Wavelength-dependent error minimization for quantitative spectroscopic photoacoustic tomography with a ring-array system

PURPOSE

Photoacoustic tomography (PAT) is a non-invasive and high-resolution imaging technique that can provide functional and molecular information from the optical properties of pathological tissues, such as cancer. Spectroscopic PAT (sPAT) is capable of supplying information such as oxygen saturation (sO2), which is an important biological indicator for diseases such as cancer. However, the wavelength dependent nature of sPAT makes it challenging to provide accurate quantitative measurements of tissue oxygenation beyond shallow depths. We have previously reported the utility of combined ultrasound tomography and PAT to achieve optical and acoustic compensated PAT images at a single wavelength and for enhanced PAT images at larger depths. In this work we further explore the utility of the optical and acoustic compensation PAT algorithm to minimize the wavelength dependency in sPAT by showcasing improvements in spectral unmixing.

MATERIALS AND METHODS

Two optically and acoustically characterized heterogenous phantoms were manufactured to test the ability of the system and developed algorithm to minimize the wavelength-dependence driven error in sPAT spectral unmixing. The PA inclusions within each phantom were composed of a mixture of two sulfate dyes, copper sulfate (CuSO4) and nickel sulfate (NiSO4), with known optical spectra. Improvements between uncompensated and optically and acoustically compensated PAT (OAcPAT) were quantified as the relative percent error between the measured results and the ground truth.

RESULTS

The results of our phantom studies demonstrate that OAcPAT can significantly improve the accuracy of sPAT measurements in a heterogenous medium and especially at larger inclusions depths which can reach to up to 12% improvement in measurement errors. This significant improvement can play a vital role in reliability of future in-vivo biomarker quantifications.

CONCLUSIONS

Utilizing UST for model-based optical and acoustic compensation of PAT images was proposed by our group previously. In this work, we further demonstrated the efficacy of the developed algorithm in sPAT by minimizing the error caused by the tissue's optical heterogeneity on improving spectral unmixing, which is a major limiting factor in reliability of sPAT measurements. Such synergistic combination of UST and PAT provides a window of opportunity to achieve bias-free quantitative sPAT measurements, which plays an important role in future pre-clinical and clinical utility of PAT.

Underwater light climate and wavelength dependence of microalgae photosynthetic parameters in a temperate sea

Studying how natural phytoplankton adjust their photosynthetic properties to the quantity and quality of underwater light (i.e. light climate) is essential to understand primary production. A wavelength-dependent photoacclimation strategy was assessed using a multi-color pulse-amplitude-modulation chlorophyll fluorometer for phytoplankton samples collected in the spring at 19 locations across the English Channel. The functional absorption cross section of photosystem II, photosynthetic electron transport (PET_λ) parameters and non-photochemical quenching were analyzed using an original approach with a sequence of three statistical analyses. Linear mixed-effects models using wavelength as a longitudinal variable were first applied to distinguish the fixed effect of the population from the random effect of individuals. Population and individual trends of wavelength-dependent PET_λ parameters were consistent with photosynthesis and photoacclimation theories. The natural phytoplankton communities studied were in a photoprotective state for blue wavelengths (440 and 480 nm), but not for other wavelengths (green (540 nm), amber (590 nm) and light red (625 nm)). Population-detrended PET_λ values were then used in multivariate analyses (partial triadic analysis and redundancy analysis) to study ecological implications of PET_λ dynamics among water masses. Two wavelength ratios based on the microalgae saturation parameter E_k (in relative and absolute units), related to the hydrodynamic regime and underwater light climate, clearly confirmed the physiological state of microalgae. They also illustrate more accurately that natural phytoplankton communities can implement photoacclimation processes that are influenced by in situ light quality during the daylight cycle in temporarily and weakly stratified water. Ecological implications and consequences of PET_λ are discussed in the context of turbulent coastal ecosystems.

Wavelength-dependent chlorine photolysis and subsequent radical production using UV-LEDs as light sources

UV-LEDs are considered as the most promising UV light sources, because it has the potential to replace conventional UV lamps in some water treatment applications in the foreseeable future. In this study, UV-LEDs at four wavelengths in the UV-C or near UV-C range (i.e., 257.7, 268, 282.3, and 301.2 nm) were used to investigate the wavelength-dependency on chlorine photolysis and its subsequent radical formation. The fluence-based photodecay rates of hypochlorous acid (HOCl) and hypochlorite (OCl^-) were monotonically correlated to their molar absorption coefficients and quantum yields, and the chlorine photodecay rates were much more significantly affected by molar absorption coefficients (β  = 0.949) than quantum yields (β  = 0.055). An empirical model that incorporated the chlorine photodecay rate constants, quantum yields, and molar absorption coefficients of HOCl and OCl^- was established, validated and then used to predict the chlorine photodecay rate at any wavelength (257.7-301.2 nm) and pH (5-10). The modelling results suggested that the maximum fluence-based rate constant (1.46 × 10^-4 m² J^-1) was obtained at 289.7 nm and pH 9.95. The wavelength dependency was larger at alkaline pH than at acidic pH, and the pH dependency was the largest at the longest wavelength. The formation of hydroxyl radicals (HO·) and reactive chlorine species (RCS) decreased with increasing wavelength at pH 6, and increased with increasing wavelength at pH 7. More HO· was formed at pH 6 than pH 7, but RCS showed the opposite pH-dependency. The findings in this study provide the fundamental information in selecting UV-LEDs with specific wavelength for enhancing/optimizing chlorine photodecay and/or its radical generation at different pHs in real-world applications.