Neuroimaging data analysis

• Main Author: Chew, Vivien Jing Wen
• Other Authors: Ke Yiping, Kelly
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Science::Medicine::Computer applications; Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/166020
• Institution: Nanyang Technological University

Neuroimaging is a discipline that utilises imaging technology to study the neural system, including its composition and function. The data obtained from neuroimaging methods such as functional magnetic resonance imaging (fMRI) and diffusion magnetic resonance imaging (dMRI) is extensive and complex, as it encompasses multiple dimensions related to structure, function, and pathology. The analysis of neuroimaging data is a complex and multi-step process that involves several stages such as pre-processing, segmentation, registration, and statistical analysis. Pre-processing is a crucial step before further analysis as the quality of the data is largely influenced by the image acquisition parameters employed. Various neuroimaging data can have different intensity values, matrix sizes, and orientations depending on the acquisition parameters used. One specific neuroimaging technique, diffusion tensor imaging (DTI) provides quantitative data on the white matter structure of the brain, allowing for the evaluation of microstructural changes. DTI has a low signal-to-noise ratio and its acquisition requires a relatively long scan time, which will affect the quality of the scans by factors such as head motion, physiological contributions, and tissue outside the scope of interest. To ensure accurate observations, it is imperative to acquire precise images in medical image processing. Without proper pre-processing, the data may contain artifacts that could result in considerable error and bias in the subsequent analysis. As a result, pre-processing DTI data is essential to guarantee that the final results of research studies are accurate, reliable, and interpretable. There are several pre-processing steps for different purposes, all with the goal of making sure that the neuroimaging data is of high quality. These steps include denoising, fiber tractography, image registration, and many others. The focus of this paper is to examine the various stages involved in the pre-processing pipeline of DTI data and to explore the potential benefits of utilising a comprehensive application called DSI Studio, which offers a unified platform for researchers and clinicians to pre-process their data.