fNIRS Optodes' Location Decider (fOLD): a toolbox for probe arrangement guided by brain regions-of-interest.
Zimeo Morais GA, Balardin JB, Sato JR. · 2018
Read fNIRS Optodes' Location Decider (fOLD): a toolbox for probe arrangement guided by brain regions-of-interest. by Zimeo Morais GA, Balardin JB, Sato JR. (2018) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5820343.txt.
Introduction
Zimeo Morais and co-authors present fOLD, a toolbox for placing fNIRS sensors. fNIRS shines near-infrared light through the scalp to track brain work. It is portable, cheap, and tolerant of movement. Each study can place only a few light probes, though.
Probe spots must match the brain areas under study. Guessing that match is a common design headache. The toolbox runs photon travel simulations on two head atlases. It then picks the predefined spots with the best aim. The free tool borrows mapping advances from MRI research. Readers get a practical way to design sharper fNIRS experiments.
Catalog record
| CREATOR | Zimeo Morais GA, Balardin JB, Sato JR. |
|---|---|
| YEAR | 2018 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
fNIRS Optodes’ Location Decider (fOLD): a toolbox for probe arrangement guided by brain regions-of-interest By Guilherme Augusto Zimeo Morais, Joana Bisol Balardin, João Ricardo Sato Published in: Scientific Reports DOI: 10.1038/s41598-018-21716-z ABSTRACT Abstract The employment of functional near-infrared spectroscopy (fNIRS) as a method of brain imaging has increased over the last few years due to its portability, low-cost and robustness to subject movement. Experiments with fNIRS are designed in the face of a limited number of sources and detectors (optodes) to be positioned on selected portion(s) of the scalp. The optodes locations represent an expectation of assessing cortical regions relevant to the experiment’s hypothesis. However, this translation process remains a challenge for fNIRS experimental design. In the present study, we propose an approach that automatically decides the location of fNIRS optodes from a set of predefined positions with the aim of maximizing the anatomical specificity to brain regions-of-interest. The implemented method is based on photon transport simulations on two head atlases. The results are compiled into the publicly available “fNIRS Optodes’ Location Decider” (fOLD). This toolbox is a first-order approach to bring the achieved advancements of parcellation methods and meta-analyses from functional magnetic resonance imaging to more precisely guide the selection of optode positions for fNIRS experiments. INTRODUCTION Functional near-infrared spectroscopy (fNIRS) is a technique capable of measuring concentration changes of oxygenated and deoxygenated hemoglobin from the variations of absorbed near-infrared light during its transportation in tissues1. The employment of fNIRS to assess brain activity has increased over the last few years due to its advantages over functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) mainly concerning its robustness to artifacts due to motion2, thus enabling a greater gamut of naturalistic experiments3–8. While fMRI is capable of measuring whole-brain functional activity along with the structural image of the individual, fNIRS experiments are designed with a limited number of sources and detectors (optodes). The optodes are positioned on selected portion(s) of the scalp with an expectation to assess the activity of a set of brain cortical regions that are relevant to the designed experiment. However, the translation of regions of interest to the placement of optodes on a measuring cap (shown in Fig. 1) remains a challenge for experimental design with fNIRS. [Figure: Challenge commonly faced when designing an fNIRS experiment to assess a set of regions of interest expected to be activated according to a study hypothesis: the translation to an fNIRS optode layout by choosing appropriate sources and detectors positions to maximize anatomical specificity to regions of interest. Illustrated are Brodmann areas 4, 9 and 19 and 21 and the fNIRS cap layout with corresponding color-coded channels.] A few recent studies have suggested approaches to overcome this challenge, e.g. for epileptic discharges9, or based on iterative probe geometry modifications10, which was extended to the voxel-space for an image-based approach11. Herein, we propose an alternative approach to automatically decide optodes positions based on 10–10 and 10–5 systems12 according to a set of brain regions of interest. This method is based on the sensitivity profile from photon transport simulations run on two head atlases (Methods section). The results were compiled into a toolbox to facilitate the definition of optodes positions, the fNIRS Optodes’ Location Decider (fOLD). METHODS Tissue segmentation As different tissues of the human head present different optical properties (e.g. absorption and scattering)13,14, it is necessary to segment the atlases to be used for photon transport simulations. The segmentation results in discrimination of five tissues: scalp, skull, cerebrospinal fluid (CSF), and gray and white matter. The segmentation algorithm implemented in SPM1215 with default parameters has been used to segment the T1 image of the Colin27 head atlas16 provided within the MRIcron software17 as “ch2.nii.gz”. Briefly, the segmentation procedure in SPM12 returns probability maps for each of the five tissues of our interest. For each tissue, a NIfTI file is generated with an image with same size and origin as in the input file (e.g. Colin27 atlas) and each voxel is set to the probability of being part of a given tissue. And the sum of the probabilities of a given voxel along head tissues plus air is 1. For example, voxel (x = 57, y = 126, z = 144) of Colin27 atlas corresponding to MNI coordinate (x = −34, y = 0, z = 72) presented probability 86.27% for skull and 13.73% for CSF. To create a single image file for the final tissue segmentation, we have defined each voxel to be part of a given tissue if its probability was higher than all other tissues and if it was greater than 0.2. The latter was particularly important for boundary voxels e.g. between scalp and air. Voxels corresponding to scalp were assigned the value 1, skull 2, CSF 3, gray matter 4 and white matter 5. All voxels that did not fulfill the comparisons for any of tissue was assigned value 0 (air). As this procedure could potentially result in unitary holes inside the tissue (i.e. tissue voxels assigned to air), the resulting image was smoothed in SPM12 with full width at half maximum (FWHM) of [2 mm, 2 mm, 2 mm]. The values in the smoothed image were rounded in Matlab2017a. The resulting segmentation of Colin27 is depicted in Fig. 2A. [Figure: Axial view of the tissue segmentation of (A) Colin27 template and (B) SPM12 tissue probability maps (TPM.nii). This resulted on five layers: scalp (blue), skull (cyan), cerebrospinal fluid (CSF, yellow), gray matter (red) and white matter (black).] In addition to the Colin27 head model, which is based on 27 averages of a single subject, we have considered a second head atlas based on the tissue probability maps provided in SPM12. These maps have been generated based on 549 subjects from the IXI dataset18. The specific subjects used to compute TPM.nii can be found in spm_templates.man19. We spatially resampled20 the TPM.nii image from 1.5 × 1.5 × 1.5 mm3 to 2 × 2 × 2 mm3 and followed the same procedure as for Colin27 to obtain the segmented image (herein called “SPM12”), illustrated in Fig. 2B. Optodes coordinates Once the tissue segmentation procedure was performed, the scalp boundaries were defined for both brain atlases of interest, and thus we proceeded with the spatial localization of optodes to be used for the photon migration simulations. We initially considered the 10–10 international system as primary reference for optodes placement as its positions can be identified from a set of fiducial points (e.g. nasion, inion and left/right preauricular points)12, which may be visually determined in head models. First, we converted the image file of the tissue segmentation into mesh by using the function ‘v2s’ provided in the iso2mesh toolbox21,22. We also corrected the coordinates to MNI space by accounting for the voxel size and origin of the image. After plotting the corrected mesh in Matlab2017a, we visually identified four fiducial points (depicted in Fig. 3A) whose spatial identification followed the definition provided by Jurcak et al.12 for nasion, inion and preauricular points. [Figure: (A) MNI space localization (in mm) of fiducial points on mesh of Colin27. (B) Left view of sources (red) and detectors (green) positions on the 10–10 international system that have been initially considered for photon transport simulation