutions based 3D network toward elimination of gadolinium in neuro‐oncology

Contact Info

Product

Resources

About