Functional human brain mapping using whole head very-high-density diffuse optical tomography
Poster No:
2421
Submission Type:
Abstract Submission
Authors:
Morgan Fogarty1, Sean Rafferty2, Anthony O'Sullivan2, Calamity Svoboda2, Zachary Markow1, Edward Richter1, Tessa George1, Kelsey King2, Dana Wilhelm2, Kalyan Tripathy2, Jason Trobaugh1, Adam Eggebrecht2, Joseph Culver2
Institutions:
1Washington University in St. Louis, St. Louis, MO, 2Washington University School of Medicine, St. Louis, MO
First Author:
Co-Author(s):
Introduction:
Functional neuroimaging has seen a rise in engaging and ecologically valid natural stimuli like podcasts and movies. While fMRI is the gold standard for functional neuroimaging, this technique involves a loud, unnatural scanning environment in the isolated magnet bore. Optical imaging with functional near infrared spectroscopy (fNIRS) allows for greater flexibility in imaging due to an open-air scanning environment, minimal (to no) acoustic noise, and portability (depending on the system). High-density diffuse optical tomography (HD-DOT) uses a dense array of optical elements to provide multiple multi-distance overlapping fNIRS light measurements. These measurements, together with tomographic reconstruction techniques, improve image quality over traditional fNIRS and recover brain activations and resting state networks with fidelity comparable to fMRI (Eggebrecht et al., 2014). However, HD-DOT has a resolution of ~15 mm, which doesn't extend to the physical limits of this imaging method (Markow et al., 2023). Here we explore the image quality improvements achieved when doubling the optode count and decreasing the inter-optode spacing from 13 mm to 9.75 mm while maximizing whole-head coverage for usage in naturalistic and decoding imaging studies.
Methods:
Our very-high-density DOT (VHD-DOT) system consists of 255 sources and 252 detectors in a regular rectangular grid of 9.75 mm nearest neighbor spacing (Fig 1A), resulting in approximately 5,000 measurements per wavelength. While only a 33% decrease in spacing from our HD-DOT system, this generates an approximately 200% increase in the number of measurements and directly translates to higher resolution and image quality as the point-spread function decreases in size with increasing grid density (Fig 1B-E). We also expanded the field of view to include bilateral aspects of occipital, temporal, parietal, and prefrontal areas (Fig 1F). The system was validated with 11 healthy adult participants completing a series of standard block-design localizer tasks (finger tapping, word hearing, verb generation, and retinotopy), and movie viewing in both VHD-DOT and MRI. The movie viewing task consisted of two viewings of the same 10-minute movie clip (Fishell et al., 2019). Beta value maps were generated using a GLM approach and responses were aggregated across subjects using random effect t-statistics. Functional localizer responses were also included in a six-way decoding task by generating a template from half of the blocks (3 blocks total) in each run and then computing the correlation between the template and the remaining 3 blocks individually. Similarly, correlations between the first and second movie viewing runs were used for template-based decoding across varying time durations.
Results:
Through simulations, we prove our VHD-DOT system outperforms our HD-DOT system using standard image quality metrics (Fig 1B-E). Data quality was assessed in measurement space for each imaging session (Fig 1G-I). Functional localizer t-statistic maps appear consistent between fMRI and VHD-DOT across all tasks (Fig 1J-K). Localizer decoding offers insight into imaging performance across the cap for each subject (Fig 2A-C). Localizer decoding performance across all subjects resulted in an accuracy of 73.7% (Fig 2D) – well above the chance level of 17% when using oxy-, deoxy- and total hemoglobin (Fig 2E-G). For movie viewing, decoding accuracies remained above chance across the varying clip lengths and with an increasing number of clip choices (Fig 2H-K).
Conclusions:
Overall, these results show that we can map brain function across the majority of the cortex using our VHD-DOT system. This builds upon our previously reported HD-DOT systems to improve the image quality and resolution of whole-head optical imaging in adults. This system is promising for future work involving whole-head studies of resting state functional connectivity and expanding our decoding paradigms to more naturalistic opportunities (Tang et al., 2023).
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Novel Imaging Acquisition Methods:
NIRS 1
Keywords:
ADULTS
Hearing
Modeling
Motor
Near Infra-Red Spectroscopy (NIRS)
OPTICAL
Optical Imaging Systems (OIS)
Vision
Other - High density diffuse optical tomography; Naturalistic stimuli
1|2Indicates the priority used for review
Provide references using author date format
Fishell, A. K., Arbeláez, A. M., Valdés, C. P., Burns-Yocum, T. M., Sherafati, A., Richter, E. J., Torres, M., Eggebrecht, A. T., Smyser, C. D., & Culver, J. P. (2020). Portable, field-based neuroimaging using high-density diffuse optical tomography. NeuroImage, 215, 116541. https://doi.org/10.1016/j.neuroimage.2020.116541
Fishell, A. K., Burns-Yocum, T. M., Bergonzi, K. M., Eggebrecht, A. T., & Culver, J. P. (2019). Mapping brain function during naturalistic viewing using high-density diffuse optical tomography. Scientific Reports, 9(1). https://doi.org/10.1038/s41598-019-45555-8
Markow, Z. E., Trobaugh, J. W., Richter, E. J., Tripathy, K., Rafferty, S. M., Svoboda, A. M., Schroeder, M. L., Burns-Yocum, T. M., Bergonzi, K. M., Chevillet, M. A., Mugler, E. M., Eggebrecht, A. T., & Culver, J. P. (2023). Ultra-high density imaging arrays for diffuse optical tomography of human brain improve resolution, signal-to-noise, and information decoding. https://doi.org/10.1101/2023.07.21.549920
Tang, J., LeBel, A., Jain, S., & Huth, A. G. (2023). Semantic reconstruction of continuous language from non-invasive brain recordings. Nature Neuroscience. https://doi.org/10.1038/s41593-023-01304-9
Tripathy, K., Markow, Z. E., Fishell, A. K., Sherafati, A., Burns-Yocum, T. M., Schroeder, M. L., Svoboda, A. M., Eggebrecht, A. T., Anastasio, M. A., Schlaggar, B. L., & Culver, J. P. (2021). Decoding visual information from high-density diffuse optical tomography neuroimaging data. NeuroImage, 226. https://doi.org/10.1016/j.neuroimage.2020.117516