Transfontanelle photoacoustic imaging for in-vivo cerebral oxygenation measurement

Abstract

The capability of photoacoustic (PA) imaging to measure oxygen saturation through a fontanelle has been demonstrated in large animals in-vivo. We called this method, transfontanelle photoacoustic imaging (TFPAI). A surgically induced 2.5 cm diameter cranial window was created in an adult sheep skull to model the human anterior fontanelle. The performance of the TFPAI has been evaluated by comparing the PA-based predicted results against the gold standard of blood gas analyzer measurements.

Figure 1

In-vivo brain oxygenation measurement setup using TFPAI. (a) Schematic of the experimental setup and process flow of monitoring vitals (prepared in Adobe Illustrator CS6 (version: 16), URL: https://www.adobe.com/products/illustrator.html), (b) Key components of the experimental setup for sheep brain oxygen saturation study, (c) TFPAI probe configuration and placement over cranial window—a magnified version of the dashed yellow box from (b), and (d) 2.5 cm diameter surgically induced cranial window representing anterior fontanelle area—gray dashed circular area. Q-Sw: Q-switch, FL: flash lamp, Trig: trigger, DAQ: data acquisition.

Figure 2

Absorption coefficient spectra of endogenous tissue chromophores and reconstructed PA images overlaid on ultrasound image. (a) HbO₂ and HbR, 150 g/L in blood; water, 80% by volume in tissue; lipid. Figure adapted with from. In the inset, an MRI cross-sectional (sagittal plane) of ex-vivo sheep brain depicting major anatomical structure and relative ROI locations (b) ROIs considered for estimating sO₂ from reconstructed PA images overlaid on US imaging plane at (i) 750 nm, (ii) 758 nm, (iii) 798 nm, (iv) 850 nm. Anatomical structures annotated in (ii) 1: skull, 2: fibrous tissue growth within the cranial window, 3: superficial vasculature, 4: lateral ventricle. (Prepared in Adobe Illustrator CS6 (version: 16), URL: https://www.adobe.com/products/illustrator.html).

Figure 3

Correlation between estimated (from PA images) and measured (blood gas analyzer) O₂ saturation for all trials, (a) ROI #1, (b) ROI #2, and (c) ROI #3. Dashed line: arterial blood gas analyzer, Solid line (red): linear fit of estimated sO₂ from PA images.

Figure 4

Photoacoustic Oxygen saturation (PA-sO₂) validation. (a) PA estimated sO₂ plotted against measured sO₂ from blood gas analyzer. Error bars represent standard deviation (SD) among the trials 1–3, (b) Bland–Altman estimation analysis plot showing error estimation for all trials. SD: standard deviation.