Computed Tomography Angiography (CTA) is an important imaging modality for the diagnosis of vascular diseases. However, it requires intravenous injection of iodine contrast agents, which carries risks such as renal injury and allergic reactions, and increases medical costs. To address these issues, this paper proposes a CT-to-CTA image translation method based on Vessel-Enhanced GAN (VE-GAN), which can generate high-quality CTA images from routine CT images without contrast agents.Based on conditional Generative Adversarial Networks (cGAN), the method designs a UNet generator integrating multi-scale vessel enhancement blocks, edge preservation blocks, and frequency-domain attention mechanisms. A spectral-normalized PatchGAN discriminator is adopted to ensure training stability, and a multi-objective loss function including vessel-specific loss and edge loss is constructed to optimize generation quality. Experiments are conducted using 1491 pairs of CT-CTA images as the training set, 100 pairs as the validation set, and 216 CT images as the test set. Results show that the generated images achieve a Peak Signal-to-Noise Ratio (PSNR) of 31.35 dB and a Structural Similarity Index Measure (SSIM) of 0.881, verifying the feasibility of the method. Ablation studies demonstrate that each innovative module effectively improves translation performance, with the vessel enhancement module and perceptual loss contributing the most significantly.Although the current performance does not yet meet clinical application standards, this study provides a new technical path for non-invasive vascular visualization and lays a solid foundation for further optimization.
Nannan Li, Miao Song· International Conference on...· 0 citations
The high prevalence of ophthalmic diseases has become a major threat to visual health. Traditional diagnosis relies heavily on manual assessment, which is time-consuming and prone to subjective variability, highlighting the urgent need for efficient and reliable automated methods. This study proposes a novel deep learning framework, FReCSA-ONFELNet, designed for the automated classification of diabetic retinopathy, glaucoma, cataract, and normal retina. The framework integrates three innovative modules: the Frequency-Regulated Channel–Spatial Attention (FReCSA), the Optic Nerve Feature Enhancement Layer (ONFEL), and the Multi-Scale Feature Fusion (MSFF). Specifically, FReCSA leverages frequency-domain information to enhance channel attention discriminability and accurately localize lesion regions; ONFEL strengthens the representation of key ophthalmic structures through a multi-scale convolutional design; MSFF captures both local and global lesion features across multiple scales. Additionally, a structural consistency loss function is introduced to effectively reduce false positives in pathological categories. Experiments conducted on a dataset of 4,180 fundus images demonstrate that the proposed model achieves an overall classification accuracy of 94%, and significantly outperforms advanced architectures such as VGG16, VGG19, AlexNet, ResNet, and InceptionNet-v4 in terms of precision, recall, and F1-score. Overall, FReCSA-ONFELNet exhibits advantages in both performance and robustness, indicating strong potential for application in early screening and clinical-assisted diagnosis of retinal diseases.
ZeKun Gao, Miao Song· International Conference on...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
