SPOT3D: Spatial positioning toolbox for head markers using 3D scans

Abstract

Recent studies have highlighted the importance of an accurate individual head model for reliably using high-density electroencephalography (hdEEG) as a brain imaging technique. Correct identification of sensor positions is fundamental for accurately estimating neural activity from hdEEG recordings. We previously introduced a method of automated localization and labelling of hdEEG sensors using an infrared colour-enhanced 3D scanner. Here, we describe an extension of this method, the spatial positioning toolbox for head markers using 3D scans (SPOT3D), which integrates a graphical user interface (GUI). This enables the correction of imprecisions in EEG sensor positioning and the inclusion of additional head markers. The toolbox was validated using 3D scan data collected in four participants wearing a 256-channel hdEEG cap. We quantified the misalignment between the 3D scan and the head shape, and errors in EEG sensor locations. We assessed these parameters after using the automated approach and after manually adjusting its results by means of the GUI. The GUI overcomes the main limitations of the automated method, yielding enhanced precision and reliability of head marker positioning.

Smartphone-based photogrammetry provides improved localization and registration of scalp-mounted neuroimaging sensors

“…Alternative approaches use photogrammetry 22 – 28 or 3D scanning 29 – 33 methods. Photogrammetry consists of the acquisition of multiple pictures of the subject’s head from different angles while the subject is wearing the cap.…”

“…Overall, sensors localization from a 3D point cloud has been shown to be more accurate, easy to operate, reliable and less sensitive to human error compared to the electromagnetic digitization technique. Furthermore, photogrammetry/3D scanning-based approaches considerably reduce the acquisition time and require minimal manual intervention during the analysis if combined with automated methods for sensor identifications 26 , 27 , 32 , 33 . Compared to electromagnetic and ultrasound digitizers, the time required to post-process the data is, however, longer.…”

RT-NET: real-time reconstruction of neural activity using high-density electroencephalography

“…Notably, whereas the MR image segmentation can be performed before the experiment, electrode positions need to be obtained from the participant before the leadfield matrix can be created. In this study we used a standard digitizing technique for extracting electrode positions, but it is worth noting that 3D scanning has been recently proposed to yield rapid and reliable electrode positioning (Taberna et al 2019a ; Taberna et al 2019b ). 3D scanning technology may be particularly useful for hdEEG systems, approximately halving the acquisition time.…”

Investigating the spectral features of the brain meso-scale structure at rest

“…The horizontal and vertical electrooculograms (EOG) were collected from the right eye for further identification and removal of ocularrelated artifacts. Prior to resting state hdEEG recordings, the three-dimensional locations of the 128 electrodes on the scalp were collected with either infrared color-enhanced 3D scanner (Taberna, Guarnieri, & Mantini, 2019) or Xensor digitizier (ANT Neuro, The Netherlands). To build each participant's highresolution head model, the participants underwent T1-weighted MRI acquisition using either a 3 T (N = 25) or a 1.5 T (N = 4) scanner (see Suppl.…”