Investigation to Improve the Display of Language Areas using Functional Magnetic Resonance Imaging

Abstract

Activated brain regions, crucial for understanding cognitive processes, are effectively visualized and localized through functional magnetic resonance imaging (fMRI). This technique relies on the hemodynamic response function (HRF) and the Blood-Oxygen-Level-Dependent (BOLD) effect, reflecting changes in blood flow in response to neural activation. In the context of language studies, task-based fMRI examinations employing language paradigms serve as a potent method for stimulating language areas. This investigation conducted a prospective subject study, systematically comparing and evaluating various language tasks and experimental settings. This approach sought to provide insights and recommendations for optimizing language fMRI examinations. Furthermore, a novel approach was taken to enhance the accuracy of activation maps. Unlike conventional preprocessing steps, which can introduce discrepancies, simple t-maps were generated using unprocessed fMRI data. This methodology aimed to provide a more realistic representation of language areas by preserving the true measured location of activations. To further refine the analysis of these activation maps, a single-voxel fMRI signal analysis was conducted, accounting for HRF dependencies. This analysis successfully led to the development of a language area-specific filter.