PSEUDO CT GENERATION FOR THE DETECTION OF BONE LESIONS BASED ON HYBRID DCNN GAN

Nowadays, medical imaging has an important role in radiotherapy and treatment planning process. Despite the increasing usage of the magnetic resonance imaging (MRI) in the external radiotherapy (RT) design process. MRI is only radiotherapy treatment planning is attractive since MRI provides superior soft tissue contrast without ionizing radiation compared with computed tomography (CT). CT-scans are the main imaging modality in external beam radiotherapy. They allow the definition of tissue electron density necessary for dose calculation. A magnetic resonance (MR)-only radiotherapy workflow can reduce cost, radiation exposure and uncertainties. However, it requires the generation of pseudo CT from MRI images for patient setup and dose calculation. A crucial prerequisite is generating the so called pseudo-CT (pCT) images for accurate dose calculation and planning. Our proposed GAN (Generative Adversarial Network) method to generate pseudo CT images has been shown to provide pseudo CT images with excellent image quality. In this study, the main aim is to allow pseudo-CT images to be generated from MRI data for the diagnosis of bone lesions and investigate the accuracy of dose calculation in brain frameless Stereotactic Radio Surgery (SRS) using pseudo CT images which are generated from MRI images using the Deep Convolutional Neural network method. The outcome of a detailed assessment of various strategies for GAN, brain bone lesion identification from magnetic resonance imaging (MRI) was exploited to select the optimal parameters and setting, with the aim of proposing a DCNN based GAN pseudo-CT generation approach.