What does this research mean for the field?

Generative adversarial networks can synthesize T2 FLAIR-like MR images from non-contrast CT scans, significantly improving the conspicuity of perilesional edema in brain metastases compared to standard CT. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to determine if generative adversarial networks (GANs) can enhance T2 FLAIR-like imaging from CT scans to improve visibility of brain metastases.

June 6, 2026Open Access

CT-to-FLAIR Style Transfer Improves Perilesional Edema Conspicuity in Brain Metastases

Key Points

This research aims to determine if generative adversarial networks (GANs) can enhance T2 FLAIR-like imaging from CT scans to improve visibility of brain metastases.
Retrospective analysis of 321 patients with CT and T2 FLAIR MR images paired by lesion stability.
GANs trained for CT-to-synthetic-FLAIR style transfer with cross-validation and discriminator comparison using different UNet-family generators.
Image quality and conspicuity rated by 12 neuroradiologists after generating synthetic MR images.
UNet++ produced the best quality with SSIM of 0.9119 and significant image quality ratings (p<0.001).
Lesion conspicuity improved in synthetic MR versus CT (mean difference 1.18, p<0.001) but was slightly less than real MR (mean difference 0.33, p<0.001).
73.3% of readers preferred real MR over synthetic MR in terms of realism (p<0.001).

Abstract

BACKGROUND AND PURPOSE: To evaluate whether style transfer using generative adversarial networks (GANs) can synthesize T2 FLAIR-like MR images from non-contrast CT to enhance lesion conspicuity of brain metastases. MATERIALS AND METHODS: This retrospective study analyzed 321 patients (235 with brain metastases, 86 without) whose non-contrast CT head and T2 FLAIR MR images were paired based on lesion stability and temporal proximity. Following pre-processing, CT-MR pairs were used to train GANs for CT-to-synthetic-FLAIR style transfer with 5-fold cross-validation. PatchGAN discriminator was used. 3 UNet-family generators (Attention-UNet, SUNet, UNet++) were compared using mean-absolute-error (MAE), mean-squared-error (MSE), and structural similarity index measure (SSIM). 20 synthetic MR images generated by the best performing generator were shown to 12 neuroradiologists, who rated image quality and lesion conspicuity for CT, real MR, and synthetic MR (scale 1-5★), and realism preference between real and synthetic MR. RESULTS: UNet++ produced the best image quality, with SSIM = 0.9119 ± 0.0008, MAE = 0.1232 ± 0.0022, and MSE = 0.0201 ± 0.0004 (95% CIs, all p4.28 ± 0.15★ for real MR and 4.10 ± 0.14★ for synthetic MR; mean lesion conspicuity ratings were 4.67 ± 0.13★ for real MR, 4.34 ± 0.18★ for synthetic MR, and 3.16 ± 0.44★ for CT (all 95% CIs). Synthetic MR significantly improved lesion conspicuity over CT (pd=2.2; while synthetic MR showed slightly reduced conspicuity compared to real MR (pd=1.6 (95% CIs). Readers selected real MR as more realistic than synthetic MR in 73.3% of comparisons (pp=0.078). CONCLUSION: GAN-generated synthetic T2 FLAIR images from CT improve edema conspicuity of brain metastases relative to CT.

CT-to-FLAIR Style Transfer Improves Perilesional Edema Conspicuity in Brain Metastases

Key Points

Abstract

Cite This Study

Also Consider

Also Consider