Quadruple ultrasound, photoacoustic, optical coherence, and fluorescence fusion imaging with a transparent ultrasound transducer

Significance Multimodal imaging based on optics and ultrasound can provide guide images and complementary structural and functional information, thus improving the accuracy of medical diagnosis and treatment monitoring. However, because conventional ultrasound transducers are opaque, in multimodal imaging with optics, the optical devices must be placed off-axis from the ultrasound transducer. This off-axis arrangement is prone to misalignment, adds complexity and bulk to the system, and can result in a low signal-to-noise-ratio. Here, we present a transparent ultrasound transducer at the heart of a quadruple fusion imaging system that seamlessly integrates ultrasound imaging, photoacoustic imaging, optical coherence tomography, and fluorescence imaging, and we demonstrate the system’s use in imaging responses to both ophthalmologic injuries and oncologic diseases.

The Krimholtz-Leedom-Matthaei (KLM) model, one of the widely used one-dimensional models for designing a piezoelectric single crystal ultrasound transducer, allows researchers to better understand and optimize the transducer's acoustic and electrical performance. Together with the KLM model, we used the PiezoCAD (Sonic Concepts, Inc., USA) simulation tool, which is the most commonly used software for transducer design. The properties of each material in the design, such as its electromechanical coupling coefficient, longitudinal velocity, density, and attenuation, were all first set to default values. The physical dimensions of the elements, such as their shape, size, and thickness, were then set as parameters. For the individual layers of the TUT, we then chose LNO as a piezoelectric material, non-conductive epoxy as a backing layer, N-SF11 plano-concave lens as an acoustic lens and also a first matching layer, and parylene as a second matching layer. We initially set the thickness of the LNO to half lambda, based on the desired center frequency of the transparent transducer. The thickness of each matching layer was set to 1/4 th lambda, corresponding to the center frequency. Iterative simulation was then carried out using the thickness of each layer and the diameter of the transducer as parameters to find the optimal value for the desired dual frequencies of 8 MHz and 30 MHz. All the simulations were conducted assuming a water medium.

Method S2. Spectral Unmixing Algorithm.
Spectral unmixing of each PA B-scan image was performed as follows 1,2 : 1) To compensate for the attenuation of the optical fluence as the tissue depth increased, the background signal was calculated for each depth, and then the fluence was normalized by using these calculated values.
2) The fluence-compensated multispectral PA images were spectrally unmixed to the components of the oxy-hemoglobin (HbO2), the deoxy-hemoglobin (HbR), and the melanin by using the pseudoinverse matrix approach: where is the concentration of the n th component, is the fluence-compensated PA image at the k th wavelength, and is the normalized optical absorption coefficient of the n th component at the k th wavelength. By using these unmixed components of the HbO2 and HbR, the relative sO2 map was computed as follows: sO 2 = HbO 2 HbO 2 +HbR

Method S3. Gold Nanorod Synthesis and PEGylation of Gold Nanorods
We purchased cetyltrimethylammonium bromide (CTAB), gold(III) chloride hydrate (HAuCl4), ascorbic acid, and silver nitrate (AgNO3) from Sigma-Aldrich (USA). Sodium borohydride (NaBH4) was supplied from Honeywell Fluka (USA). mPEG-SH (2kDa) was purchased from SunBio (Republic of Korea). All reagents were used as received without further purification. The GNRs were prepared with a seeded-growth mechanism previously described 3,4 . Gold seeds were prepared by adding 0.6 mL of 0.01M pre-chilled NaBH4 solution to a 5 mL mixed solution containing 0.1 M CTAB and 0.25 mM HAuCl4. Then the mixture was shaken for 2 minutes with vigorous stirring. The growth solution was prepared by mixing 125 mL of 0.2 M CTAB, 125 mL of 1 mM HAuCl4, and 8 mL of 4 mM AgNO3. This orange-colored solution became transparent after adding 1.75 mL of 78.8 mM ascorbic acid for Au ion reduction. Finally, 0.3 mL of a brown-colored seed solution was added to the transparent growth solution, and the mixture became purple/brownish over 60 minutes. After the reaction, the GNR solution was purified by centrifugation at 16,000 rcf for 20 minutes. The purification step was repeated twice more to eliminate the remaining CTAB. The molar extinction coefficient, calculated from previously published literature 5 , was 3.46 × 10 9 M -1 cm - PEGylation of the GNRs was conducted by adding mPEG-SH (2kDa) to exchange the CTAB capped at the surface of the GNRs. A solution of mPEG-SH in distilled water (DW) was added to the GNRs in DW, and then the solution was stirred for 12 hr to obtain PEG-GNRs. After the reaction, the PEG-GNRs were purified by centrifugation at 16,000 rcf for 20 minutes. The characteristics of the PEG-GNRs, including their optical absorption properties, zeta potential, and transmission electron microscope (TEM) images, are summarized in SI Appendix, Fig. S14. The zeta potential was measured using a Nano Z (Malvern Instruments, UK). High-resolution TEM images with electron energy loss spectroscopy elemental mapping were obtained using a JEM-2200FS electron microscope (JEOL, Japan). All UV-Vis absorption spectra were obtained using a SpectraMax i3 plate reader (Molecular Devices, USA).            S11. ICP-MS results of Au in the 4T1 tumor and skin of mice at 8-hr after intravenous injection of the PEG-GNRs. Statistical differences were analyzed by performing one-way ANOVA. Data represent the mean ± SD (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). ICP-MS, inductively coupled plasma mass spectrometry; PEG-GNR, PEGylated gold nanorod; hr, hours and SD, standard deviation.    Table S1. Imaging specifications for individual systems. USI, ultrasound imaging; PAI, photoacoustic imaging; OCT, optical coherence tomography; OCTA, optical coherence tomography angiography; and FLI, fluorescence imaging.
Movie S1 (separate file). In vivo 3D photoacoustic imaging of a rat's eye before and after alkali burn.
Movie S2 (separate file). In vivo 3D optical coherence tomography of a rat's eye before and after alkali burn.