Non-contact photoacoustic characterization of retinal vasculature for brain-injury assessment

Abstract

This paper describes use of a non-contact photoacoustic method to characterize blood vessels in the retina for brain injury assessment. Retinal hemorrhages (RHs) and vascular abnormalities can serve as diagnostic indicators of head trauma. Ensuring an accurate and complete diagnosis of the trauma requires a quantitative description of the hemorrhages. In this work, we describe an optical interferometry-based photoacoustic (PA) microscope (PAM) to remotely evaluate retinal tissue at sub-micron resolution with an ultra-broadband detection bandwidth (1 GHz). To demonstrate feasibility, ex-vivo porcine retina specimens containing typical blood-vessel networks were imaged using our all-optical PAM. PA signals were generated in the tissue by a 532-nm pulsed laser, and PA-induced broadband ultrasound signals were detected at the surface of the specimen using an optical interferometer operating at 532 nm. K-space reconstruction methods were employed to obtain the tissue properties by time-reversing and back-propagating the measured signals into the medium. A tunable pulsed laser was employed to perform multispectral PA imaging of the vasculature to demonstrate feasibility of measuring blood volume, oxygen concentration and lipids. Multispectral PA data were obtained to a depth of 5 mm, in raster-scanned steps of 50 microns over the wavelength range of 680 to 950 nm.