A Cost-Efficient Multiwavelength LED-Based System for Quantitative Photoacoustic Measurements

The unique ability of photoacoustic (PA) sensing to provide optical absorption information of biomolecules deep inside turbid tissues with high sensitivity has recently enabled the development of various novel diagnostic systems for biomedical applications. In many cases, PA setups can be bulky, complex, and costly, as they typically require the integration of expensive Q-switched nanosecond lasers, and also presents limited wavelength availability. This article presents a compact, cost-efficient, multiwavelength PA sensing system for quantitative measurements, by utilizing two high-power LED sources emitting at central wavelengths of 444 and 628 nm, respectively, and a single-element ultrasonic transducer at 3.5 MHz for signal detection. We investigate the performance of LEDs in pulsed mode and explore the dependence of PA responses on absorber's concentration and applied energy fluence using tissue-mimicking phantoms demonstrating both optical absorption and scattering properties. Finally, we apply the developed system on the spectral unmixing of two absorbers contained at various relative concentrations in the phantoms, to provide accurate estimations with absolute deviations ranging between 0.4 and 12.3%. An upgraded version of the PA system may provide valuable in-vivo multiparametric measurements of important biomarkers, such as hemoglobin oxygenation, melanin concentration, local lipid content, and glucose levels.

Automated summary

The article describes a compact, cost-efficient, multiwavelength PA sensing system for quantitative measurements using high-power LED sources emitting at central wavelengths of 444 and 628 nm, and a single-element ultrasonic transducer at 3.5 MHz for signal detection. It explores the performance of LEDs in pulsed mode and the dependence of PA responses on absorber's concentration and applied energy fluence in tissue-mimicking phantoms. The developed system is capable of accurate estimations with absolute deviations ranging between 0.4 and 12.3%, demonstrating potential for in-vivo multiparametric measurements of important biomarkers.