Pixelwise Gradient Model with GAN for Virtual Contrast Enhancement in MRI Imaging

Background: The development of advanced computational models for medical imaging is crucial for improving diagnostic accuracy in healthcare. This paper introduces a novel approach for virtual contrast enhancement (VCE) in magnetic resonance imaging (MRI), particularly focusing on nasopharyngeal cancer (NPC). Methods: The proposed model, Pixelwise Gradient Model with GAN for Virtual Contrast Enhancement (PGMGVCE), makes use of pixelwise gradient methods with Generative Adversarial Networks (GANs) to enhance T1-weighted (T1-w) and T2-weighted (T2-w) MRI images. This approach combines the benefits of both modalities to simulate the effects of gadolinium-based contrast agents, thereby reducing associated risks. Various modifications of PGMGVCE, including changing hyperparameters, using normalization methods (z-score, Sigmoid and Tanh) and training the model with T1-w or T2-w images only, were tested to optimize the model’s performance. Results: PGMGVCE demonstrated a similar accuracy to the existing model in terms of mean absolute error (MAE) (8.56 (Formula presented.) 0.45 for Li’s model; 8.72 (Formula presented.) 0.48 for PGMGVCE), mean square error (MSE) (12.43 (Formula presented.) 0.67 for Li’s model; 12.81 (Formula presented.) 0.73 for PGMGVCE) and structural similarity index (SSIM) (0.71 (Formula presented.) 0.08 for Li’s model; 0.73 (Formula presented.) 0.12 for PGMGVCE). However, it showed improvements in texture representation, as indicated by total mean square variation per mean intensity (TMSVPMI) (0.124 (Formula presented.) 0.022 for ground truth; 0.079 (Formula presented.) 0.024 for Li’s model; 0.120 (Formula presented.) 0.027 for PGMGVCE), total absolute variation per mean intensity (TAVPMI) (0.159 (Formula presented.) 0.031 for ground truth; 0.100 (Formula presented.) 0.032 for Li’s model; 0.153 (Formula presented.) 0.029 for PGMGVCE), Tenengrad function per mean intensity (TFPMI) (1.222 (Formula presented.) 0.241 for ground truth; 0.981 (Formula presented.) 0.213 for Li’s model; 1.194 (Formula presented.) 0.223 for PGMGVCE) and variance function per mean intensity (VFPMI) (0.0811 (Formula presented.) 0.005 for ground truth; 0.0667 (Formula presented.) 0.006 for Li’s model; 0.0761 (Formula presented.) 0.006 for PGMGVCE). Conclusions: PGMGVCE presents an innovative and safe approach to VCE in MRI, demonstrating the power of deep learning in enhancing medical imaging. This model paves the way for more accurate and risk-free diagnostic tools in medical imaging.