Mapping functional hemodynamic and metabolic responses to dementia: a broadband spectroscopy pilot study

Broadband near-infrared spectroscopy (bNIRS) can simultaneously monitor several chromophores, including the oxidative state of cytochrome c-oxidase (oxCCO), an oxygen metabolism biomarker, the activity of which is altered in Alzheimer's disease. Being a portable and noninvasive neuromonitoring technique, bNIRS could provide accessibility to brain-specific biomarkers and aid in the dementia diagnostic pathway.

We use bNIRS-recorded functional hemodynamic and oxCCO changes to assess their relevance in Alzheimer's disease diagnosis.

Using a visual stimulus paradigm, we recorded functional changes in oxy-, deoxy-hemoglobin and oxCCO in three similarly aged cohorts: healthy controls ( $n=5$ ), individuals with mild cognitive impairment ( $n=7$ ), and individuals with early Alzheimer's dementia ( $n=7$ ). We then selected features from these functional responses to find the best correlation with clinical cognitive markers (cognitive and behavioral test scores and clinical diagnoses) using canonical correlation analysis (CCA).

We found individual variations in peak amplitude and time-to-peak for all the stimulus-evoked bNIRS signals across the three cohorts. CCA showed a strong correlation between bNIRS features and the clinical cognitive markers ( $r=0.902$ ). However, repeating the same analysis by excluding the bNIRS oxCCO features leads to a significantly lower correlation ( $r=0.687$ ) with the clinical markers.

oxCCO could be a crucial biomarker, partly explaining cognitive differences with dementia. bNIRS uniquely provides a portable and noninvasive technique to monitor several chromophores simultaneously, including oxCCO, with potential future applications in diagnosing and tracking dementia progression.