Diagnosis of infirm tissues and determination of cognitive power of brain from fMRI images

Abstract

newline One of the most important medical applications is brain MRI segmentation. The human brain is the main organ of the central nervous system and controls the overall voluntary and involuntary activities of the entire body. The major functions of the human brain are, controlling the higher and more complex activities like retaining memory, sensory perception, attention to detail, and a lot more. Which is why, it is of prime importance to safeguard human brain. When a tumor occurs in the brain, it influences the activities of the brain and stops the entire body from functioning, or results in sudden death. Functional Magnetic Resonance Imaging (fMRI) is used to diagnose the functions of the brain. The fMRI is a medical imaging application used to detect and identify the blood flow and oxygenation variations in the brain. The blood flow and oxygenation respond to neural activity which helps map the brain for investigating its functions. Several earlier methods have focused on diagnosing the MRI image segmentation to identify and detect the MRI tumor and its severity level. Few research works have tried to analyze fMRI images for cognitive calculation, but the accuracy is not satisfactory. Machine learning algorithms are semiautomatic in feature extraction and classification, and recent applications have moved to deep learning algorithms. Deep Learning is an essential advanced machine learning algorithm, which is used to analyze massive and complex data speedily and deeply. Deep learning algorithms have a proven record of increasing the efficiency of various kinds of data analytics. Healthcare, genes, natural disasters, e-commerce and customer review data are examples of complex and ever-growing big data where deep learning algorithms are used iv to analyze them. Deep learning algorithms learn the data thoroughly and provide higher accuracy in classification and prediction than machine learning algorithms.