Speaker
Description
Mental illness is becoming more and more commonplace in today's environment, affecting a growing number of people. Functional Magnetic Resonance Image (fMRI) is the most effective technique for preventing and detecting mental disease [2]. Specialists can quickly diagnose patients by using fMRI analysis to detect changes in the brain [3]. fMRI data processing helps map neural activity in specific brain regions, enabling accurate localization of cognitive functions and improving our understanding of functional connectivity and brain region interactions [1][5]. This paper focuses on the use of machine learning techniques for fMRI analysis with the aim of identifying different brain features that may be associated in young people with the later development of Bipolar Disorder (BD). Our study uses publicly available data collected from a total of 119 symptomatic youth aged 12 to 17 years with a first-degree relative with BD and naive antidepressant treatment from the University of Cincinnati and Stanford University [4]. The first step in the proposed approach is the preprocessing of fMRI data using two distinct methods, the first one combining Statistic Parametric Mapping and CONN-Box for a better outcome and the second one focusing on Melodic software, comparing them to determine which yields the cleanest and most reliable results. This is essential for ensuring accurate analyses and stronger study outcomes. Given that BD is associated with abnormal brain interconnectivity, machine learning techniques such as Support Vector Machines and Random Forests are well-suited for analyzing this complex data. These approaches are implemented and analyzed in the current work to facilitate a detailed comparison and identify the optimal model or combination of models, aiming to enhance the accuracy and depth of our analysis of BD-related neural patterns.
References
[1] - Jenkinson, M., & Chappell, M. (2018). Introduction to neuroimaging analysis. Oxford University Press.
[2] - Kwong, K. K. (2012). Record of a single fMRI experiment in May of 1991. Neuroimage, 62(2), 610-612.
[3] - Ogawa, S. (2012). Finding the BOLD effect in brain images. Neuroimage, 62(2), 608-609.
[4] - Pan, N., Qin, K., Patino, L. R., Tallman, M. J., Lei, D., Lu, L., ... & DelBello, M. P. (2024). Aberrant brain network topology in youth with a familial risk for bipolar disorder: a task‐based fMRI connectome study. Journal of Child Psychology and Psychiatry.
[5] - Sörös, P., & Witt, K. (2018). Book review: Introduction to neuroimaging analysis.
