@inproceedings{bartkowski_towards_2022,
    author = {Bartkowski, C and Nandori, A and von L{\"u}hmann, Alexander and Schmitz, Christoph},
    address = "Boston, USA",
    title = "Towards a fully integrated {Smart} {Textile} patch-based cap for multi-distance {CW} {fNIRS} whole-head imaging",
    copyright = "All rights reserved",
    abstract = "Introduction: Whole-head brain imaging with fNIRS becomes increasingly relevant in the neurosciences, in particular in fields that require high usability (easy set-up) and wearability, e.g., measurements in naturalistic environments. Solutions need to be comfortable for a range of head sizes and -shapes while simultaneously ensuring a good contact interface, stability, and signal quality for both deep and superficial measurements. To meet these requirements, we integrated Smart Textile techniques with opto-electronics, shielding, and mechanical coupling. The achieved reduction in weight and bulkiness greatly enhances usability and reduces preparation time. Methods: We created functional prototypes of a modular flexible patch-based multi-optode solution. Each patch is made from industry-standard flex PCB and arranges eight bi-color-LEDs and eight Si-PD based detectors in a rectilinear pattern, forming 23 30-mm source-detector-channels. Additional Si-PD detectors are placed next to each source to form eight additional short-separation (6 mm) channels. To assure good hair penetration and stable skin contact, the long-distance optodes are individually spring-loaded, and the short-separation detectors feature smalldiameter front-ends. (see Fig. 1 A). The circuit boards employ a meandering shape and jersey-fabric lamination as described in [1] to aid flexibility (Fig. 1 B). Up to four patches are clipped into magnetic elastic holders in an adapted EEG cap (EASCYAP) and driven by two NIRSport2 imagers (NIRx Medizintechnik GmbH, Germany). Holders provide a secondary spring loading mechanism to improve optical contact and anatomical adaptability. Signal quality and comfort were measured in 9 subjects (3♀, head sizes 55–60, 6 min resting state) and compared with a standard NIRx setup of matching optode layout (EASYCAP; dual-tip, 32 source/28 detector montage; 32 short distance sensors). Results: Fig. 1 (C) shows the resulting whole-head setup with four patches (fully prepared within 30min) yielding 127 Fig. 1 (A) Double spring-loaded optode (B) Flextextile patch (C) Whole-head setup (D) Signal quality after calibration in NIRx Aurora standard- and 32 Software short-distance channels. The 8x8 textile patches have a 26\\% lower weight compared to the traditional setup and show no mechanical weaknesses or vulnerability to water and alcohol (Isopropanol 70\\%) under appropriate use. Signal quality of long channels showed comparable performance and sensing density as traditional, individually wired optodes, except in occipital areas for subjects with long hair. Short-channel signal quality varied strongly across subjects. Conclusion: Our approach of a Smart Textile-based fNIRS probe shows promising results regarding mechanical robustness, signal quality, and comfort. Long channel signal quality is excellent; however, contact-pressure needs to be optimized in specific areas of the head. Our novel short channel design has passed the proof-of-concept stage but requires more detailed investigations. Preparation time for a whole head imaging setup was massively improved (30 min vs 4 h).",
    language = "en",
    booktitle = "Proc. {Biennial} {Meeting} of the {Society} for {fNIRS} 2022",
    publisher = "SfNIRS",
    month = "October",
    year = "2022",
    pages = "1"
}