Main Article Content
Background/Objectives: Glioma is a serious brain tumor affecting the nervous system and is often found only after symptoms such as headache, vomiting, seizures and cerebral neuropathy appear.
Methods/Statistical analysis: In medical image analysis, digital image processing is a process to diagnose a disease or prediction of survival rate of patient or so on using medical images like MRI, CT scan, PET scan, X-ray or ultrasound using machine learning algorithms. In this paper an overview over digital image processing, machine learning is discussed and we propose a comparison study between state-of-the-art supervised deep learning algorithms.
Findings: We discuss about Glioma tumor and its severity. In this paper we present medical image processing techniques are helpful in glioma tumor diagnosis and overall survival time prediction for glioma patients. Furthermore, we present an overall introduction about digital image processing, history, tasks and architecture; machine learning, deep learning and its classification. Later on we discuss about medical images used for brain tumor detection, we describe about MRI, CT scans, ultrasound, X-ray and PET scans. We investigate about open data sets which is also known as public data sets. We present the cancer imaging archive (TCIA), the whole brain atlas (TWBT) and about brain tumor segmentation challenges (BraTS) which are public brain tumor datasets available free on internet for researchers.
Improvements/Applications: we explore about traditional Korean medicine and traditional Indian medicine ayurveda which are alternative therapies and also known as pseudoscience.