Publication

Abstract : This study concerns the use of complex imaging methods like integrating ResNet, Generative Adversarial Networks (GAN), and U- Net with a single set of variables. Improving medical image analysis for lung-related diseases is the primary objective. This study performs in- depth analysis of lung CT data, aiming at improving UNet parameters, GANs and ResNet for better segmentation experience effects on 3D volume rendering images classification. The major effort here is tuning the parameters of the U-Net architecture, training GANs for generative tasks and applying ResNet to improve feature extraction and classification. To provide more detailed understanding of structures in the lung, advanced 3D visualization techniques are used to give greater insights into abnormalities. With the use of U-Net, GAN and ResNet together can classify lung CT scans. This provides a way to distinguish between general patterns and specific problems. The main goal is to fuse these solid building blocks in order to process medical pictures for better results, that will increase the effectiveness and accuracy of diagnosing diseases related with lungs.