To ensure a fair comparison with Mostafa’s proposed CAE model, we maintained consistency in data processing, using T1-weighted MRI slice images of the healthy control group for autoencoder ...

This project leverages SAM 2.0's advanced segmentation capabilities to automate the creation of annotated datasets. The system significantly reduces the time and effort required for manual annotation, ...

This paper focuses on the application of adversarial autoencoders for improving the robustness of image generation. Therefore, in this study, adversarial training is proposed to be incorporated into ...

Class imbalance is a significant challenge in medical image analysis, particularly in lung ultrasound (LUS), where severe patterns are often underrepresented. Traditional oversampling techniques, ...

Built and trained a Variational Autoencoder (VAE) on the CelebA dataset to generate realistic celebrity faces. Used a CNN-based encoder-decoder, KL divergence, and reconstruction loss to learn facial ...