Open Access
Add to BibSonomyAbstract
The conventional optical coherence tomography (OCT) images based on enhanced scattering and the photothermal (PT) images based on enhanced absorption of the localized surface plasmon (LSP) resonance of Au nanorings (NRIs) in a bio-tissue sample are demonstrated with the scans of an OCT system (1310-nm system), in which the spectral range covers the LSP resonance peak wavelength, and another OCT system (1060-nm system), in which the spectral range is away from the LSP resonance peak wavelength. A PT image is formed by evaluating the modulation frequency (400 Hz) response of an excitation laser with its wavelength (1308 nm) close to the LSP resonance peak at 1305 nm of the Au NRI solution. With the scan of the 1310-nm OCT system, the Au NRI distribution in the bio-tissue sample can be observed in both conventional OCT and PT images. However, with the scan of the 1060-nm OCT system, the Au NRI distribution can be clearly observed only in the PT image. The diffusion process of Au NRIs in the bio-tissue sample can be traced with the scan of either OCT system. Based on phantom experiments, it is shown that the PT image can help in resolving the ambiguity of a conventional OCT image between the enhanced scattering of Au NRIs and the strong scattering of a tissue structure in the 1310-nm OCT scanning. Also, under the condition of weak intrinsic sample scattering, particularly in the scan of the 1060-nm system, the PT signal can be lower than a saturating level, which is determined by the excitation power. By increasing OCT system signal-to-noise ratio or M-mode scan time, the PT signal level can be enhanced.
© 2014 Optical Society of America
1. Introduction
Localized surface plasmon (LSP) represents locally collected oscillation of excited electrons on a confined metal structure, such as a metal nanoparticle (NP) [1, 2]. The LSP resonance of a metal NP shows two important behaviors, including enhanced scattering and enhanced absorption around the resonance wavelength. Both behaviors can be used as the mechanisms of contrast enhancement in optical imaging, particularly in optical coherence tomography (OCT). At the LSP resonance of a metal NP, the enhanced coherent scattering of incident OCT light can significantly increase the interfered signal intensity of an OCT system [3–6]. On the other hand, the enhanced absorption at the LSP resonance of a metal NP can heat up the surrounding medium to produce a photothermal (PT) effect. This effect has been considered for the inactivation of cancer cells [6–14]. Through a bio-conjugation process, metal NPs can be adsorbed by cancer cells or even internalized by the cells. By illuminating the cells with an excitation laser of a wavelength close to the LSP resonance peak, the induced PT effect can inactivate the cancer cells (the PT therapy). Such a PT effect can also be applied to optical imaging through the generation of optical path variation (through thermal expansion or refractive index change) in OCT scanning [15–20]. By applying a periodically modulated excitation laser to the sample for heating the scanning spot, a modulated phase signal can be monitored in M-mode OCT scanning [21]. The PT OCT technique can be used to track metal NPs delivered into a bio-tissue for monitoring the function of bio-conjugation and the distribution of metal NPs or cancer cells. For deep tumor scanning, a catheter or needle OCT configuration with the excitation laser integrated into the system needs to be developed.
Because of the inert property of Au, Au NPs are widely used in biomedical application. In the past, the LSP resonance wavelengths of the used Au NPs for PT therapy or PT OCT applications focused at the spectral range between 500 and 900 nm. The used Au NPs include sphere-like solid NPs [22], nanorods [23–25], silica/Au nanoshells [6, 26, 27], hollow NPs [28, 29], and nanocages [15, 30, 31]. Generally speaking, at the LSP resonance of an Au NP with a shorter resonance wavelength, the absorption cross section is relatively larger than the scattering cross section. Therefore, the enhanced scattering at the LSP resonances of the aforementioned Au NPs is weak for OCT application. Recently, on-substrate Au nanorings (NRIs) have been fabricated for sensing application [32–34]. Also, bio-conjugated Au NRI solution has been successfully fabricated for demonstrating the contrast enhancement of an OCT image and the effective PT inactivation of cancer cells [35–37]. Compared with other Au NPs, an Au NRI has more geometry parameters for controlling the LSP resonance wavelength [38]. In particular, it can extend the LSP resonance wavelength into the spectral range between 1000 and 1300 nm, in which tissue scattering is weaker such that light penetration is deeper. Meanwhile, an Au NRI can be designed for its scattering and absorption cross sections to be comparable such that the imaging applications of a larger variety are feasible.
In forming a PT image based on OCT scanning, the phase information of OCT signal is to be acquired for analysis. Therefore, as long as the wavelength of the excitation laser matches the LSP resonance spectrum of Au NP for resonance excitation, the OCT operation wavelength can be away from the spectral range of LSP resonance. However, this condition may lead to an ambiguity in tracking Au NPs in a tissue with OCT scanning. In this paper, we apply Au NRI solution with the LSP resonance wavelength around 1300 nm to a pig liver sample for demonstrating the PT OCT operation. In particular, we compare the scanning results of two OCT systems with the operation central wavelengths at 1310 and 1060 nm. In the scanning operation of the 1310-nm OCT system, both enhanced scattering and absorption of LSP resonance can contribute to the contrast enhancements of scanning results. However, in the scanning operation of the 1060-nm OCT system, only enhanced absorption of LSP resonance makes the contribution. Their differences in conventional OCT and PT images are evaluated with bio-tissue and phantom experiments. In section 2 of this paper, the procedures of Au NRI fabrication and sample preparation are described. Also, the setups of the two OCT systems are introduced. Then, the scanning results, including the conventional OCT and PT images, are reported in section 3. Next, discussions with phantom experiments are presented in section 4. Finally, conclusions are drawn in section 5.
2. Au nanoring fabrication, sample preparation, and optical coherence tomography systems
The fabrication of Au NRI is based on the technique combination of nano-imprint lithography, reactive ion etching (RIE), Au deposition, and secondary metal sputtering on a polymer substrate. Briefly speaking, first a nanopillar array is formed on a polymer substrate through nano-imprint lithography. Then, RIE with O2 plasma is applied for adjusting the height and diameter of the nanopillars. Next, a thin Au layer is deposited onto the substrate to serve as the source of secondary sputtering. The secondary sputtering is implemented through an RIE process with CHF3 plasma. In this process, the Au atoms on the tops of the nanopillars are removed. Meanwhile, the Au atoms on the substrate surface in the gap regions between nanopillars are sputtered onto the sidewalls of the nanopillars to form a ring shape. Then, another step of O2 RIE is applied for removing the polymer inside the Au ring structure. In this stage, the background substrate level is also lowered to form new pillars with the Au ring structures at the tops such that the formed Au NRIs can be easily lifted off. To avoid aggregation after liftoff, the substrate with Au NRIs is immersed in a biolinker solution to form the carboxyl groups on the surface of Au NRIs before liftoff. The biolinker solution is prepared by mixing a Nanothinks acid16 (5mM in ethanol, Sigma-Aldrich) solution of 20 μL with 50 mL de-ionized water. Then, the sample is rinsed in de-ionized water to remove the residual biolinker. Finally, with ultrasonic vibration, Au NRIs can be transferred into de-ionized water. The detailed fabrication procedures for bio-conjugated Au NRI solution can be found in an earlier publication [37]. The insert of Fig. 1 shows the tilted scanning electron microscopy (SEM) image of the fabricated Au NRI array before liftoff. The outer diameter, inner diameter, and height of the Au NRI are ~155, ~130, and ~50 nm, respectively. Hence, the ring thickness is about 12.5 nm. Figure 1 shows the normalized extinction spectrum of the Au NRI solution in the spectral range of 400-1370 nm. The major peak of the extinction spectrum, which is located at 1305 nm in wavelength, as indicated by the vertical (pink) dashed line, corresponds to the cross-ring LSP dipole resonance of the Au NRI in water. This is the LSP resonance mode we will use for producing enhanced scattering and absorption. The broad secondary hump between 600 and 850 nm in the extinction spectrum originates from the merge of the higher-order cross-ring LSP resonance mode and the axial LSP dipole resonance mode. The normalized light source spectra of the two OCT systems are also plotted in Fig. 1 with the central wavelengths at 1310 and 1060 nm. Also, the thick (green) arrow is drawn to indicate the excitation laser wavelength at 1308 nm for generating the PT effect. Here, one can see that the LSP resonance strength in the spectral range of the 1310-nm OCT system is significantly stronger than that in the spectral range of the 1060-nm OCT system.
Fig. 1 Normalized extinction spectrum of the Au NRI solution in the spectral range of 400-1370 nm with the major LSP resonance peak at 1305 nm in wavelength, as indicated by the vertical (pink) dashed line. The insert shows the tilted SEM image of the fabricated Au NRI array before liftoff. The light source spectra of the two OCT systems are also plotted here with the central wavelengths at 1310 and 1060 nm. Meanwhile, the thick (green) arrow is drawn to indicate the excitation laser wavelength at 1308 nm.
Au NRI solution is delivered into a pig liver sample through a hollow fiber for OCT scanning, as shown in the picture of Fig. 2(a). The hollow fiber is connected with an injection needle through a pipette tip for injecting Au NRI solution. The OCT scanning and excitation laser illumination are aligned to be at the same location on the pig liver sample, which is near the distal end of the hollow fiber. The OCT B-mode scan is along the direction perpendicular to the hollow fiber. The particle concentration of the injected Au NRI solution is estimated to be 2 x 109 cm−3. The volume of injected Au NRI solution is 2 micro-liters. To confirm the LSP resonance effect of Au NRI, phosphate buffered saline (PBS) of also 2 micro-liters is injected at a different location of the pig liver sample for OCT scanning. To investigate certain details regarding the LSP resonance effects on OCT scanning, we use agar as the host material for preparing six phantom samples. In samples A-E, Au NRIs of a fixed concentration and TiO2 NPs of various concentrations are homogeneously mixed into agar before it is solidified. For preparing samples A-E, 2-mg agar powder and 0, 0.3, 0.6, 0.9, and 1.2-mg TiO2 powders, respectively, are dissolved in 1-mL Au NRI solution (concentration at ~2 x 109 cm−3). For sample F (a control sample), 2-mg agar powder and 0.6-mg TiO2 powder are dissolved in 1-mL de-ionized water. Before solidification, the agar solution is injected into a tapering capillary tube with the distal end sealed with glue. After the injection of agar solution, the proximal end is also sealed with glue. As shown in Fig. 2(b), the capillary tube with the solidified agar mixture inside is glued onto a microscope slide for cross-sectional OCT scanning with excitation laser illumination at the same location.
Fig. 2 (a): Photograph of the used pig liver sample with Au NRI solution or PBS delivered into the sample through a hollow fiber. The collocated OCT scanning (represented by the continuous arrow) and excitation laser illumination (represented by the dashed arrow) are close to the distal end of the hollow fiber. (b): Photograph of the phantom setup. The OCT scanning light (cross-sectional scanning of the capillary tube) and the excitation laser are incident from the top to illuminate the sample at the same location.
The two OCT systems with the operation central wavelengths at 1310 and 1060 nm have the similar setup of a swept-source OCT configuration, as schematically shown in Fig. 3(a) [39]. In the 1310-nm system, the swept source has a sweeping wavelength range of 170 nm (Santec, HSL2100). The central wavelength and scanning rate of the light source are 1310 nm and 20 kHz, respectively. In the 1060-nm system, the swept source has a sweeping wavelength range of 100 nm (Axsun, SS-1060 nm). The central wavelength and scanning rate of the light source are 1060 nm and 100 kHz, respectively. In the OCT setup, besides the fiber Mach-Zehnder interferometer to form the main body of the OCT system, 5% of the light source power is sent to a fiber Bragg grating (Lead Fiber Optics Co.) with the Bragg wavelength at 1260 nm for the 1310-nm OCT system (at 998 nm for the 1060-nm OCT system) through a circulator. The returned signal from the fiber Bragg grating is monitored by a detector (Thorlabs, PDB145C) for synchronizing the OCT signal acquisition with the frequency sweeping of the swept source. The interfered spectral signals are monitored by a balanced detector (Thorlabs, PDB150C-AC) and stored in a personal computer (PC) through a data acquisition card (Alazar Tech, ATS-9350). All the fiber couplers and circulators in the system are manufactured by the company of Thorlabs. In the sample arm of the 1310-nm OCT system [see Fig. 3(b)], before the OCT signal light hits a scanning galvanometer, a beam splitter is used for combining the excitation laser with the OCT light beam such that both light beams can be scanned through the scanning galvanometer (Thorlabs, GVS002) and focused onto the sample at the same location. In the sample arm of the 1060-nm OCT system [see Fig. 3(c)], a dichroic mirror, instead of a beam splitter, is used for combining the excitation laser with the OCT light beam. The excitation laser is carefully controlled such that the power illuminating the sample is the same in the two OCT systems at ~15 mW (unless specified). The focal spot size of the excitation laser at the sample depth of ~600 μm is ~11 μm and hence the applied power density is ~158 W/mm2. Unless specified, the modulation frequency of the excitation laser is 400 Hz and the M-mode scan time is 50 ms. In the 1310-nm OCT system, without using the beam splitter for combining the excitation laser into the sample arm, the system signal-to-noise ratio (SNR) is ~92 dB, which is obtained from the measurement of mirror reflection. After the use of the beam splitter, the system SNR is reduced to ~86 dB. In this situation, the phase sensitivity is ~0.008 rad. Although the available source power of the 1060-nm OCT system can be higher for achieving a higher system SNR, for a reasonable comparison between the scan results of the two OCT systems, the source power of the 1060-nm system is controlled for maintaining the same system SNR at ~86 dB as that of the 1310-nm system. In this situation, the phase sensitivity of the 1060-nm system is ~0.006 rad. The incident OCT light powers onto the sample in the two systems are the same at ~1.3 mW. The axial resolutions of the 1310-nm and 1060-nm OCT systems are ~6.4 and ~7.3 μm, respectively. The spot sizes of OCT light beams at the focal depth (~600 μm in depth) are ~13 and ~11 μm in the 1310-nm and 1060-nm systems, respectively. They correspond to the individual lateral resolutions. The power densities of the OCT light beams at the focal depth are 9.8 and 13.7 W/mm2 in the 1310-nm and 1060-nm systems, respectively.
Fig. 3 (a): Common optical setup of the two OCT systems with the operation central wavelengths at 1310 and 1060 nm. (b): Setup of the sample arm in the 1310-nm OCT system, in which a beam splitter is used for combining the excitation laser with the OCT light beam. (c): Setup of the sample arm in the 1060-nm OCT system, in which a dichroic mirror is used for combining the excitation laser with the OCT light beam.
3. Scanning results of optical coherence tomography
Figure 4(a) shows the M-mode scan image for 50 ms of an A-mode scan near the distal end of the injection hollow fiber by using the 1310-nm OCT system when PBS is injected into the pig liver sample. During OCT scanning, the sample is simultaneously illuminated by the modulated excitation laser. The M-mode-scan intensity and differential phase profiles at the depth indicated by the (red) arrow are shown in Figs. 4(b) and 4(c), respectively. The differential phase of a pixel is obtained by taking the difference between its phase value and that in the pixel right above it in the same A-mode scan [40]. The phase difference between the adjacent pixels is obtained through the multiplication of the complex signal of a pixel with the complex conjugated signal of the neighboring pixel. With this procedure for obtaining the phase difference, the ambiguity of phase wrapping can be minimized. As shown in Figs. 4(b) and 4(c), no periodic feature can be seen in both intensity and differential phase variations with time. Figures 4(d) and 4(e) show the corresponding spectra obtained by Fourier-transforming the M-mode-scan signals in Figs. 4(b) and 4(c), respectively. Except the dc component, there is no feature in those spectra. The results indicate that although the sample may somewhat absorb the excitation laser, the used laser power level (~15 mW) is not strong enough for generating a PT effect to be monitored by the OCT scanning when no Au NRI is delivered into the sample.
Fig. 4 (a): M-mode scan image for 50 ms of an A-mode scan near the distal end of the injection hollow fiber by using the 1310-nm OCT system when PBS is injected into the pig liver sample and the excitation laser is turned on. (b) and (c): M-mode-scan intensity and differential phase profiles, respectively, at the depth indicated by the (red) arrow in part (a). (d) and (e): Corresponding spectra obtained by Fourier-transforming the M-mode-scan signals in parts (b) and (c), respectively.
Figure 5(a) shows the M-mode scan image for 50 ms of an A-mode scan near the distal end of the injection hollow fiber by using the 1310-nm OCT system when Au NRI solution is injected into the pig liver sample. However, during OCT scanning, the modulated excitation laser is turned off. The M-mode-scan intensity and differential phase profiles at the depth indicated by the (red) arrow are shown in Figs. 5(b) and 5(c), respectively. Also, Figs. 5(d) and 5(e) show the corresponding spectra obtained by Fourier-transforming the M-mode-scan signals in Figs. 5(b) and 5(c), respectively. Without modulated excitation laser, there is no feature in those spectra except the dc component. Then, Figs. 6(a)-6(e) show the similar results to Figs. 5(a)-5(e) except that the modulated excitation laser is now turned on. Although the spectrum of intensity variation in Fig. 6(d) shows no feature except the dc component, that of differential phase in Fig. 6(e) does demonstrate a clear feature at 400 Hz. After subtracting the average noise level, the spectral density at 400 Hz is defined as the PT signal intensity and will be used for plotting the PT image. Figures 7(a)-7(e) show the similar results to Figs. 6(a)-6(e) except that now the 1060-nm OCT system is used for scanning. With both conditions of Au NRI injection and excitation laser illumination, a feature at 400 Hz can be seen in the spectrum of differential phase, as shown in Fig. 7(e). Compared with Fig. 6(e), the PT SNR ratio in Fig. 7(e) based on the scanning of the 1060-nm system is lower.
Fig. 5 (a): M-mode scan image for 50 ms of an A-mode scan near the distal end of the injection hollow fiber by using the 1310-nm OCT system when Au NRI solution is injected into the pig liver sample and the excitation laser is turned off. (b) and (c): M-mode-scan intensity and differential phase profiles, respectively, at the depth indicated by the (red) arrow in part (a). (d) and (e): Corresponding spectra obtained by Fourier-transforming the M-mode-scan signals in parts (b) and (c), respectively.
Fig. 6 (a): M-mode scan image for 50 ms of an A-mode scan near the distal end of the injection hollow fiber by using the 1310-nm OCT system when Au NRI solution is injected into the pig liver sample and the excitation laser is turned on. (b) and (c): M-mode-scan intensity and differential phase profiles, respectively, at the depth indicated by the (red) arrow in part (a). (d) and (e): Corresponding spectra obtained by Fourier-transforming the M-mode-scan signals in parts (b) and (c), respectively.
Fig. 7 (a): M-mode scan image for 50 ms of an A-mode scan near the distal end of the injection hollow fiber by using the 1060-nm OCT system when Au NRI solution is injected into the pig liver sample and the excitation laser is turned on. (b) and (c): M-mode-scan intensity and differential phase profiles, respectively, at the depth indicated by the (red) arrow in part (a). (d) and (e): Corresponding spectra obtained by Fourier-transforming the M-mode-scan signals in parts (b) and (c), respectively.
Figures 8(a) and 8(g) show the conventional OCT and the corresponding PT images, respectively, of the pig liver sample near the distal end of the injection fiber scanned by the 1310-nm OCT system when Au NRIs are injected and the modulated excitation laser is turned on. In Fig. 8(a), one can see the strong signal area around a hole-shaped structure. Here, the (pink) arrow roughly indicates the position of the hole formed by the hollow fiber. The strong signal area corresponds to the distribution region of Au NRIs. With the 1310-nm OCT system, we can monitor the enhanced scattering of the Au NRIs at their LSP resonance. In the corresponding PT image shown in Fig. 8(g), a bright area can be observed in the distribution region of Au NRIs in contrast to the dark background. Figure 8(b) shows the conventional OCT image when the modulated excitation laser is turned off. The clear image of Au NRI distribution is almost identical to that in Fig. 8(a), indicating that the excitation laser is not needed for conventional OCT imaging based on the LSP-enhanced scattering of Au NRIs. The enhanced scattering of Au NRI at its LSP resonance induced by the OCT light source results in the image feature of the Au NRI distribution in OCT scanning. However, when the modulated excitation laser is turned off, the PT signal disappears, as demonstrated by the PT image in Fig. 8(h). For comparison, Figs. 8(c) and 8(i) show the conventional OCT and PT images of the pig liver sample near the distal end of the hollow fiber when PBS is injected into the sample. Here, without Au NRI, neither enhanced scattering nor enhanced absorption can be monitored even though the excitation laser is turned on. It is noted that the hole-shaped structure shown in Figs. 8(a), 8(b), and 8(g) is unclear in Figs. 8(c) and 8(i). This difference can be due to the different OCT scanning locations or the internal structure change only after the injection of Au NRI solution. The clearer hole-shaped structure in Figs. 8(a), 8(b), and 8(g) can also be attributed to the shielding effect of Au NRIs. With LSP resonance-induced enhanced scattering and absorption of Au NRIs, the OCT scanning signal below the Au NRI distribution region becomes weak. Figures 8(d)-8(f) and 8(j)-8(l) show the similar conventional OCT and PT images to Figs. 8(a)-8(c) and 8(g)-8(i), respectively, except that the images are obtained with the scan of the 1060-nm OCT system. The major difference between these two image groups is that the feature of Au NRI distribution is not clearly seen in the conventional OCT images based on the scan of the 1060-nm OCT system, as shown in Figs. 8(d) and 8(e). This is so because the spectral range of the 1060-nm OCT system is away from the LSP resonance peak of Au NRI and hence enhanced scattering is weak. However, the induced heating due to enhanced absorption when the modulated excitation laser is turned on can still be monitored by the scan of the 1060-nm OCT system, as shown in Fig. 8(j).
Fig. 8 (a) and (b) [(d) and (e)]: Conventional OCT images of the pig liver sample with Au NRIs injected near the distal end of the injection hollow fiber scanned by the 1310-nm [1060-nm] OCT system when the excitation laser is turned on and off, respectively. (c) and (f): Conventional OCT images scanned by the 1310-nm and 1060-nm OCT systems, respectively, when PBS is injected into the sample and the excitation laser is turned on. (g)-(l): PT images corresponding to the conventional OCT images in (a)-(f), respectively. The (pink) arrows indicate a hole-like structure formed by Au NRI injection.
Figures 9(a)-9(g) show the conventional OCT images of the 1310-nm system at 0, 20, 40, 60, 80, 100, and 120 min, respectively, after Au NRI solution is injected into the pig liver sample. The modulated excitation laser is turned on for generating the PT effect. Figures 9(h)-9(n) show the corresponding PT images of the conventional OCT images in Figs. 9(a)-9(g), respectively. In the conventional OCT images, the region of strong signal above the dark area indicates the distribution of Au NRIs. This distribution region becomes larger and larger with time demonstrating the diffusion process of Au NRIs in the pig liver sample. The diffusion process can also be observed in the PT images. It is noted that in the PT images, tail-like features can be observed below the Au NRI distribution regions, as indicated by the circled region in Fig. 9(j). Such a tail-like feature is not seen in the corresponding conventional OCT image. It is caused by the effect of accumulated phase during the round-trip of an OCT signal through the Au NRI distribution region. This effect cannot be completely removed when it is strong even though we use the differential phase for evaluating the PT signal. It is also noted that in a certain region of strong signal in the conventional OCT images (at the upper-left corner of the Au NRI distribution region in the conventional OCT images), there is no noticeable signal in the corresponding PT image, as indicated by the circled region in Fig. 9(c). This feature in the conventional OCT images can be due to a denser structure of the pig liver sample formed during the intrusion of the hollow fiber and the injection of Au NRI solution. There is no Au NRI in this region such that no PT signal can be observed here.
Fig. 9 (a)-(g): Conventional OCT images of the 1310-nm system at 0, 20, 40, 60, 80, 100, and 120 min, respectively, after Au NRI solution is injected into the pig liver sample with the excitation laser being turned on. (h)-(n): Corresponding PT images of the conventional OCT images in parts (a)-(g), respectively.
Figures 10(a)-10(g) show the conventional OCT images of the 1060-nm system at 0, 10, 20, 30, 40, 50, and 60 min, respectively, after Au NRI solution is injected into the pig liver sample with the excitation laser being turned on. Figures 10(h)-10(n) show the corresponding PT images of the conventional OCT images in Figs. 10(a)-10(g), respectively. Similar to the image in Fig. 8(d), the Au NRI distribution cannot be clearly seen in the conventional OCT images. The PT images also show clearly the diffusion of Au NRIs. However, they show the Au NRI distribution as vertically striped patterns with tail-like features. The relatively stronger tail-like features in Figs. 10(h)-10(n), when compared with Figs. 9(h)-9(n), are due to the lower SNR ratio in a PT image when we scan the sample with the 1060-nm OCT system, whose spectrum is away from the LSP resonance peak of the used Au NRI. This issue will be further discussed in the next section.
Fig. 10 (a)-(g): Conventional OCT images of the 1060-nm system at 0, 10, 20, 30, 40, 50, and 60 min, respectively, after Au NRI solution is injected into the pig liver sample with the excitation laser being turned on. (h)-(n): Corresponding PT images of the conventional OCT images in parts (a)-(g), respectively.
4. Discussions-phantom study
In a bio-tissue with Au NP distribution in a certain region for inducing LSP resonance, an OCT scan with its spectral range covering the LSP resonance wavelength may show the image features of the similar signal intensities from the region of Au NP distribution and another region of a tissue structure of strong scattering. Such an ambiguity may lead to an error in tracking Au NPs in a tissue sample. The corresponding PT image can help in resolving such an ambiguity. The other issue of great concern is the relationship between the SNR of conventional OCT signal and the PT signal level. With a low SNR of OCT signal, the precise calibration of the phase information for PT imaging is difficult. To shed some light on these two issues, we scan phantom samples A-F with the two OCT systems for comparing their results.
Figures 11(a)-11(f) show the conventional OCT images of phantom samples A-F, respectively, obtained from the scan of the 1310-nm OCT system. Figures 11(g)-11(l) show the PT images corresponding to the conventional OCT images in Figs. 11(a)-11(f), respectively. Then, with the scan of the 1060-nm OCT system, Figs. 12(a)-12(l) show the similar images to those in Figs. 11(a)-11(l), respectively. In Fig. 11(a) for sample A, without TiO2 NP, the LSP induced scattering of Au NRIs results in a clear conventional OCT image of the agar mixture (within the quasi-circular area). As the concentration of TiO2 NP increases in samples B-E, which are used to simulate bio-tissue portions of stronger intrinsic sample scattering, the signal intensity in the conventional OCT image becomes stronger and stronger, as demonstrated in Figs. 11(b)-11(e). Such an increasing trend is also shown in Fig. 13, in which with the right ordinate, the average signal intensity in the agar mixture area of a conventional OCT image is plotted as a function of TiO2 NP concentration. Here, each data point includes six OCT scanning results for forming the error bar. For the OCT signal intensity, because the standard deviation of the multiple-scanning results is very small, the error bars are unclear in Fig. 13. For sample F, as shown in Fig. 11(f), the scattering of TiO2 NPs results in a conventional OCT image with the signal intensity lying between those of samples A and B. The average OCT signal intensity of sample F is shown with the upper horizontal dashed line in Fig. 13. Again, its error bar is negligibly small. In a practical situation of OCT scanning of a bio-tissue containing a local distribution of Au NPs, an OCT image similar to Fig. 11(f) may lead to a wrong judgment that sample F contains Au NPs like samples A and B. In this situation, the PT image can help in differentiating sample F from sample A or B. By comparing Fig. 11(l) with Fig. 11(g) or 11(h), one can see the significant difference between the samples with and without Au NRIs. The average PT signal intensity with the error bar based on six OCT scans as a function of TiO2 NP concentration (samples A-E) is also shown in Fig. 13. The error bar for the average PT signal intensity of sample F is depicted by the left I-shaped notation in Fig. 13. Here, one can see that the PT signal level of sample F is significantly lower than those of samples A-E. Among samples A-E, we can see that when the intrinsic sample scattering is weak, such as that in sample A, the average PT signal intensity is reduced and the error bar becomes wider. As intrinsic sample scattering becomes stronger, like those in samples B-E, the average PT signal intensity reaches an almost constant level. Such a variation trend is attributed to the weak OCT signal and hence the low SNR in sample A such that the calibration for differential phase and hence the evaluation for PT signal become less effective. Therefore, below a certain level of intrinsic sample scattering or OCT signal intensity, stronger intrinsic sample scattering can help us in obtaining a clearer PT image. When intrinsic sample scattering is strong enough, the PT signal intensity is expected to reach the accurate level determined by the excitation laser power [21]. The variation trends of the conventional OCT images in Figs. 12(a)-12(f) and the corresponding PT images in Figs. 12(g)-12(l) obtained from 1060-nm OCT scans are similar to those based on 1310-nm OCT scans. The average OCT signal intensity and average PT signal intensity as functions of TiO2 NP concentration are also shown in Fig. 13. Here, with the enhanced scattering in the case of 1310-nm OCT scan, its average OCT signal intensity is higher than that of 1060-nm OCT scan by an almost constant value in increasing TiO2 NP concentration. In Fig. 13, the average OCT signal intensity and the error bar of average PT signal intensity of sample F with 1060-nm OCT scan are depicted by the lower dashed line and right I-shaped notation, respectively. With the lower OCT signal intensity in scanning sample A with the 1060-nm OCT system, its average PT signal intensity becomes even lower and the corresponding error bar becomes even wider. This trend is consistent with the aforementioned attribution to the lower SNR and hence ineffective differential phase calibration. The significantly larger difference in PT signal intensity between samples A and B with the scan of the 1060-nm OCT system, when compared with that based on the scan of the 1310-nm OCT system, may lead to a misjudgment of Au NP concentration in the sample. The stronger intrinsic sample scattering in sample B, leading to the higher PT signal intensity, may be mistakenly thought to have a higher Au NP concentration if the PT signal is used for evaluating Au NP concentration. In practical application, this mistake can be made if Au NRIs diffuse into a tissue portion of high transparency. Such a mistake can be minimized when the 1310-nm OCT system is used for scanning the sample. With the stronger scattering due to LSP resonance around 1310 nm, the PT signal intensity can be closer to its accurate value, as shown in Fig. 13.
Fig. 11 (a)-(f): Conventional OCT images of phantom samples A-F, respectively, obtained from the scan of the 1310-nm OCT system. (g)-(l): PT images corresponding to the conventional OCT images in (a)-(f), respectively.
Fig. 12 (a)-(f): Conventional OCT images of phantom samples A-F, respectively, obtained from the scan of the 1060-nm OCT system. (g)-(l): PT images corresponding to the conventional OCT images in (a)-(f), respectively.
Fig. 13 Average conventional OCT signal intensity (right ordinate) and average PT signal intensity (left ordinate) in the agar mixture region as functions of TiO2 NP concentration in the samples with Au NRIs (samples A-E) when the 1310-nm and 1060-nm OCT systems are used for scanning the samples. The average OCT and PT signal intensities of sample F are depicted by the two horizontal dashed lines and the two I-shaped notations, respectively.
Besides the factor of intrinsic sample scattering, the SNR of an OCT image is related to the system sensitivity. To understand the effects of OCT system sensitivity on PT signal level, we vary the power levels of the light sources in the two OCT systems for scanning phantom sample E. For the results discussed so far, we control the individual OCT source powers to maintain the SNRs of the two systems the same at ~86 dB. Here, we change the source powers to vary system SNR from this level for scanning phantom sample E. It is noted that the system SNR is defined based on OCT scanning of a mirror surface. Figures 14(a)-14(d) show the conventional OCT images of sample E scanned with the 1310-nm OCT system when the system SNRs are 77, 80, 83, and 86 dB, respectively. Figures 14(e)-14(j) show the conventional OCT images of sample E scanned with the 1060-nm OCT system when the system SNRs are 77, 80, 83, 86, 89, and 92 dB, respectively. Because the available maximum source power at the sample in the 1310-nm OCT system is smaller, when compared with that in the 1060-nm OCT system, no OCT scanning result at a higher SNR from the 1310-nm OCT system is shown. Figures 14(k)-14(t) show the corresponding PT images of the conventional OCT images in Figs. 14(a)-14(j), respectively. Here, one can see that in either group of image with the two OCT systems, both conventional OCT and PT signal intensities increase with system SNR. Figure 15 shows the average OCT signal intensity (right ordinate) and average PT signal intensity (left ordinate) as functions of system SNR with the scans of the two OCT systems. Here, one can see that the average OCT signal intensity increases super-linearly with system SNR in either OCT scan. When the system SNR is larger than 83 dB, the PT signal intensity reaches an almost constant level. At a lower system SNR or a lower OCT signal intensity, the average PT signal intensity drops and its error bar becomes wider. This trend is consistent with that in Fig. 13. When OCT signal intensity is too low, the accurate calibration for PT signal becomes difficult. In comparing the results between the 1060-nm and 1310-nm OCT systems in Fig. 15, one can see that with the enhanced scattering of Au NRIs, the OCT signal intensity of the 1310-nm system is always higher than that of the 1060-nm system in varying system SNR. However, their difference diminishes as system SNR becomes small. Their difference is expected to keep increasing if we can further increase the source power of the 1310-nm system. Although when system SNR is 80 or 83 dB, the PT signal intensity of the 1310-nm system is higher than that of the 1060-nm system, their difference diminishes when system SNR becomes lower or higher. The diminishing trend in the small system SNR limit is due to the diminishing difference of OCT signal intensity. The diminishing trend on the larger system-SNR side is attributed to the saturating behavior of PT signal calibration. When OCT signal intensity reaches a certain level for effective PT signal calibration, a further increase of OCT signal intensity cannot lead to a higher PT signal level because the PT effect is fixed by the excitation power. This variation behavior is consistent with the related conclusion in [21].
Fig. 14 (a)-(d): Conventional OCT images of sample E scanned with the 1310-nm OCT system when the system SNRs are 77, 80, 83, and 86 dB, respectively. (e)-(j): Conventional OCT images of sample E scanned with the 1060-nm OCT system when the system SNRs are 77, 80, 83, 86, 89, and 92 dB, respectively. (k)-(t): Corresponding PT images of the conventional OCT images in (a)-(j), respectively.
Fig. 15 Average OCT signal intensity (right ordinate) and average PT signal intensity (left ordinate) as functions of system SNR of sample E with the scans of the two OCT systems.
The lower PT signal intensities in the sample without TiO2 NP (sample A) with the scans of both OCT systems, as shown in Fig. 13, when compared with other samples of TiO2 NP mixtures (samples B-E), can be improved by increasing the M-mode scan time [16]. Figures 16(a)-16(f) show the PT images of sample A when the M-mode scan times are 25, 50, 75, 100, 125, and 150 ms, respectively, scanned by the 1310-nm OCT system. Figures 16(g)-16(l) show the similar PT images of sample A to those in Figs. 16(a)-16(f), respectively, scanned by the 1060-nm OCT system. Here, one can see that with the scan of either OCT system, the PT image becomes clearer as the M-mode scan time increases. With the left ordinate in Fig. 17, we show the average PT signal and noise intensities based on the scanning results in Figs. 16(a)-16(l) and another five similar scans (for error bar evaluation). Also, with the right ordinate, the PT SNR data, i.e., the ratios of the PT signal intensity over the noise intensity, are shown in Fig. 17. Here, one can see that both PT signal and noise intensities increase with the M-mode scan time. However, the increasing slope of PT signal intensity is larger such that the PT SNR also increases with the M-mode scan time. Therefore, by increasing the M-mode scan time, the PT image quality can be improved. However, the increase of the M-mode scan time leads to the increases of scan and data processing times proportionally.
Fig. 16 (a)-(f): PT images of sample A when the M-mode scan times are 25, 50, 75, 100, 125, and 150 ms, respectively, scanned by the 1310-nm OCT system. (g)-(l): PT images of sample A when the M-mode scan times are 25, 50, 75, 100, 125, and 150 ms, respectively, scanned by the 1060-nm OCT system.
Fig. 17 Average PT signal and noise intensities (with the left ordinate), and PT SNR (with the right ordinate) as functions of M-mode scan time based on the scanning results in Figs. 16(a)-16(l) and other similar measurements.
Two more factors can affect the PT signal level, including the excitation laser power and its modulation frequency. In Figs. 18 and 19, we show the average OCT signal intensity (right ordinate) and the average PT signal intensity (left ordinate) as functions of excitation laser power and modulation frequency, respectively, when the 1060-nm OCT system is used for scanning sample E. Here, one can see that the OCT signal intensity is essentially independent of either excitation laser parameter. The PT signal intensity increases with decreasing modulation frequency and with increasing excitation power. The use of a lower modulation frequency implies a lower speed of PT OCT scanning because a longer M-mode scan time is required to obtain a reasonable number of modulation periods for accurate phase analysis when the modulation frequency is low. On the other hand, the use of a lower excitation power can minimize the thermal damage to a sample.
Fig. 18 Average OCT signal intensity (right ordinate) and the average PT signal intensity (left ordinate) as functions of excitation laser power when the 1060-nm OCT system is used for scanning sample E.
Fig. 19 Average OCT signal intensity (right ordinate) and the average PT signal intensity (left ordinate) as functions of modulation frequency when the 1060-nm OCT system is used for scanning sample E.
5. Conclusions
In summary, we have demonstrated the conventional OCT images based on enhanced scattering and the PT images based on enhanced absorption of the LSP resonance of Au NRIs in a bio-tissue sample with the scans of an OCT system (1310-nm system) with its spectral range covering the LSP resonance peak wavelength and another OCT system (1060-nm system) with its spectral range away from the LSP resonance peak wavelength. A PT image was formed by evaluating the modulation frequency response of an excitation laser with its wavelength close to the LSP resonance peak at 1305 nm of the Au NRI solution. By using the 1310-nm OCT system, the Au NRI distribution in the bio-tissue sample could be observed in both conventional OCT and PT images. However, by using the 1060-nm OCT system, the Au NRI distribution could be observed only in the PT image. The diffusion process of Au NRIs in the bio-tissue sample could be traced with the scan of either OCT system. Based on phantom experiments, it was shown that the PT image could help us in resolving the ambiguity of a conventional OCT image between the enhanced scattering of Au NRIs and the strong intrinsic sample scattering of a tissue structure in the 1310-nm OCT scanning. Also, under the condition of weak intrinsic sample scattering, particularly in the scan with the 1060-nm system, the PT signal could be lower than a saturating level, which was determined by the excitation power. By increasing OCT system SNR or M-mode scan time, the PT signal level could be enhanced.
Acknowledgments
This research was supported by National Science Council (grants NSC 102-2218-E-002-012-MY3 and NSC 102-2221-E-002-199) and National Health Research Institute (grant NHRI-EX102-10043EI), Taiwan. Also, it is sponsored by the NTU Excellent Research Project (102R890951 and 102R890952) of National Taiwan University.
References and links
1. K. L. Kelly, E. Coronado, L. L. Zhao, and G. C. Schatz, “The optical properties of metal nanoparticles: The influence of size, shape, and dielectric environment,” J. Phys. Chem. B 107(3), 668–677 (2003). [CrossRef]
2. S. A. Maier, Plasmonics: Fundamentals and Applications (Springer, 2007).
3. E. V. Zagaynova, M. V. Shirmanova, M. Y. Kirillin, B. N. Khlebtsov, A. G. Orlova, I. V. Balalaeva, M. A. Sirotkina, M. L. Bugrova, P. D. Agrba, and V. A. Kamensky, “Contrasting properties of gold nanoparticles for optical coherence tomography: phantom, in vivo studies and Monte Carlo simulation,” Phys. Med. Biol. 53(18), 4995–5009 (2008). [CrossRef] [PubMed]
4. C. S. Kim, P. Wilder-Smith, Y. C. Ahn, L. H. L. Liaw, Z. Chen, and Y. J. Kwon, “Enhanced detection of early-stage oral cancer in vivo by optical coherence tomography using multimodal delivery of gold nanoparticles,” J. Biomed. Opt. 14(3), 034008 (2009). [CrossRef] [PubMed]
5. M. Kirillin, M. Shirmanova, M. Sirotkina, M. Bugrova, B. Khlebtsov, and E. Zagaynova, “Contrasting properties of gold nanoshells and titanium dioxide nanoparticles for optical coherence tomography imaging of skin: Monte Carlo simulations and in vivo study,” J. Biomed. Opt. 14(2), 021017 (2009). [CrossRef] [PubMed]
6. A. M. Gobin, M. H. Lee, N. J. Halas, W. D. James, R. A. Drezek, and J. L. West, “Near-infrared resonant nanoshells for combined optical imaging and photothermal cancer therapy,” Nano Lett. 7(7), 1929–1934 (2007). [CrossRef] [PubMed]
7. X. Huang and M. A. El-Sayed, “Gold nanoparticles: Optical properties and implementations in cancer diagnosis and photothermal therapy,” J. Advert. Res. 1(1), 13–28 (2010). [CrossRef]
8. B. Jang, J. Y. Park, C. H. Tung, I. H. Kim, and Y. Choi, “Gold nanorod-photosensitizer complex for near-infrared fluorescence imaging and photodynamic/photothermal therapy in vivo,” ACS Nano 5(2), 1086–1094 (2011). [CrossRef] [PubMed]
9. X. Huang, I. H. El-Sayed, W. Qian, and M. A. El-Sayed, “Cancer cell imaging and photothermal therapy in the near-infrared region by using gold nanorods,” J. Am. Chem. Soc. 128(6), 2115–2120 (2006). [CrossRef] [PubMed]
10. S. Lal, S. E. Clare, and N. J. Halas, “Nanoshell-enabled photothermal cancer therapy: Impending clinical impact,” Acc. Chem. Res. 41(12), 1842–1851 (2008). [CrossRef] [PubMed]
11. S. Jelveh and D. B. Chithrani, “Gold nanostructures as a platform for combinational therapy in future cancer therapeutics,” Cancers (Basel) 3(4), 1081–1110 (2011). [CrossRef] [PubMed]
12. E. S. Day, J. G. Morton, and J. L. West, “Nanoparticles for thermal cancer therapy,” J. Biomech. Eng. 131(7), 074001 (2009). [CrossRef] [PubMed]
13. X. Huang, P. K. Jain, I. H. El-Sayed, and M. A. El-Sayed, “Determination of the minimum temperature required for selective photothermal destruction of cancer cells with the use of immunotargeted gold nanoparticles,” Photochem. Photobiol. 82(2), 412–417 (2006). [CrossRef] [PubMed]
14. R. J. Bernardi, A. R. Lowery, P. A. Thompson, S. M. Blaney, and J. L. West, “Immunonanoshells for targeted photothermal ablation in medulloblastoma and glioma: An in vitro evaluation using human cell lines,” J. Neurooncol. 86(2), 165–172 (2008). [CrossRef] [PubMed]
15. J. Chen, F. Saeki, B. J. Wiley, H. Cang, M. J. Cobb, Z. Y. Li, L. Au, H. Zhang, M. B. Kimmey, X. Li, and Y. Xia, “Gold nanocages: Bioconjugation and their potential use as optical imaging contrast agents,” Nano Lett. 5(3), 473–477 (2005). [CrossRef] [PubMed]
16. D. C. Adler, S. W. Huang, R. Huber, and J. G. Fujimoto, “Photothermal detection of gold nanoparticles using phase-sensitive optical coherence tomography,” Opt. Express 16(7), 4376–4393 (2008). [CrossRef] [PubMed]
17. C. Zhou, T. H. Tsai, D. C. Adler, H. C. Lee, D. W. Cohen, A. Mondelblatt, Y. Wang, J. L. Connolly, and J. G. Fujimoto, “Photothermal optical coherence tomography in ex vivo human breast tissues using gold nanoshells,” Opt. Lett. 35(5), 700–702 (2010). [CrossRef] [PubMed]
18. M. C. Skala, M. J. Crow, A. Wax, and J. A. Izatt, “Photothermal optical coherence tomography of epidermal growth factor receptor in live cells using immunotargeted gold nanospheres,” Nano Lett. 8(10), 3461–3467 (2008). [CrossRef] [PubMed]
19. H. Cang, T. Sun, Z. Y. Li, J. Chen, B. J. Wiley, Y. Xia, and X. Li, “Gold nanocages as contrast agents for spectroscopic optical coherence tomography,” Opt. Lett. 30(22), 3048–3050 (2005). [CrossRef] [PubMed]
20. C. Pache, N. L. Bocchio, A. Bouwens, M. Villiger, C. Berclaz, J. Goulley, M. I. Gibson, C. Santschi, and T. Lasser, “Fast three-dimensional imaging of gold nanoparticles in living cells with photothermal optical lock-in optical coherence microscopy,” Opt. Express 20(19), 21385–21399 (2012). [CrossRef] [PubMed]
21. J. M. Tucker-Schwartz, T. A. Meyer, C. A. Patil, C. L. Duvall, and M. C. Skala, “In vivo photothermal optical coherence tomography of gold nanorod contrast agents,” Biomed. Opt. Express 3(11), 2881–2895 (2012). [CrossRef] [PubMed]
22. M. C. Daniel and D. Astruc, “Gold nanoparticles: assembly, supramolecular chemistry, quantum-size-related properties, and applications toward biology, catalysis, and nanotechnology,” Chem. Rev. 104(1), 293–346 (2004). [CrossRef] [PubMed]
23. M. Eghtedari, A. V. Liopo, J. A. Copland, A. A. Oraevsky, and M. Motamedi, “Engineering of hetero-functional gold nanorods for the in vivo molecular targeting of breast cancer cells,” Nano Lett. 9(1), 287–291 (2009). [CrossRef] [PubMed]
24. W. I. Choi, J. Y. Kim, C. Kang, C. C. Byeon, Y. H. Kim, and G. Tae, “Tumor regression in vivo by photothermal therapy based on gold-nanorod-loaded, functional nanocarriers,” ACS Nano 5(3), 1995–2003 (2011). [CrossRef] [PubMed]
25. G. von Maltzahn, J. H. Park, A. Agrawal, N. K. Bandaru, S. K. Das, M. J. Sailor, and S. N. Bhatia, “Computationally guided photothermal tumor therapy using long-circulating gold nanorod antennas,” Cancer Res. 69(9), 3892–3900 (2009). [CrossRef] [PubMed]
26. A. M. Gobin, J. J. Moon, and J. L. West, “EphrinAl-targeted nanoshells for photothermal ablation of prostate cancer cells,” Internal. J. Nanomed. 3, 351–358 (2008).
27. L. B. Carpin, L. R. Bickford, G. Agollah, T. K. Yu, R. Schiff, Y. Li, and R. A. Drezek, “Immunoconjugated gold nanoshell-mediated photothermal ablation of trastuzumab-resistant breast cancer cells,” Breast Cancer Res. Treat. 125(1), 27–34 (2011). [CrossRef] [PubMed]
28. A. M. Schwartzberg, T. Y. Olson, C. E. Talley, and J. Z. Zhang, “Synthesis, characterization, and tunable optical properties of hollow gold nanospheres,” J. Phys. Chem. B 110(40), 19935–19944 (2006). [CrossRef] [PubMed]
29. J. Z. Zhang, “Biomedical applications of shape-controlled plasmonic nanostructures: A case study of hollow gold nanospheres for photothermal ablation therapy of cancer,” J. Phys. Chem. Lett. 1(4), 686–695 (2010). [CrossRef]
30. J. Chen, D. Wang, J. Xi, L. Au, A. Siekkinen, A. Warsen, Z. Y. Li, H. Zhang, Y. Xia, and X. Li, “Immuno gold nanocages with tailored optical properties for targeted photothermal destruction of cancer cells,” Nano Lett. 7(5), 1318–1322 (2007). [CrossRef] [PubMed]
31. J. Chen, C. Glaus, R. Laforest, Q. Zhang, M. Yang, M. Gidding, M. J. Welch, and Y. Xia, “Gold nanocages as photothermal transducers for cancer treatment,” Small 6(7), 811–817 (2010). [CrossRef] [PubMed]
32. E. M. Larsson, J. Alegret, M. Käll, and D. S. Sutherland, “Sensing characteristics of NIR localized surface plasmon resonances in gold nanorings for application as ultrasensitive biosensors,” Nano Lett. 7(5), 1256–1263 (2007). [CrossRef] [PubMed]
33. F. Hao, E. M. Larsson, T. A. Ali, D. S. Sutherland, and P. Nordlander, “Shedding light on dark plasmons in gold nanorings,” Chem. Phys. Lett. 458(4-6), 262–266 (2008). [CrossRef]
34. J. Aizpurua, P. Hanarp, D. S. Sutherland, M. Käll, G. W. Bryant, and F. J. García de Abajo, “Optical properties of gold nanorings,” Phys. Rev. Lett. 90(5), 057401 (2003). [CrossRef] [PubMed]
35. H. Y. Tseng, C. K. Lee, S. Y. Wu, T. T. Chi, K. M. Yang, J. Y. Wang, Y. W. Kiang, C. C. Yang, M. T. Tsai, Y. C. Wu, H. Y. E. Chou, and C. P. Chiang, “Au nanorings for enhancing absorption and backscattering monitored with optical coherence tomography,” Nanotechnology 21(29), 295102 (2010). [CrossRef] [PubMed]
36. C. K. Lee, H. Y. Tseng, C. Y. Lee, S. Y. Wu, T. T. Chi, K. M. Yang, H. Y. E. Chou, M. T. Tsai, J. Y. Wang, Y. W. Kiang, C. P. Chiang, and C. C. Yang, “Characterizing the localized surface plasmon resonance behaviors of Au nanorings and tracking their diffusion in bio-tissue with optical coherence tomography,” Biomed. Opt. Express 1(4), 1060–1073 (2010). [CrossRef] [PubMed]
37. H. Y. Tseng, W. F. Chen, C. K. Chu, W. Y. Chang, Y. Kuo, Y. W. Kiang, and C. C. Yang, “On-substrate fabrication of a bio-conjugated Au nanoring solution for photothermal therapy application,” Nanotechnology 24(6), 065102 (2013). [CrossRef] [PubMed]
38. S. Y. Wu, W. M. Chang, H. Y. Tseng, C. K. Lee, T. T. Chi, J. Y. Wang, Y. W. Kiang, and C. C. Yang, “Geometry for maximizing localized surface plasmon resonance of Au nanorings with random orientations,” Plasmonics 6(3), 547–555 (2011). [CrossRef]
39. T. T. Chi, C. K. Lee, C. T. Wu, C. C. Yang, M. T. Tsai, and C. P. Chiang, “Motion-insensitive optical coherence tomography based micro-angiography,” Opt. Express 19(27), 26117–26131 (2011). [CrossRef] [PubMed]
40. R. K. Wang and A. L. Nuttall, “Phase-sensitive optical coherence tomography imaging of the tissue motion within the organ of corti at a subnanometer scale: a preliminary study,” J. Biomed. Opt. 15(5), 056005 (2010). [CrossRef] [PubMed]
