A A study of MRI data synthesis with Convolutional Networks, KNN regression and Generative Adversarial Network (GAN)
Sunanda Das · SPAST Abstracts · 2021
Magnetic resonance imaging (MRI) is a technology mainly used for disease prediction and treatment. Sometimes doctors advised for computed tomography (CT) for diagnosis or therapy. But, the ionizing radiations for CT are the reason for damaging DNA which become vulnerable for repetition of usage. As, MRI does not use this kind of radiations, so it is quite good and safe for clinical testing. Practically, due to poor quality of MRI images, sometimes it is advised to repeat the scan test again which causes some unavoidable situations with increase of costs. Therefore, only the improvement of the quality of MRI images can give us the relief from these unnecessary problems. Some studies [1-4] have shown that the signal to noise ratio, image resolution, contrast sensitivity and artifacts are the main key factors for enhancing image quality. So, we need an automated supervised machine learning algorithm to generate high resolution data without more efforts. The exponentially growth of uses of MRI data [5] from past 1970’s makes a big platform for the researchers for analysing medical images. Applications of machine learning algorithms in medical field have proven its tremendous successful results to diagnose medical image. The most important is that machine learning algorithms for medical image analysis is able to find the significant relationships among data in a short time and accurate results. The aim of this study is to design a comparative analysis of MRI images for the betterment of image quality for developing treatment procedure in medical field. In this paper, some computational techniques like convolutional networks, KNN regression and generative adversarial network (GAN) are applied on the MRI images to get the high-resolution based MRI images. The methodology follows medical image localization, detection, segmentation and classification. The validation results on real data of MRI data fundamentally determines its usefulness and demonstrates the effectiveness in compared to state-of-the-art super-resolution techniques.