Mapping Human Brain Function with Diffuse Optical Tomography

Background:  Mapping human brain function has been transformative in systems neuroscience. However, traditional functional neuroimaging, with positron emission tomography (PET) or functional magnetic resonance imaging (fMRI), cannot be used when applications require portability, or are contraindicated because of ionizing radiation (PET) or implanted metal (fMRI). Optical neuroimaging offers a noninvasive alternative that is radiation free and compatible with implanted metal and electronic devices (e.g., pacemakers). However, optical imaging technology has heretofore lacked the combination of spatial resolution and wide field-of-view sufficient to map distributed high level brain functions. 

Objectives: In this paper we present a large field-of-view high-density diffuse optical tomography (HD-DOT) system with anatomical modeling approaches that collectively provide new imaging capabilities and enable novel milestones for optical neuroimaging. Higher order brain functions, specifically area MT and language processing areas, are mapped using tasks as are a set of resting state networks (RSNs), including the default mode, dorsal attention, and fronto-parietal control networks.  

Methods:  Eight subjects underwent non-concurrent HD-DOT and fMRI sessions. The large field-of-view high-density DOT instrument, based on previous opto-electronic circuitry (Zeff et al., 2007), uses 96 optode source positions and 92 optode detector positions in an imaging array that wraps around the subject’s head covering the occipital, parietal, temporal, and prefrontal cortex. Data was averaged across subjects and t-maps were created using one-sided random effects t-tests. Subjects were presented tasks to stimulate higher-order cognitive areas including visual area MT (static and moving dots; Swallow et al., 2003) and language processing areas  (Petersen et al., 1988) associated with hearing, reading, speaking, and generation of words. Seed-voxel maps of functional connectivity were generated from resting state data (10 minutes; subjects fixated on a small cross on a screen and were instructed to think of nothing in particular). Pearson r-values were converted to Fisher-z vaues and were averaged across subjects.

Results:  To measure higher order brain function, we first attempted imaging higher order visual processing in region MT. Non-concurrently recorded fMRI activations show strong correlation in their temporal response. Also, the regions detected show excellent spatial co-localization. Second, the language processing centers of the brain were broken down into sensory (visual and auditory), motor (motor mouth), and cognitive association areas (verb generation in Broca’s area) with spatial correspondence between HD-DOT and fMRI. We believe this is the first reported mapping of the hierarchy of the language system with optical neuroimaging. Third, strong bilateral patterns of functional connectivity are co-registered between HD-DOT ΔHbO and fMRI BOLD RSN maps in the sensory motor networks. Additionally, in areas associated with higher order RSNs, function connections are evident not only between bilateral regions, but also between differentiable anterior-posterior constellations of regions. The seed-seed correlation values shows qualitative and quantitative agreement between the modalities. 

Conclusions:  Collectively these maps of tasks and resting state function in adult humans clearly indicate that high-density DOT can be a practical and powerful tool for functional brain mapping of children with Autism Spectrum Disorders.