LIONirs: flexible Matlab toolbox for fNIRS data analysis

• Published in: Journal of neuroscience methods 2022-03, Vol.370, p.109487-109487, Article 109487
• Main Authors: Tremblay, Julie, Martínez-Montes, Eduardo, Hüsser, Alejandra, Caron-Desrochers, Laura, Lepage, Charles, Pouliot, Philippe, Vannasing, Phetsamone, Gallagher, Anne
• Format: Article
• Language: English
• Subjects: Brain - diagnostic imaging; Brain - physiology; Child; Data Analysis; Electroencephalography; Functional connectivity; Functional near infrared spectroscopy (fNIRS); Humans; Matlab toolbox; Multimodal; Semi-automated artifact correction; Spectroscopy, Near-Infrared - methods; Task-based analysis
• ISSN: 0165-0270; 1872-678X
• DOI: 10.1016/j.jneumeth.2022.109487

Functional near-infrared spectroscopy (fNIRS) is a suitable tool for recording brain function in pediatric or challenging populations. As with other neuroimaging techniques, the scientific community is engaged in an evolving debate regarding the most adequate methods for performing fNIRS data analyses. We introduce LIONirs, a neuroinformatics toolbox for fNIRS data analysis, designed to follow two main goals: (1) flexibility, to explore several methods in parallel and verify results using 3D visualization; (2) simplicity, to apply a defined processing pipeline to a large dataset of subjects by using the MATLAB Batch System and available on GitHub. Within the graphical user interfaces (DisplayGUI), the user can reject noisy intervals and correct artifacts, while visualizing the topographical projection of the data onto the 3D head representation. Data decomposition methods are available for the identification of relevant signatures, such as brain responses or artifacts. Multimodal data recorded simultaneously to fNIRS, such as physiology, electroencephalography or audio-video, can be visualized using the DisplayGUI. The toolbox includes several functions that allow one to read, preprocess, and analyze fNIRS data, including task-based and functional connectivity measures. Several good neuroinformatics tools for fNIRS data analysis are currently available. None of them emphasize multimodal visualization of the data throughout the preprocessing steps and multidimensional decomposition, which are essential for understanding challenging data. Furthermore, LIONirs provides compatibility and complementarity with other existing tools by supporting common data format. LIONirs offers a flexible platform for basic and advanced fNIRS data analysis, shown through real experimental examples. [Display omitted] •The LIONirs toolbox is designed for fNIRS data inspection and visualization.•Methods are integrated for isolation of relevant activity and correction of artifacts.•Multimodal auxiliary, EEG or audio-video are visualized alongside the fNIRS data.•Task-based and functional connectivity measure analysis tools are available.•The code structure allows to automated and standardized analysis of large data set.