Dense121GAN: transfer learning-enhanced conditional generative adversarial network with DenseNet121 for reliable and efficient segmentation in medical and industrial imaging
Dense121GAN, a conditional generative adversarial network (cGAN) that integrates a pre-trained DenseNet121 as a frozen encoder that integrates a pre-trained DenseNet121 as a frozen encoder, provides a practical and versatile solution for both clinical diagnostic and industrial inspection applications.
Abstract
Accurate image segmentation in medical and industrial domains remains challenging due to small object sizes, complex textures, and diverse defect morphologies. To address these limitations, we propose Dense121GAN, a conditional generative adversarial network (cGAN) that integrates a pre-trained DenseNet121 as a frozen encoder. The proposed architecture promotes effective feature reuse and stable information flow through residual skip connections. By leveraging transfer learning, Dense121GAN captures rich hierarchical representations and improves training stability, particularly in data-constrained settings. Extensive experiments were conducted using five-fold cross-validation across 16 heterogeneous datasets. Dense121GAN consistently outperformed UNet and ResNet-based generative adversarial networks (GANs) across multiple evaluation metrics, including Intersection over Union (IoU), Dice coefficient (Dice), Matthews Correlation Coefficient (MCC), Cohen’s Kappa, Structural Similarity Index Measure (SSIM), coefficient of determination (
R
2
), and Symmetric Mean Absolute Percentage Error (SMAPE). The model also demonstrated stable training behavior and mitigated common GAN failure modes such as mode collapse. In addition, Dense121GAN achieved faster convergence compared to the evaluated baseline architectures. These results highlight the effectiveness, robustness, and computational efficiency of Dense121GAN. The proposed framework provides a practical and versatile solution for both clinical diagnostic and industrial inspection applications.
A hybrid GAN-guided diffusion framework that uses a pretrained Wasserstein GAN with gradient penalty (WGAN-GP) as a feature prior for conditional diffusion-based image restoration that consistently improves the quality of both degraded and low-resolution images.
Saif Ahmed, Ashadullah Galib, S. R. R. Antu et al.· 0 citations
Large-scale, publicly available datasets have driven advances in deep learning, but privacy and legal restrictions often limit data sharing in medical imaging. Synthetic data generation offers a privacy-friendly alternative to enable the training of high-performance models on health data. While most state-of-the-art generative models produce high-quality images, they remain computationally expensive, which limits their applicability on resource-constrained hardware. We propose StyleGANCA, the first lightweight general-purpose NCA-based generative adversarial network. The architecture integrates a StyleGAN-inspired mapping network and adaptive style modulation into a multi-scale NCA synthesis process, enabling latent-controlled image generation through iterative local interactions. We evaluate StyleGANCA on BloodMNIST and PathMNIST against adversarial, variational, diffusion, and NCA-based baselines. Experimental results demonstrate that StyleGANCA achieves competitive image quality with substantially fewer parameters than baseline architectures, achieving the best FID and KID scores on PathMNIST with only 617k parameters. Furthermore, downstream experiments show that the generated images preserve class-specific information and effectively support the training of multi-class classifiers. Our code is publicly available at: https://github.com/MECLabTUDA/StyleGANCA
Anh Thi Luu, Nick Lemke, Anirban Mukhopadhyay· 0 citations
This study builds and test a Deep Convolutional Generative Adversarial Network (DCGAN) that can produce realistic portraits of people's faces and proves that DCGANs are capable of creating realistic facial representations.
K. N. Reddy, A. Renuka· International Journal for Re...· 0 citations
Generative Adversarial Networks (GANs) have become a prominent approach for image generation; however, they often suffer from training instability, mode collapse, and limited controllability of generated outputs. This study investigates the role of mutual information in improving generative modeling through a comparative analysis of Vanilla GAN, Conditional GAN (CGAN), and InfoGAN. Experiments were conducted using two image datasets with different levels of complexity, namely MNIST and Anime Face, under comparable training configurations. The evaluation focused on training behavior, convergence characteristics, generated image quality, and latent representation learning. The results revealed notable differences among the evaluated models. Vanilla GAN exhibited unstable convergence behavior at higher training epochs, while CGAN provided conditional control over generated outputs but did not fully mitigate training instability. In contrast, InfoGAN maintained more balanced generator and discriminator loss dynamics and produced visually consistent outputs across both datasets. Furthermore, latent code manipulation experiments showed that InfoGAN learned more structured and interpretable latent representations, enabling controllable feature variation in generated images. These findings indicate that incorporating mutual information improves representation learning, controllability, and training stability in GAN-based image generation. This study provides an empirical comparison of GAN, CGAN, and InfoGAN under a unified experimental framework and demonstrates that mutual information regularization contributes to improved training stability, controllable generation, and more interpretable latent representations. The findings highlight the potential of information-theoretic regularization for enhancing generative modeling performance.
Empirical support is provided for the utility of structure-aware perturbation refinement in improving black-box adversarial transferability across heterogeneous visual architectures.
Qi-Rui Lu, Liansong Zong, Fu-Ran Liu et al.· Neural Networks· 0 citations
Medical image segmentation requires high accuracy and robustness, yet practical commercial deployment also demands privacy preservation and computational efficiency. In this context, the U-Net architecture, which can be inherently decoupled into independent encoder and decoder components, serves as a natural commercial choice. However, pure Transformer-based variants like Swin-UNet often suffer from insufficient local detail capture and limited interpretability. In this paper, we propose a lightweight hybrid architecture built upon the Swin-UNet framework. Our model integrates a parallel CNN encoder to complement the shallow layer reasoning of Swin Transformers with local texture features. To bridge the semantic gap and enhance fine-grained spatial detail recovery, we design an asymmetric feature fusion strategy and introduce cross-layer skip (XSkip) connections that explicitly propagate shallow CNN features into the decoder. We further incorporate novel loss functions and an auxiliary supervision head (Aux-Head) to strengthen training stability, boundary delineation, and intermediate feature interpretability. Extensive experiments on the Synapse multi-organ segmentation dataset demonstrate that our approach achieves state-of-the-art competitive Dice scores and Hausdorff distances, offering an accurate, efficient, and interpretable solution for clinical deployment.
Bin Dong, Jing-Hong Chen· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
