Computed Tomography (CT) Generation from Magnetic Resonance Imaging (MRI) using Multi Cycle GAN’s

Computed Tomography (CT) plays an important role in planning of radiation therapy (RT) but CT scans expose patients to additional ionizing radiation repeatedly, which brings potentially harmful effects to patients. Magnetic Resonance Imaging (MRI) is another method used for auxiliary diagnosis and treatment. Compared with CT, MRI is safer since the patients would not be exposed to ionizing radiation in the image scanning process. It also reduces the cost of treatment because only one set of images is required for scanning. But the drawback of MRI is that it can’t obtain the patient’s bone information. So, our proposed system is going to convert MRI images to CT using multi-cycle GAN (generative adversarial network).We have used various deep learning models like Alexnet and Resnet.