Open Access
Add to BibSonomy
Abstract
Retinal laser photocoagulation is used to treat several ophthalmic diseases. However, it is associated with damage to surrounding healthy tissue. Local tissue alteration during coagulation laser illumination was measured using phase-resolved optical coherence tomography (OCT) M-mode scan as a change in the local optical path length (LOPL). A metric that represents global net tissue alteration was defined using the LOPL change. The visibility of a laser lesion was assessed by three-dimensional OCT volume measurement. Multiple logistic regression analysis was performed to investigate the association between the introduced metric and the laser lesion visibility. The metric was found to be a statistically significant predictor of the laser lesion visibility independent to laser condition. The proposed method based on an LOPL change is thus promising for retinal photocoagulation monitoring.
© 2017 Optical Society of America
1. Introduction
Retinal laser photocoagulation is used to treat several ophthalmic diseases, such as extrafoveal choroidal neovascularization, central serous chorioretinopathy, and diabetic retinopathy [1–3]. The photothermal effect is used to heat tissue and coagulate abnormal tissues or harm tissues to solve hypoxia. Hence, the treatment is inherently associated with damage to normal surrounding tissue. The monitoring and/or prediction of the treatment is important in terms of avoiding unnecessary tissue malfunction.
There are difficulties in predicting or assessing photocoagulation. First, each subject possesses variable pigmentation, which is a main absorber that produces heat in retinal photocoagulation. Second, the light power on the retina is unknown owing to the vignetting of light and/or opacity of the anterior chamber. The standard method is to ophthalmoscopically observe the whitening of tissue. The monitoring of tissue denaturation has been demonstrated by detecting a change in the reflectance of tissue [4–6]. These methods are based on changes in optical properties due to tissue alteration [4]. The optoacoustic method monitors a change in the temperature of tissue [7]. Meanwhile, optical coherence tomography (OCT) has been used to detect a change in internal tissue due to photocoagulation [8–11]. Localized tissue monitoring will be suitable for the high-sensitivity detection of tissue damage.
The real-time monitoring of tissue deformation and the prediction of laser lesion formation perhaps are essentials in feedback for laser treatment. A high frequency of feedback will be needed to avoid long latency. The duration of retinal laser photocoagulation is about several tens to hundreds of milliseconds [12,13]. Hence, data acquisition and signal processing might need to be around or less than millisecond order to achieve a sufficiently high frequency of feedback.
In this study, we present the detection of tissue denaturation during laser photocoagulation by an M-mode OCT-based measurement of the local tissue alterations in the axial strain and/or refractive index; i.e., local optical path length (LOPL) change [14]. Depth-resolved tissue alteration monitoring with high temporal resolution is available. The LOPL change method is immune to the axial motion artifact and has been applied to in vivo skin photothermal imaging [14]. We introduce a metric based on the LOPL change that might reveal the starting time point of tissue denaturation. The association between the metric and laser lesion formation is evaluated in ex vivo porcine eye experiments.
2. Methods
In this study, the LOPL change is measured during the coagulation laser illumination on ex vivo porcine eyes by employing M-mode OCT. The LOPL change is then compared with the visibility of the laser lesion.
2.1. Subjects
Seven ex vivo porcine eyes were used in the photocoagulation experiments. The porcine eyes were obtained from a local abattoir and experiments were completed within 24 hours after enucleation. Saline solution was put on the cornea and a contact lens was applied to reduce aberrations and to suppress dehydration.
2.2. Combined OCT – coagulation laser system
The combined OCT and coagulation laser system of this study has been outlined previously in detail [11]. Spectral-domain OCT in the 1-μm wavelength band was combined with a 532-nm laser (GYC-1000, Nidek Co., Ltd., Japan). The coagulation laser and the OCT probe beam were combined using a dichroic mirror. The galvanometric scanner installed between the laser output and the dichroic mirror was used as a high-speed shutter to synchronize data acquisition with the coagulation laser illumination.
The OCT scanned each axial profile at 91,911 lines/s. The axial resolution was 4.0 μm (full width at half-maximum) in the tissue. The lateral 1/e2 spot diameter of the Gaussian probe beam on the retina is expected to be around 16 μm. The probe power of the OCT was 1.8 mW on the cornea. The size of the coagulation laser on the retina exceeded at least 80 μm in diameter.
2.3. Protocol
The coagulation laser illumination was applied with several combinations of power and duration. For each eye, each laser illumination condition was applied at a different location on the retina. The laser power was set to 400, 300, 200, 100, or 50 mW on the cornea while the duration was set to 200, 100, 50, or 25 ms.
Before and after each laser illumination, a three-dimensional OCT volume scan was taken as shown in the time chart of Fig. 1. An M-mode OCT scan was taken during the laser illumination. The M-mode scan started 1 ms before the laser illumination began.
Fig. 1 Timing chart of laser illumination and M-mode OCT measurement. The start time of coagulation laser illumination ts is 1 ms after the start of M-mode OCT scan.
2.4. Measurement of the LOPL change and quantification of the tissue alteration
We define the LOPL change rate as
where w(t, l) is the optical path length (OPL) change at time t and OPL l. The LOPL change rate [Eq. (1)] is computed from the mixed second-order partial derivative of phase-resolved OCT signals as where λc is the central wavelength of the light source and ϕ is the phase of the OCT signal and related to the OPL change as . ti and lj respectively denote time and the OPL for the i-th axial scan and j-th pixel. Δt is the operator of the temporal differentiation with a time separation of mδt while Δl is the operator of the axial differentiation with an axial separation of dδl. d and m are respectively pixel separations along axial and temporal directions. δl = 3.2 μm is the OPL per pixel (i.e., the axial distance per pixel in air). δt = 10.9 μs (91,911 lines/s) is the period of the axial profile measurement. Before the axial differentiation and after the temporal differentiation, complex axial averaging was applied as same as Ref. [14]. This technique provides the instantaneous OPL change at the local point. This study uses m = 20 and d = 10, which correspond to approximately 218 μs and 32 μm, respectively. A window size of 5 pixels is used for the complex axial averaging. These parameters are determined empirically. Masking with the threshold level of 8 dB higher than the noise level is applied [Ref. [14]]. When the axial separation dδl is sufficiently larger than the temporal thickness change of local tissue, the LOPL change rate [Eq. (1)] can be approximated as [14] where ∊z is axial strain and n is the refractive index. The LOPL change rate r is an approximation of the sum of the axial strain change and refractive index change rate.To represent the tissue alteration, we introduce a metric based on the LOPL change rate. First, the magnitude of the calculated change rate [Eq. (2)] is integrated along the axial direction as
This value is interpreted as the net global axial alteration rate. The noise offset is calibrated by the averaged LOPL change rate above the threshold level before the start of laser illumination. ts is the start time of coagulation laser illumination. The integrated net LOPL change rate [Eq. (4)] is then accumulated over time to calculate the total alteration as We use this M(ti) as a metric representing the net global tissue alteration appearing along the OPL.2.5. Laser lesion visibility assessment
Coagulation due to laser illumination on the retina causes alterations at photoreceptor cells, mainly at the outer nuclear layer (ONL) [15]. En face slices at the ONL were extracted from three-dimensional OCT volumes before and after coagulation laser illumination. Slices immediately above the retinal pigment epithelium (RPE) were selected. These en face images were compared before and after the coagulation laser illumination. A case is classified as “laser lesion visible” if there is a hyper-scattering spot after laser illumination.
2.6. Statistical analysis
Multiple logistic regression analysis is performed to evaluate the association between the introduced metric M and laser lesion formation. The outcome y is the visibility of a laser lesion (i.e., whether the lesion is visible (y = 1) or not (y = 0)), which is evaluated as described in Section 2.5. To exclude the effect of the coagulation laser condition, coagulation laser power P, duration τ, and the interaction between the laser power and duration P × τ, which corresponds to the laser energy, are included as nuisance variables. The regression model is expressed as
where Pr(y = 1) is the probability that the visible laser lesion is occurred (y = 1). β0, βM, βP, βτ and βE are the regression coefficients of the constant (intercept), metric, power, duration and energy, respectively.A Wald test on the regression coefficients is performed. A p-value less than or equal to 0.05 is regarded as statistically significant. To evaluate the contribution of the metric M, a likelihood ratio (LR) test is performed and a difference of Akaike’s information criterion (AIC) is calculated between two models with the metric M [Eq. (6)] and without M.
3. Results
3.1. OCT measurement results
3.1.1. M-scan results
Examples of M-mode OCT and LOPL change rate images of one porcine eye (Eye07) are shown in Figs. 2 and 3. OCT intensity [Fig. 2] exhibits the alteration of the speckle pattern and the elevation of the neural retina. The temporal LOPL change rate at the beginning (orange rectangles) and around the end (blue rectangles) of the coagulation laser illumination are shown in Fig. 3. The time-resolved LOPL change rate [Fig. 3(a)–3(c)] shows that there were three phases of conditions at the beginning of the laser illumination. No response is observed from the time point of turning on the coagulation laser (0 ms) to a few milliseconds later (I in Fig. 3). This is perhaps because of the delayed response of the coagulation laser shutter. The galvanometric shutter delay was measured to be about 3 ms. This value is in good agreement with the duration of this no response phase. The LOPL change rate then gradually increased at the RPE depth (II). This might be due to thermal expansion of the tissue. After several milliseconds, the LOPL change rate increased and exhibit bidirectional values around the RPE (III). After this time point, a hyper-scattering lesion gradually appeared at the ONL. Hence, this large and abrupt LOPL change may be due to tissue denaturation. As clearly observed in Fig. 2(a)–2(c), there was a high-scattering lesion above the RPE in OCT cross sections after coagulation laser illumination. In the case of 100-mW illumination, there was only a small LOPL change rate immediately after the laser was turned on[Fig. 3(d)]. No clear hyper-scattering lesion was observed after the illumination [Fig. 2(d)]. The LOPL change rate of only the front part of the duration of laser illumination is shown. After that, the magnitude of LOPL change rate decreases with time (not shown).
Fig. 2 OCT M-mode scan images of Eye07 with coagulation laser duration of 200 ms and power of (a) 400 mW, (b) 300 mW, (c) 200 mW, and (d) 100 mW. Yellow ticks in OCT cross sections denote the location of the M-mode scan. The orange and blue boxes are the location of M-mode LOPL change rate shown in Fig. 3. White arrows indicate the stop time of the coagulation laser.
Fig. 3 M-mode LOPL change rate corresponding to Fig. 2. Yellow ticks in OCT cross sections denote the location of the M-mode scan.
After turning off the laser illumination, rather negative LOPL change rates were observed at the RPE. Probably, the contraction of tissue is predominantly caused by a decrease in temperature. The time lags from the laser stop timing (200 ms) are perhaps due to the laser shutter delay as mentioned above.
In cases with high laser power [Figs. 3(a)–3(c)], periodic patterns in LOPL change rate appeared at the retina. We currently consider three possible explanations for the LOPL change rate; i.e., retinal absorption, heat propagation from the RPE, and mechanical strain. Blood is the main absorber in the inner retina, however, its contribution to photothermal effect is negligible as is outlined in Section 4.5. The measurement of the LOPL change [Eq. (2)] without a change in the refractive index is equivalent to a measurement of axial strain [14,16] [See Eq. (3)]. Perhaps a large deformation due to thermal expansion and/or tissue denaturation applies pressure to the retina and thus generates mechanical strain. Thus, thermal change due to heat propagation from the RPE and/or the mechanical strain due to a large deformation around the RPE may be reasons of the LOPL change in the retina. Maybe the propagation of heat and strain interfered and caused the periodic pattern of LOPL change.
3.1.2. Metric M
The obtained metric M(t) is shown for all subjects and all coagulation laser illumination configurations in Fig. 4. The metric curve is displayed with dashed line in cases that there is no clearly visible laser lesion in the OCT cross section [Section 3.1.3]. The metric M does not change with time in some cases; e.g., the case of Eye06 at power of 300 mW and duration of 50 ms [Fig. 4(f)]. No laser lesion was observed after laser illumination in these cases.
Fig. 4 Obtained metric M(t) for seven porcine eyes and 20 laser configurations. Solid lines indicate cases in which there is a visible laser lesion, and dashed lines indicate cases in which there is no visible laser lesion.
3.1.3. Visible lesion in OCT
Several cases of Eye01 with and without a visible laser lesion are shown in Fig. 5. The first case is laser irradiation at 400 mW for 200 ms. A hyper-scattering spot appears in the en face slice [Fig. 5(b)]. In addition, tissue alteration is visible in the cross-sectional image [Fig. 5(d)] (M(200 ms) = 14.7 μm). The second case of illumination at 200 mW for 100 ms is shown in Fig. 5(e)–(h). In this case, there is no apparent laser lesion in the cross-sectional image [Fig. 5(h)]; however, a hyper-scattering spot is clearly visible in the en face slice after laser illumination [Fig. 5(f)] (M(100 ms) = 7.81 μm). In the third case of illumination at 50 mW for 50 ms [Fig. 5(i)–(l)], no laser lesion is observed in either the en face slice [Fig. 5(j)] or cross section [Fig. 5(l)] (M(50 ms) = 0.0889 μm). The first two cases are classified to have a visible laser lesion.
Fig. 5 Visualization of laser lesions with OCT volumetric imaging (Eye01): (a–d) duration of 200 ms and power of 400 mW; (e–h) duration of 100 ms and power of 200 mW; (i–l) duration of 50 ms and power of 50 mW; (a). (e), (i): en face OCT slices at the ONL before laser illumination; (b), (f), (j): en face OCT slices at the ONL after laser illumination; (c), (g), (h): cross-sectional OCT images before laser illumination; (d), (h), (l): cross-sectional OCT images after laser illumination.
3.2. Statistical analysis of the metric M
Multiple logistic regression analysis is performed to investigate the association between the metric and laser lesion visibility. The maximum value of the metric M(t) is used as an alternative to the metric M at the end of laser illumination. The maximum value during the laser illumination is used because some non-response cases decrease the metric [e.g., Figs. 4(o, p, s, t)]. The maximum value of the metric M(t) during laser illumination and the power and duration of laser illumination are used as explanatory variables.
All subjects and all laser conditions are used for multiple logistic regression analysis. Table 1 gives the results of multiple logistic regression analysis. A Wald test on the coefficients of each parameter is performed.
Table 1. Results of multiple logistic regression analysis. *P < 0.05, **P < 0.01, ***P < 0.001.
The coefficient of the metric M is statistically significant in terms of rejecting the null hypothesis (i.e., a coefficient of a variable is zero). The metric M thus explains the variation in laser lesion visibility when the coagulation laser condition is fixed.
The intercept is the offset of log-odds. When all explanatory variables are zero, the probability of laser lesion formation Pr(y = 1|P = 0, τ = 0, M = 0) = logit−1[β0]. The large negative value is thus reasonable, and the probability of laser lesion formation is estimated as being very low (logit−1[−5.94] = 0.0026) when there is no laser illumination.
Note that we build the model [Eq. (6)] to investigate the association between the metric M and laser lesion visibility. Multiple variables of the coagulation laser condition are included in the model. Hence, no statistical significance in Wald tests on each laser condition parameter does not mean these parameters are not associated with the laser lesion visibility.
The LR test between the models with and without the metric M suggests that the contribution of the metric M in the regression model [Eq. (6)] is statistically significant (p ≪ 0.001). The AIC increases by 38.3 when excluding the metric M. These results indicate that the metric M will be useful in predicting the formation of laser lesion.
4. Discussions
4.1. Difficulty of prediction using the laser illumination condition
As described in Section 3.2, the power and duration of the coagulation laser are not good explanatory variables with which to explain the formation of a laser lesion. This may be due to the variable size of the laser spot. One reason for this variation is chromatic aberration. Chromatic aberration will provide a large focal shift between the OCT probe and coagulation laser beams because the wavelengths of these beams are largely separated. The focus was adjusted for OCT imaging, and the spot size of the coagulation laser might be affected by the chromatic aberration and variable dioptric power of the individual porcine eye.
4.2. Formation of a laser lesion
In the presented experiments, almost half of the laser illumination conditions generated visible laser lesions. However, according to a numerical simulation of temperature (which is not described in this paper), the temperature increases by more than 30 degrees and the Arrhenius damage integral [17] exceeds 1 for all laser conditions. The numerical simulation was conducted in the same way as in the previous report (with the porcine explant model in Ref. 18) and the consistency of the simulation results was confirmed with the same simulation condition. The discrepancy between experimental results and numerical simulation may be due to the laser spot variation as described above and the transparency of anterior tissue. The postmortem eye may have large attenuation in the cornea and lens due to opacity. Light attenuation due to opacity was not accounted for in the numerical simulation and it may reduce the power of the coagulation laser on the retina.
4.3. Comparison with previous methods
Employing titration may avoid variation of the inter-eye condition [19, 20]; first find a laser power to form a visible laser lesion, then adjust the laser condition for treatment based on the titrated power. However, this method ignores variation of the intra-eye condition, such as the pigmentation distribution. The metric introduced in this study may instantaneously and directly represent tissue alteration under coagulation laser illumination. It already accounts for variations in inter- and intra-eye conditions.
The scattering of tissue is affected by tissue denaturation. This characteristic is used to monitor the laser lesion. A similar approach may be developed by directly observing OCT intensity information. However, such an approach depends not only on denaturation but also on several other factors, such as attenuation due to vignetting and/or opacity of the tissue. The en face projection of the OCT intensity before and after coagulation laser illumination is shown in Fig. 6. The OCT intensity projection is analogous to fundus reflectance. However, the difference is not obvious. A more quantitative approach, such as a measurement of attenuation, might provide a more reliable measure of the change in scattering.
Fig. 6 En face projections of three-dimensional volumes presented in Fig. 5: (a–c) en face projections before laser coagulation; (d–f) en face projections after laser coagulation. The coagulation laser was illuminated at (d) 400 mW for 200 ms, (e) 200 mW for 100 ms, and (f) 50 mW for 50 ms. Scale bars indicate 500 μm.
A local expansion method based on OCT was previously demonstrated [8]. It assumes that the axial shift in OCT is proportional to displacement and the total displacement along time is calculated. The discrepancy between the OPL change and displacement thus accumulates. Changes in the temperature and volume (density) affect the refractive index [21, 22]. Protein denaturation alters the molecular structure and affects optical properties, including the refractive index [23,24]. Additionally, it is difficult to predict the dominant factor of a change in the LOPL (i.e., the tissue expansion or refractive index change) because it depends on the mechanical property of the tissue [25]. The LOPL change rate thus exhibits both positive and negative values as shown in Fig. 3. The method based on total displacement integrates both positive and negative LOPL changes. These bidirectional LOPL changes cancel each other out. Therefore, the introduced metric that integrates the magnitude of the instantaneous LOPL change rate might be stable in evaluating tissue alteration.
4.4. Evaluation of tissue damage
In the case of the laser lesion, outer segments are stretched, inner segments are condensed, and nuclei are pyknotic [15]. The abnormality thus appears at the ONL. Some experimental results in the present study seem to show a “humping reaction” according to the coagulation laser spot size [26]. The OCT examination of laser lesions (Section 2.5) may therefore work well. In some cases, no apparent hyper-scattering spot was observed in the cross section [Fig. 5(h)], while a spot was observed in the en face slice [Fig. 5(f)]. This may be because of RPE thickening due to vacuolation [26]. It is expected that these different reactions affect the treatment. The classification and management of laser lesion will be a future topic of research.
4.5. Blood absorption
As shown in Fig. 3, the LOPL change rate exhibits non-zero values even in the sensory retina in cases of high-power illumination. Tissue deformation around retinal blood vessels due to blood absorption in the retina is one of possible reasons for this change.
Blood in the sensory retina could obstruct the coagulation laser for the posterior tissue because of attenuation. When there is a large retinal vessel, there will be a photothermal effect and tissue denaturation at and around the retinal vessel, not at the RPE, as shown in Fig. 7. Even at a location without a large retinal vessel, retinal capillaries may act as an absorber and generate heat. At a wavelength of 532 nm, the absorption coefficient of hemoglobin is, however, one-tenth of that of melanin [27,28]. Furthermore, the exposure area of the retinal capillary layer is smaller than that of the RPE. It is difficult to consider tissue denaturation due to laser absorption at retinal capillaries.
Fig. 7 Laser photocoagulation at a large retinal vessel with a power of 400 mW and a duration of 200 ms. Cross-sectional OCT image (a) before and (b) after coagulation laser illumination. (c) LOPL change rate during laser illumination at the location indicated by orange lines in (a) and (b). Scale bars indicate 500 μm.
Hence, the LOPL change measured in the current experiment, which avoided large retinal vessels, was possibly due to heat generated at the RPE. The LOPL change at the sensory retina may be due to other reasons.
4.6. Limitations
We introduced a metric M based on the measurement of the LOPL change. However, there is a limitation in the current study. The LOPL change rate was summed along the depth to provide a global metric, which removes spatial localization. In future work, a depth-resolved metric will be useful in detecting depth-dependent tissue alteration. However, a depth-resolved measurement may decrease the signal-to-noise ratio. A possible solution to the reduction in the signal-to-noise ratio is to combine measurements of the LOPL change and other local tissue properties, such as attenuation [29].
The M-mode scan limits the measurement in the one-dimensional axial profile. Although a two- or three-dimensional measurement is interesting, there will be a long latency between the measurement and feedback. The ultrahigh-speed measurement system and low latent processing may extend the capability of the current method.
The experiment has been done for ex vivo porcine eyes. For clinical application, eye motion when imaging in vivo will be one of the main problems. As shown theoretically and experimentally, the LOPL change rate measurement is not affected by axial motion [14]. Lateral motion could cause a displacement between the beam spot locations of coagulation laser and OCT probe beam due to chromatic aberration. However, the displacement may not cause an apparent change on the LOPL change rate because of the relatively large beam spot size of the coagulation laser. Hence, eye motion should not be a significant problem.
5. Conclusion
In this study, we demonstrated the measurement of the LOPL change during the retinal laser photocoagulation of ex vivo porcine eyes. A metric that represents tissue alteration was introduced and its relationship with the visibility of a laser lesion was analyzed. Statistical analysis showed that the introduced metric is a good predictor for the formation of a visible laser lesion. Retinal photocoagulation monitoring based on the measurement of the LOPL change is thus promising.
Funding
Japan Society for the Promotion of Science (JSPS) KAKENHI 15K13371 (all authors); Ministry of Education, Culture, Sports, Science, and Technology (MEXT) Program for Building Innovation Ecosystem (all authors).
Acknowledgments
Research in this publication was supported in part by Nidek Co., Ltd.
Disclosures
SM: Topcon Corporation (F), Tomey Corporation (F); YY: Topcon Corporation (F), Tomey Corporation (F).
References and links
1. Macular Photocoagulation Study Group, “Argon laser photocoagulation for neovascular maculopathy: Five-year results from randomized clinical trials,” Arch. Ophthalmol. 109(8), 1109–1114 (1991). [CrossRef] [PubMed]
2. D. M. Robertson and D. Ilstrup, “Direct, indirect, and sham laser photocoagulation in the management of central serous chorioretinopathy,” Am. J. Ophthalmol. 95(4), 457–466 (1983). [CrossRef] [PubMed]
3. Early Treatment Diabetic Retinopathy Study Research Group, “Early photocoagulation for diabetic retinopathy: ETDRS report number 9,” Ophthalmology 98(5, Supplement), 766–785 (1991). [CrossRef]
4. W. Weinberg, R. Birngruber, and B. Lorenz, “The change in light reflection of the retina during therapeutic laser photocoagulation,” IEEE J. Quantum Electron. 20(12), 1481–1489 (1984). [CrossRef]
5. K. P. Pflibsen, F. C. Delori, O. Pomerantzeff, and M. M. Pankratov, “Fundus reflectometry for photocoagulation dosimetry,” Appl. Opt. 28(6), 1084–1096 (1989). [CrossRef] [PubMed]
6. M. R. Jerath, R. Chundru, S. F. Barrett, H. Rylander III, and A. Welch, “Reflectance feedback control of photocoagulation in vivo,” Arch. Ophthalmol. 111(4), 531–534 (1993). [CrossRef] [PubMed]
7. K. Schlott, S. Koinzer, L. Ptaszynski, M. Bever, A. Baade, J. Roider, R. Birngruber, and R. Brinkmann, “Automatic temperature controlled retinal photocoagulation,” J. Biomed. Opt. 17(6), 0612231 (2012). [CrossRef]
8. H. H. Müller, L. Ptaszynski, K. Schlott, C. Debbeler, M. Bever, S. Koinzer, R. Birngruber, R. Brinkmann, and G. Hüttmann, “Imaging thermal expansion and retinal tissue changes during photocoagulation by high speed OCT,” Biomed. Opt. Express 3(5), 1025–1046 (2012). [CrossRef] [PubMed]
9. K. Kurokawa, S. Makita, Y.-J. Hong, and Y. Yasuno, “Two-dimensional micro-displacement measurement for laser coagulation using optical coherence tomography,” Biomed. Opt. Express 6(1), 170–190 (2015). [CrossRef] [PubMed]
10. K. Kurokawa, S. Makita, Y.-J. Hong, and Y. Yasuno, “In-plane and out-of-plane tissue micro-displacement measurement by correlation coefficients of optical coherence tomography,” Opt. Lett. 40(9), 2153–2156 (2015). [CrossRef] [PubMed]
11. K. Kurokawa, S. Makita, and Y. Yasuno, “Investigation of thermal effects of photocoagulation on retinal tissue using fine-motion-sensitive dynamic optical coherence tomography,” PLOS ONE 11(6), e0156761 (2016). [CrossRef] [PubMed]
12. A. Jain, M. S. Blumenkranz, Y. Paulus, M. W. Wiltberger, D. E. Andersen, P. Huie, and D. Palanker, “Effect of pulse duration on size and character of the lesion in retinal photocoagulation,” Arch. Ophthalmol. 126(1), 78–85 (2008). [CrossRef] [PubMed]
13. C. Sanghvi, R. McLauchlan, C. Delgado, L. Young, S. J. Charles, G. Marcellino, and P. E. Stanga, “Initial experience with the Pascal photocoagulator: a pilot study of 75 procedures,” Br. J. Ophthalmol. 92(8), 1061–1064 (2008). [CrossRef] [PubMed]
14. S. Makita and Y. Yasuno, “In vivo photothermal optical coherence tomography for non-invasive imaging of endogenous absorption agents,” Biomed. Opt. Express 6(5), 1707–1725 (2015). [CrossRef]
15. R. Birngruber, V. P. Gabel, and F. Hillenkamp, “Experimental studies of laser thermal retinal injury,” Health Phys. 44(5), 519–531 (1983). [CrossRef] [PubMed]
16. B. F. Kennedy, S. H. Koh, R. A. McLaughlin, K. M. Kennedy, P. R. T. Munro, and D. D. Sampson, “Strain estimation in phase-sensitive optical coherence elastography,” Biomed. Opt. Express 3(8), 1865–1879 (2012). [CrossRef] [PubMed]
17. R. Birngruber, F. Hillenkamp, and V. P. Gabel, “Theoretical investigations of laser thermal retinal injury,” Health Phys. 48(6), 781–796 (1985). [CrossRef] [PubMed]
18. C. Sramek, Y. Paulus, H. Nomoto, P. Huie, J. Brown, and D. Palanker, “Dynamics of retinal photocoagulation and rupture,” J. Biomed. Opt. 14(3), 034007 (2009). [CrossRef] [PubMed]
19. D. Lavinsky, C. Sramek, J. Wang, P. Huie, R. Dalal, Y. Mandel, and D. Palanker, “SUBVISIBLE RETINAL LASER THERAPY: Titration Algorithm and Tissue Response,” Retina 34(1), 87–97 (2014). [CrossRef]
20. D. Lavinsky, J. Wang, P. Huie, R. Dalal, S. J. Lee, D. Y. Lee, and D. Palanker, “Nondamaging retinal laser therapy:: rationale and applications to the macula,” Invest. Ophthalmol. Vis. Sci. 57(6), 2488–2500 (2016). [CrossRef] [PubMed]
21. I. Thormählen, J. Straub, and U. Grigull, “Refractive index of water and its dependence on wavelength, temperature, and density,” J. Phys. Chem. Ref. Data 14(4), 933–945 (1985). [CrossRef]
22. A. H. Harvey, J. S. Gallagher, and J. M. H. L. Sengers, “Revised formulation for the refractive index of water and steam as a function of wavelength, temperature and density,” J. Phys. Chem. Ref. Data 27(4), 761–774 (1998). [CrossRef]
23. F. Kreith, ed., The CRC Handbook of Thermal Engineering (Springer, 2000).
24. R. Barer and S. Tkaczyk, “Refractive index of concentrated protein solutions,” Nature 173(4409), 821–822 (1954). [CrossRef] [PubMed]
25. J. Kim, J. Oh, and T. E. Milner, “Measurement of optical path length change following pulsed laser irradiation using differential phase optical coherence tomography,” J. Biomed. Opt. 11(4), 041122 (2006). [CrossRef] [PubMed]
26. J. Marshall, “Thermal and Mechanical Mechanisms in Laser Damage to the Retina,” Invest. Ophthalmol. 9(2), 97–115 (1970). [PubMed]
27. S. L. Jacques, “Optical Absorption of Melanin,” (1998). {http://omlc.org/spectra/melanin/index.html}.
28. S. Prahl, “Optical Absorption of Hemoglobin,” (1999). {http://omlc.org/spectra/hemoglobin/index.html}.
29. K. A. Vermeer, J. Mo, J. J. A. Weda, H. G. Lemij, and J. F. de Boer, “Depth-resolved model-based reconstruction of attenuation coefficients in optical coherence tomography,” Biomed. Opt. Express 5(1), 322–337 (2014). [CrossRef] [PubMed]
