"Predicting Brain Tumor from MRI-based Medical Image Synthesis using DCGAN and CNN" is a deep learning technique. The paper focuses on generating synthetic medical images through DCGAN (Deep Convolutional Generative Adversarial Network) and subsequently employs a CNN (Convolutional Neural Network) for accurate brain tumor prediction.
** The brain tumor detection in this case was done using cnn and the image synthesis was done using feed forward Gan as dcgan was not running on my machine as it takes to much ram you can try reducing the input and out image size also reducing the batch size and number of epoches. also yo can try using sas viya and intel dev cloud to run dcgan on cloud.
** create additional files as names given in the code to store output images
** both the files(feed forward gan and dcgan) are attached and also the data set used has been provided.
** Also we have tried using feed forward gan with and without discriminator(that is simple feed forward network) surprisingly the one without discriminator was generating better results.