Peripheral Artery Disease (PAD) is a prevalent and complex medical condition affecting the vascular system, most notably the lower extremities, including the iliac, femoral, and femoropopliteal arteries [1]. Advancements in imaging technology have positioned computer-tomography angiography (CTA) as a highly reliable alternative for digital subtraction angiography, particularly for assessing arterial disease in the femoral arteries [2]. Current challenges in diagnosing atherosclerosis often stem from the limited resolution of femoral artery CT scans. This research aims to improve the quality and resolution of critical medical images by employing a specialized super-resolution generative adversarial network (SRGAN) model. The dataset was comprised of 1683 CT images from 8 patients diagnosed with atherosclerosis. The GAN architecture consists of two main components: the Generator and the Discriminator accompanied by a feature extractor to facilitate content loss calculation during training. The generator is based on a residual network (ResNet) architecture, specifically tailored for super-resolution tasks. The GAN is trained through an alternating optimization process where the generator aims to minimize both the adversarial loss and the content loss, and the discriminator focuses on correctly classifying real and generated high-resolution images. A range of evaluation metrics was employed to quantify the performance of this super-resolution model, such as mean square error (MSE), root mean square error (RMSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), universal quality index (UQI), enhanced tri-axial gray-scale anisotropy score (ETGAS), spectral angle mapper (SAM), multi-scale structural similarity index (MSSSIM), and visual information fidelity in pixel-based images (VIF). PSNR and SSIM scores, of 13.16 and 0.2957 respectively, validated the achievement of superior image quality. The UQI and ETGAS metrics of 0.6023 and 19.38 respectively support the preservation of texture and global structural information in the up-sampled images. In conclusion, this research demonstrates significant real-world applicability, particularly in the realm of cardiovascular imaging. By using this SRGAN model for super-resolution, substantial improvements in image quality, which hold considerable promise for more accurate diagnosis and nuanced treatment planning were achieved.