Clinical monitoring of blood oxygen saturation (sO2) is traditionally performed using optical techniques, such as pulse oximetry and diffuse reflectance spectroscopy (DRS), which lack spatial resolution. Photoacoustic imaging (PAI) is a rapidly developing biomedical imaging technique that is superior to previous techniques in that it combines optical excitation and acoustic detection, providing a map of chromophore distribution in the tissue. Hitherto, PAI has primarily been used in preclinical studies, and only a few studies have been performed in patients. Its ability to measure sO2 with spatial resolution during local vasoconstriction after adrenaline injection has not yet been investigated. Using PAI and spectral unmixing we characterize the heterogeneous change in sO2 after injecting a local anesthetic containing adrenaline into the dermis on the forearm of seven healthy subjects. In comparison to results obtained using DRS, we highlight contrasting results obtained between the two methods arising due to the so-called ‘window effect’ caused by a reduced blood flow in the superficial vascular plexus. The results demonstrate the importance of spatially resolving sO2 and the ability of PAI to assess the tissue composition in different layers of the skin.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
The most commonly employed non-invasive techniques for monitoring oxygen saturation (sO2) in the clinical setting are pulse-oximetry and diffuse near-infrared reflectance spectroscopy, both relying on optical signals to determine the difference in concentration of oxygenated (HbO2) and deoxygenated (HbR) hemoglobin through the skin surface [1,2]. Unfortunately, these methods only provide an average measure of sO2 in a specific volume, and lack inherent spatial resolution.
Other techniques can be used to measure sO2 with spatial resolution, but these all have drawbacks. Diffuse optical tomography (DOT) and blood-oxygenation-level-dependent (BOLD) contrast imaging are non-invasive techniques that can provide spatially resolved information on sO2 in, for example, the brain [3,4]. DOT, which uses near-infrared radiation, provides topographic reconstruction, but is not capable of providing layer-specific anatomical information . BOLD-contrast imaging in functional magnetic resonance imaging can provide some degree of spatial resolution, [3,6] but with reduced temporal resolution . There is thus a need for a non-invasive method of monitoring sO2 with high spatial resolution to discriminate local changes in sO2. This is particularly important in medical conditions that are not systemic, for example, monitoring local changes in peripheral artery disease, tumors, flaps, and reconstructive surgery [7–10].
Photoacoustic imaging (PAI) is currently one of the most promising non-invasive biomedical imaging techniques, and unlike existing techniques, it provides label-free molecular imaging with high spatial resolution and an extended measurement depth. It combines the benefits of spectroscopic-based specificity with ultrasound detection imaging depth. Photons transmitted from a laser source are absorbed in the tissue where a thermo-elastic response is generated, giving rise to acoustic waves that can be measured with an ultrasound probe. The detection of sound to determine the light absorption affords PAI the feature of high spatial resolution. The absorption spectra obtained with PAI can be analyzed by applying spectral unmixing, allowing the extraction of molecular information, including relative concentrations of HbO2 and HbR [11–14].
PAI has so far mostly been used preclinically, [15,16] but has major potential for clinical applications in the future [7,17–19]. It is the first non-invasive technique capable of providing spatially resolved information on sO2 in tissue with the resolution necessary to differentiate between different skin tissue layers . PAI has been used in vascular imaging of human feet and the results were compared to those from traditional duplex ultrasonography . While duplex ultrasonography measures perfusion, PAI provides information on sO2 with high spatial resolution, and can thus provide a more detailed view of smaller vasculature . PAI has also been used successfully for pre-operative vascular mapping in human thigh flap surgery,  and to measure oxygen saturation in patients with skin burns, as a useful indicator of the degree of tissue damage and for monitoring .
The applicability of PAI to monitor local vasoconstriction with high spatial resolution has not been explored. The aim of this study was therefore to monitor sO2 in the human forearm with PAI where a local anesthetic containing adrenaline was injected superficially in the dermis of the skin, resulting in vasoconstriction of the superficial vascular plexus. Since the deep vascular plexus remains unaffected, this creates a depth-dependent reduction in sO2. The effects in the different skin layers were mapped using PAI, and compared to the results from a commercial oxygen monitor, based on diffuse reflectance spectroscopy (DRS), which yields an average signal from the entire tissue volume examined.
2. Materials and methods
The experimental protocol for this study was approved by the Ethics committee at Lund University, Sweden. The research adhered to the tenets of the Declaration of Helsinki as amended in 2008. All the subjects were thoroughly informed about the study, and the voluntary nature of participation, and gave their informed written consent.
Inclusion criteria were healthy subjects with a skin type II-III on the Fitzpatrick scale,  to reduce variability in the results as a consequence of melanin. The exclusion criterion was the presence of any advanced medical condition that could contra-indicate the injection of local anesthetics containing adrenaline, such as ischemic heart disease, heart arrhythmia, lung disease, asthma, or previous adverse reaction to local anesthetics. Subjects who were identified as being smokers or having microangiopathy, i.e. resulting from diseases such as diabetes, kidney or cardiovascular disease, were also excluded. The subjects were asked to refrain from caffeine-containing drinks and food for at least two hours, as well as strenuous exercise for 24 hours, prior to the measurements in order to stabilize their peripheral circulation. The subjects rested lying down for 10 minutes before the start of the experiment. Their blood pressure and pulse were measured before and after the procedure to assure that they were normotensive and had a stable pulse. Studies were performed in a room where the temperature was maintained around 22 °C. Seven healthy adult volunteers, three men and four women, were included in the study. The median age of the subjects was 34 years (range 20–75 years).
2.3 Photoacoustic imaging
PAI was performed using a Vevo LAZR-X instrument (VisualSonics Inc., Toronto, ON, Canada) where a laser excitation source generates a pulse of light every 50 ms that is guided to the skin surface via optical fibers. As these photons penetrate the skin and become absorbed at various depths, governed by the optical properties of the tissue, a thermo-elastic response is generated. The acoustic signal is measured with an ultrasound transducer operating at a central frequency of 30 MHz with a bandwidth of 20–46 MHz, which provides the spatial contrast. The spectral contrast is obtained by repeating measurements at different excitation wavelengths between 680 nm and 970 nm in steps of 10 nm, indirectly relating the signal to the light absorption, with 30 spectral components at every point in a vertical cross-section. The axial and lateral spatial resolutions are 50 µm and 110 µm, respectively. The maximum measurement depth of the system is 20 mm, determined by the 40 MHz ultrasonic transducer and ultimately light fluence and strength of absorption. In this study, the depth range is 5 mm since this is sufficient to contrast the PAI signals from the superficial and deep vascular plexus. Extraction of sO2 with PAI has previously been performed using as few as two excitation wavelengths, and up to 59, as discussed by Merdasa et al. .
2.4 Diffuse reflectance spectroscopy
A commercial oxygen monitor system (moorVMS-OXY, Moor Instruments, Devon, UK), based on DRS, was used to obtain relative concentrations of HbR and HbO2, from which total hemoglobin (HbT) and sO2 are determined. Photons in the spectral range between 500 and 650 nm are directed into the tissue via an optical fiber in an excitation probe placed on the skin surface. The light scattered from the skin is collected by another optical fiber placed 1 mm laterally from the excitation probe, and detected with a photodiode. The wavelength and separation between the light source and the detector governs the depth of the measurements; greater separation resulting in a greater depth . The DRS system measures a signal integrated over a depth of ∼1 mm into the tissue, and operates at a frequency of 1 kHz. Since the signal is an average from the entire volume that is measured, no vertical spatial information can be obtained from the measurements .
2.5 Experimental procedure
Both the subject’s arms were stabilized using vacuum positioning pillows (Germa Protec, Germa AB, Kristianstad, Sweden), and the subjects were specifically asked not to move during the procedure. The examiner paid special attention to this issue. The PAI probe was mounted on an adjustable arm such that it could easily be moved, but also remain rigidly fixed in space for an extended period of time. The probe was adjusted so that it just touched the skin of one arm, and care was taken to ensure there was no compression of the tissue. The DRS probe was secured on the skin of the other arm, directly above the injection site, using a double-sided adhesive O-ring. Measurements were performed prior to the injection in order to establish a baseline signal in unaffected tissue. An injection of 0.5 ml lidocaine (20 mg/ml) + adrenaline (12.5 µg/ml) (Xylocaine Dental Adrenaline, Dentsply Ltd., York, PA, USA), (henceforth denoted “adrenaline” in the text and figures) was then given in each arm. The solution was preheated to 37°C before injection. The injection was administered into the dermis, on the volar side of the forearm (avoiding hair and veins) to cause vasoconstriction in the superficial vascular plexus, avoiding the deep vascular plexus, thereby creating a depth-dependent sO2 profile. A volume of 0.5 ml was chosen since it represents a clinically significant amount, without resulting in tissue compression. The injections were performed by the same surgeon, as uniformly as possible in one arm, followed by the contralateral arm. The order of examining the right and left arm and monitoring with PAI or DRS was randomized, and all subjects were monitored with both techniques. Immediately after the injection, monitoring with PAI or DRS resumed for 6–10 min, during which the vasoconstrictive effect of adrenaline unfolded.
2.6 Data analysis
The commercial DRS instrument automatically provides HbO2, HbR and HbT in arbitrary units, as well as sO2 in percent. The PA data require analysis to extract these, and was performed as follows. Each pixel in a photoacoustic (PA) image contains a spectrum. The data are prepared for analysis by spatially averaging a region of the PA image for each spectral component separately, after which a single spectrum is constructed by repeating this procedure for the same region in each excitation-wavelength-specific photoacoustic image. Spectral unmixing was performed as described previously,  using absorption spectra of five endmember spectra, for HbO2, HbR, melanin, fat, and water . In order to minimize bias toward a preconceived notion of the expected tissue constituents at a particular depth, all endmember spectra were included in the spectral unmixing analysis of all the extracted PA spectra. Inclusion of too many endmembers may cause incorrect fitting of the data with endmember spectra that should not be present. A non-negative matrix factorization (NNMF) approach is therefore used in the linear spectral unmixing procedure, which forces the contribution from all endmembers to be positive (or zero). This limits the incorrect inclusion of endmember spectra that should not be considered.
Due to the wavelength dependent attenuation of the fluence with increasing depth (spectral coloring),  no quantification of the molecular components from the spectral unmixing was possible. Instead, the fractional abundance of each endmember was compared at each depth, from which HbT was obtained by adding the contributions from HbO2 and HbR, while sO2 was obtained by dividing the contribution from HbO2 by that from HbT. Since sO2 relies on the relative abundance of HbO2 and HbR extracted from a single measured PA spectrum, it is less affected by the decreasing fluence with depth into tissue, allowing comparison of sO2 at different depths.
The data given (Figs. 2–4) are median values, averaged from a ROI on each of the seven subjects’ forearms (solid lines in the figures) with 95% confidence intervals (shaded area in the figures). The size of ROI was chosen so as to maximize the area of each layer (see Fig. 1), but also leaving some margin for error. A representative example of a PA image including spectral unmixing (Fig. 1) is also given in the results. The spectral unmixing analysis was performed using a NNMF algorithm (‘lsqnonneg.m’) in MATLAB (The Mathworks Inc. South Natick, MA, USA).
3.1 Chromophores in the three skin layers
Spectral unmixing of the PA data, using the five endmember spectra representing absorption by HbO2, HbR, melanin, fat, and water, showed variations in the molecular composition between the three skin layers. In the epidermis, the PA spectrum is dominated by absorption by melanin, with only a small contribution from HbR. Further into the dermis, the PA spectrum has a distinctly different appearance; the HbO2 component is prominent, which represents the anatomical location of the superficial vascular plexus, with a contribution from fat. In the hypodermis, the PA spectrum is best fitted by absorption spectra from primarily HbO2, which is the location of the well-perfused deep vascular plexus. A representative example is shown in Fig. 1.
3.2 Results from PAI monitoring of the effects of adrenaline
PAI monitoring after the injection of adrenaline showed a reduction in HbT over time in the dermis at the location of the superficial vascular plexus where the adrenaline was injected. Furthermore, the amount of HbO2 decreased simultaneously with an increase in HbR, indicating a gradual reduction in sO2, as expected following adrenaline-induced vasoconstriction (Figs. 2(a)-(d)). Deeper in the hypodermis the HbT signal remained stable, indicating that sO2 was unaffected in this layer. This demonstrates the ability of PAI to monitor sO2 with spatial resolution.
The fractional abundance of melanin was analyzed in the epidermis in order to validate the spectral unmixing results. The melanin signal was stable over time, in contrast to the substantially larger changes in HbO2, demonstrating that PAI monitoring was not sensitive to measurement artefacts, such as fluctuations in the laser or motion of the subject (Fig. 2(e)).
3.3 Results of DRS measurements of the effects of adrenaline
DRS, unlike PAI, lacks vertical spatial resolution, and instead measures the ‘global’ change in sO2 throughout a volume determined by the wavelength and separation between the source and detector fibers, as described in the Methods section. DRS monitoring showed a paradoxical increase in HbT (dominated by an increase in HbO2) throughout the course of the measurements (Fig. 3), which is inconsistent with the vasoconstrictive effect of adrenaline. Analyzing the DRS graphs in detail revealed an initial decrease in HbO2 and sO2 during the first 40 s, after which they increased markedly throughout the remaining monitoring period. This may be attributed to a ‘window effect’, as discussed below.
3.4 Rate of change in sO2 measured with DRS and PAI
Analysis of the PAI data showed a steady decrease in sO2 in the dermis, reflecting the expected effect of constriction of the superficial vascular plexus by adrenaline. A decrease in sO2 was also observed during the initial 40 s when measured with DRS, although the signal subsequently increased markedly. However, the rate at which sO2 changed was similar with the two techniques. The sO2 data obtained in the dermis with PAI was fitted with a single exponential function, giving a time constant (τPAI) of 123 s. Applying the same fit to the sO2 values determined with DRS, from the point at which the curve started to increase (around 40 s), yielded a similar time constant (τDRS) of 109 s. This indicates that PAI and DRS reflect the same biological phenomenon, i.e. the effect of adrenaline on sO2. The discrepancy between the observed sO2 trends of the two techniques is likely the result of the so-called ‘window effect’, where reduced blood flow in the superficial plexus causes blanching of the skin, which alters its optical properties . Light then penetrates deeper into the tissue, and the DRS measurements reflect sO2 in the deeper layers of the tissue, as illustrated in Fig. 4.
3.5 Evolution of spatially resolved sO2 after adrenaline injection
We took advantage of the spatial resolution of PAI and calculated sO2 on a pixel-by-pixel basis. Figure 5 shows how sO2 develops with time after adrenaline injection in the different skin layers. A considerable reduction in sO2 is observed in the dermis, while that in the hypodermis remains largely unchanged, consistent with the results presented in Fig. 4.
4.1 Multi-wavelength PAI and spectral unmixing
The results of the present study show the ability of PAI to monitor sO2 with spatial resolution, detecting local changes in oxygenation that occur after the injection of adrenaline in the skin. Spectral unmixing employed over a broad spectral range (680–970 nm) enabled identification of the chromophores in each specific skin layer, which facilitated an understanding of the effects of adrenaline injection on human tissue.
These results are supported by a few previous studies on oxygenation in humans using PAI, although these considered global changes in sO2 as a result of occluding an extremity, rather than the absorption of multiple tissue chromophores using spectral unmixing. sO2 in the forearm muscle has been assessed with PAI using only a few excitation wavelengths to determine relative HbO2 (850 nm), HbR (750 nm), and HbT (800 nm) concentrations obtained from excitation wavelengths and the isosbestic point of HbT (800 nm), finding that the results were comparable to those obtained using near-infrared spectroscopy [28,29]. The results showed an overall steady level of HbO2, although arterial occlusion was expected to lead to a decrease in intramuscular oxygenation. Unprovoked subcutaneous tissue in the human arm has also been studied using PAI at nine wavelengths, where good-to-excellent interclass correlation coefficients and good reproducibility for HbT and sO2 using spectral unmixing were found .
In this study, 30 wavelengths were used over a broad spectral range (680–970 nm), which provided more detailed absorption features than in previous studies. This enabled spectral unmixing involving multiple endmember spectra with distinct spectral features, which provided information on the chromophores present at different tissue depths. Although only HbO2 and HbR were used to determine sO2, we emphasize the importance of considering as many light-absorbing tissue components as possible when employing spectral unmixing. Not only are more accurate measures of HbO2 and HbR obtained from a measured volume but, as shown in Fig. 1, other endmembers such as fat and water aid in identifying the tissue layers where sO2 is monitored beyond the capability of ultrasound alone. Correct identification of the various layers of tissue aids in determining which vascular plexus is affected by adrenaline.
Spectral unmixing of the PA data showed fractional abundances of HbO2, HbR, melanin, fat, and water in good agreement with the anatomical location of these substances in the skin. The PA spectrum from the epidermis is dominated by absorption by melanin, with only a small contribution from HbR. This is consistent with the epidermis being avascular and the location of melanocytes at the basal membrane . The PA spectrum from the dermis had a distinctly different shape; the HbO2 component being more prominent at the location of the superficial vascular plexus, together with a significant contribution from fat. Absorption by water is minimal in this spectral range, although this should still be included in the spectral unmixing analysis. The absorption by HbO2 is even more prominent in the hypodermis, presumably due to the well-perfused deep vascular plexus.
4.2 Spectral coloring
A problem associated with PAI is spectral coloring . As the incident light at the surface propagates deeper into the tissue, the probability of absorption increases. Thus, not only is the light attenuated at all excitation wavelengths with increasing tissue depth, but the attenuation may depend on the excitation wavelength as a result of the absorbing chromophores in the layers above. It is therefore unreliable to compare the absolute abundances of endmembers (chromophores) extracted at different depths. However, this problem is reduced when observing sO2 since it is a relative measure based on the fractional abundances extracted from the same PA spectrum, and have therefore been equally attenuated with depth. In other words, if the relative intensity between all the wavelengths in the entire fluence spectrum remain the same, there should be no change in the calculated value of sO2. However, should the shape of the spectrum change, this may affect sO2. Analysis methods have been suggested to correct for spectral coloring,  although it has also been shown that this had no practical significance when assessing sO2 in vivo in humans .
4.3 Adrenaline-induced vasoconstriction and the window effect
Injection of adrenaline induced vasoconstriction and a decrease in sO2 that could be monitored with spatial resolution using PAI. The reduction in sO2 was observed particularly in the dermis, while little effect was seen in the hypodermis, reflecting the fact that the adrenaline injection was given at the level of the superficial vascular plexus, and did not affect the deeper vascular plexus. However, DRS monitoring of the response to adrenaline showed a paradoxical increase in HbO2 and HbT, which is inconsistent with the expected reduction in blood perfusion and oxygenation of adrenaline . However, the rate of change of sO2 obtained from a single exponential fit to the two curves was similar (τPAI = 123 s and τDRS = 109 s). This indicates that the two techniques monitor the same biological phenomenon, i.e. adrenaline-induced vasoconstriction. The time of the vasoconstrictive effect of adrenaline has been presented the time it takes for the reducing blood flow to reach a steady state, however, the time constant used here may constitute a more robust measure allowing reliable comparisons between different studies. The time to maximum vasoconstrictive effect of adrenaline was found to be around 2 minutes with both methods, which is in agreement with previous reports of 2.6–10.0 min [34–37].
The difference between the two techniques may be attributed to the window effect. Adrenaline is injected into the dermis causing vasoconstriction of the superficial vascular plexus, evacuating hemoglobin from the superficial layers of the skin. Hemoglobin is a strong light absorber in the spectral range 500–650 nm,  and its reduction will allow light to penetrate deeper into the vascular plexus where there is a larger blood volume . In other words, the upper tissue layers become successively more transparent to light providing a ‘window’ into the deeper layers of the tissue. Blanching of the skin could be seen by the naked eye in the region around the adrenaline injection site. The window effect has been described previously, [38,39] where inconsistencies were seen when employing laser speckle contrast imaging to measure hypoperfusion in the skin . The shape of the sO2 curve given by DRS can therefore be explained as follows. The initial 40 s during which a decrease is seen represents the actual reduction in sO2, as confirmed by PAI, and the increase seen thereafter is the result of the increased transparency of the upper layers of tissue allowing the transmission of light into the deeper vascular plexus, which is less affected by adrenaline.
In a previous study, we used a DRS-based instrument with an extended wavelength range (EW-DRS) to study the effects of adrenaline primarily on perfusion, but also on sO2 . The EW-DRS device is similar to the DRS instrument used in the present study, but monitors other wavelengths and has a larger physical separation between the two fibers that deliver the excitation light to the skin and detect the diffuse reflectance, resulting in a greater measurement depth with the EW-DRS device . With respect to the window effect, the extended detection volume obtained with EW-DRS over which the signal is averaged may minimize the deviating behavior of the sO2 trend observed here. Nonetheless, it can be concluded that purely spectroscopic techniques based on so-called ‘light-in, light-out’ methods, such as DRS, are appropriate when a uniform change in HbT is expected in the entire tissue, for example, during arterial occlusion, [2,20] traumatic shock or compartment syndrome [41,42]. In cases where local changes in hemoglobin levels and sO2 can be expected in different tissue layers, techniques providing depth-resolved information, such as PAI, are needed.
In conclusion, this is the first study in humans to have demonstrated the ability of PAI to produce a spatially resolved map of sO2 as a result of adrenaline injection. Applying spectral unmixing over a broad and detailed spectral range provides an anatomically accurate picture of the architecture of the skin in which tissue layers are contrasted. Monitoring the reduction in sO2 in the anatomical structure enabled identification of the vascular plexus affected by adrenaline injection. We found that the decrease in sO2 was restricted to the superficial vascular plexus in the dermis, while blood flow remained unaffected in deeper skin layers, consistent with the location of adrenaline administration. In contrast, DRS monitoring showed a paradoxical increase in sO2 upon the administration of adrenaline, which we suggest is the result of the so-called window effect in which vasoconstriction causes blanching of the skin, changing its optical properties. The results of the presented study confirm and extend the applicability of PAI for use in spatially resolved non-invasive monitoring of localized changes in sO2, which would be of value, for example, in treating burn wounds and monitoring the perfusion in skin flaps. Being able to determine the rate of oxygen consumption in different tissue layers is a potential field for future studies and will be of clinical relevance to patients with diabetes.
Swedish Government Grant for Clinical Research (ALF); Skånes universitetssjukhus; Region Kronoberg; Skåne County Council's Research and Development Foundation; Lund University Grant for Research Infrastructure; Swedish Cancer Foundation; Stiftelsen Kronprinsessan Margaretas Arbetsnämnd för Synskadade; Friends of the Visually Impaired Association in the county of Gävleborg (KMA); Lund Laser Center Research Grant; IngaBritt och Arne Lundbergs Forskningsstiftelse; Ögonfonden; Cronqvist Foundation; Swedish Medical Association; Lund University grant for Research Infrastructure; Stiftelsen för Synskadade i f.d. Malmöhus län.
The authors declare no conflicts of interest.
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
1. M. W. Wukitsch, M. T. Petterson, D. R. Tobler, and J. A. Pologe, “Pulse oximetry: analysis of theory, technology, and practice,” J Clin Monitor Comput 4(4), 290–301 (1988). [CrossRef]
2. H. Liu, M. Kohl-Bareis, and X. Huang, “Design of a tissue oxygenation monitor and verification on human skin,” Opt. InfoBase Conf. Pap.2011.
3. M. D. Wheelock, J. P. Culver, and A. T. Eggebrecht, “High-density diffuse optical tomography for imaging human brain function, Rev. Sci. Instrum. 90, 051101 (2019) [CrossRef] .
4. E. Amaro and G. J. Barker, “Study design in fMRI: Basic principles,” Brain Cogn. 60(3), 220–232 (2006). [CrossRef]
5. Y. Hoshi and Y. Yamada, “Overview of diffuse optical tomography and its clinical applications,” J. Biomed. Opt. 21(9), 091312 (2016). [CrossRef]
6. A. Cyrous, B. O’Neal, and W. D. Freeman, “New approaches to bedside monitoring in stroke,” Expert Rev. Neurother. 12(8), 915–928 (2012). [CrossRef]
7. J. R. Eisenbrey, M. Stanczak, F. Forsberg, F. A. Mendoza-Ballesteros, and A. Lyshchik, “Photoacoustic oxygenation quantification in patients with Raynaud's: First-in-human results,” Ultrasound Med. Biol. 44(10), 2081–2088 (2018). [CrossRef]
8. M. Petri, I. Stoffels, K. Griewank, J. Jose, P. Engels, A. Schulz, H. Pötzschke, P. Jansen, D. Schadendorf, J. Dissemond, and J. Klode, “Oxygenation status in chronic leg ulcer after topical hemoglobin application may act as a surrogate marker to find the best treatment strategy and to avoid ineffective conservative long-term therapy,” Mol Imaging Biol 20(1), 124–130 (2018). [CrossRef]
9. A. A. Oraevsky, B. Clingman, J. Zalev, A. T. Stavros, W. T. Yang, and J. R. Parikh, “Clinical optoacoustic imaging combined with ultrasound for coregistered functional and anatomical mapping of breast tumors,” Photoacoustics 12, 30–45 (2018). [CrossRef]
10. K. Tenland, J. V. Berggren, C. D. Ansson, J. Hult, U. Dahlstrand, S. Lindstedt, R. Sheikh, and M. Malmsjö, “Blood perfusion in rotational full-thickness lower eyelid flaps measured by laser speckle contrast imaging,” Ophthal. Plast. Reconstr. Surg. 36(2), 148–151 (2020). [CrossRef]
11. J. Xia, A. Danielli, Y. Liu, L. Wang, K. Maslov, and L. V. Wang, “Calibration-free quantification of absolute oxygen saturation based on the dynamics of photoacoustic signals,” Opt. Lett. 38(15), 2800 (2013). [CrossRef]
12. A. Taruttis, A. C. Timmermans, P. C. Wouters, M. Kacprowicz, G. M. Van Dam, and V. Ntziachristos, “Optoacoustic imaging of human vasculature: feasibility by using a handheld probe,” Radiology 281(1), 256–263 (2016). [CrossRef]
13. V. Grasso, J. Holthof, and J. Jose, “An automatic unmixing approach to detect tissue chromophores from multispectral photoacoustic imaging,” Sensors 20, 3235 (2020). [CrossRef]
14. B. Cox, J. G. Laufer, and ArridgeS. R. and P. C. Beard, “Quantitative spectroscopic photoacoustic imaging: a review,” J. Biomed. Opt. 17(6), 061202 (2012). [CrossRef]
15. C. Liu, Y. Liang, and L. Wang, “Single-shot photoacoustic microscopy of hemoglobin concentration, oxygen saturation, and blood flow in sub-microseconds,” Photoacoustics 17, 100156 (2020). [CrossRef]
16. L. J. Rich and M. Seshadri, “Photoacoustic imaging of vascular hemodynamics: validation with blood oxygenation level-dependent MR imaging,” Radiology 275(1), 110–118 (2015). [CrossRef]
17. F. Knieling, C. Neufert, A. Hartmann, J. Claussen, A. Urich, C. Egger, M. Vetter, S. Fischer, L. Pfiefer, A. Hagel, C. Kielisch, R. Görtz, D. Wildner, M. Engel, J. Röther, W. Uter, J. Siebler, R. Atreya, W. Rasher, D. Strobel, N. Markus, and M. Waldner, “Multispectral optoacoustic tomography for assessment of Crohn's disease activity,” N. Engl. J. Med. 376(13), 1292–1294 (2017). [CrossRef]
18. A. Becker, M. Masthoff, J. Claussen, S. J. Ford, W. Roll, M. Burg, P. J. Barth, W. Heindel, M. Schäfers, M. Eisenblätter, and M. Wildgruber, “Multispectral optoacoustic tomography of the human breast: characterisation of healthy tissue and malignant lesions using a hybrid ultrasound-optoacoustic approach,” Eur Radiol 28(2), 602–609 (2018). [CrossRef]
19. G. Diot, S. Metz, A. Noske, E. Liapis, B. Schroeder, S. V. Ovsepian, R. Meier, E. Rummeny, and V. Ntziachristos, “Multispectral optoacoustic tomography (MSOT) of human breast cancer,” Clin. Cancer Res. 23(22), 6912–6922 (2017). [CrossRef]
20. A. Merdasa, J. Bunke, M. Naumovska, J. Albinsson, T. Erlöv, M. Cinthio, N. Reistad, R. Sheikh, and M. Malmsjö, “Photoacoustic imaging of the spatial distribution of oxygen saturation in an ischemia-reperfusion model in humans,” Biomed. Opt. Express 12(4), 2484 (2021). [CrossRef]
21. I. Tsuge, S. Saito, G. Yamamoto, H. Sekiguchi, A. Yoshikawa, Y. Matsumoto, S. Suzuki, and M. Toi, “Preoperative vascular mapping for anterolateral thigh flap surgeries: A clinical trial of photoacoustic tomography imaging,” Microsurgery 40(3), 324–330 (2020). [CrossRef]
22. Z. Wu, F. Duan, J. Zhang, S. Li, H. Ma, and L. Nie, “In vivo dual-scale photoacoustic surveillance and assessment of burn healing,” Biomed. Opt. Express 10(7), 3425 (2019). [CrossRef]
23. T. B. Fitzpatrick, “The validity and practicality of sun-reactive skin types I through VI,” Arch. Dermatol. 124(6), 869 (1988). [CrossRef]
24. R. Hennessy, W. Goth, M. Sharma, M. K. Markey, and J. W. Tunnell, “Effect of probe geometry and optical properties on the sampling depth for diffuse reflectance spectroscopy,” J. Biomed. Opt. 19(10), 107002 (2014). [CrossRef]
25. U. Dahlstrand, R. Sheikh, C. D. Nguyen, J. Hult, N. Reistad, and M. Malmsjö, “Photoacoustic imaging for three-dimensional visualization and delineation of basal cell carcinoma in patients,” Skin Res Technol 24(4), 667–671 (2018). [CrossRef]
26. S. L. Jacques, “Optical properties of biological tissues: a review,” Phys. Med. Biol. 58(11), R37–R61 (2013). [CrossRef]
27. R. Sheikh, Techniques for Measuring Perfusion during Reconstructive Surgery and Effects of Epinephrine in Local Anesthetics, Lund University, 2018.
28. J. Yang, G. Zhang, W. Chang, Z. Chi, Q. Shang, M. Wu, T. Pan, L. Huang, and H. Jiang, “Photoacoustic imaging of hemodynamic changes in forearm skeletal muscle during cuff occlusion,” Biomed. Opt. Express 11(8), 4560 (2020). [CrossRef]
29. A. Karlas, M. Kallmayer, N. A. Fasoula, E. Liapis, M. Bariotakis, M. Krönke, M. Anastasopoulou, J. Reber, H. H. Eckstein, and V. Ntziachristos, “Multispectral optoacoustic tomography of muscle perfusion and oxygenation under arterial and venous occlusion: A human pilot study,” J. Biophotonics 13(6), 1 (2020). [CrossRef]
30. A. Helfen, M. Masthoff, J. Claussen, M. Gerwing, W. Heindel, V. Ntziachristos, M. Eisenblätter, M. Köhler, and M. Wildgruber, “Multispectral optoacoustic tomography: intra- and interobserver variability using a clinical hybrid approach,” J. Clin. Med. 8(1), 63 (2019). [CrossRef]
31. J. Kanitakis, “Anatomy, histology and immunohistochemistry of normal human skin,” Eur. J. Dermatology 12, 390 (2002).
32. S. Tzoumas, A. Nunes, I. Olefir, S. Stangl, P. Symvoulidis, S. Glasl, C. Bayer, G. Multhoff, and V. Ntziachristos, “Eigenspectra optoacoustic tomography achieves quantitative blood oxygenation imaging deep in tissues,” Nat. Commun. 7(1), 12121 (2016). [CrossRef]
33. R. Sheikh, K. Memarzadeh, C. Torbrand, J. Blohmé, and M. Malmsjö, “Hypoperfusion in response to epinephrine in local anaesthetics: Investigation of dependence on epinephrine concentration, spread of hypoperfusion and time to maximal cutaneous vasoconstriction,” J. Plast. Reconstr. Aesthetic Surg. 70(3), 322–329 (2017). [CrossRef]
34. R. Sheikh, J. Bunke, R. L. Thorisdottir, J. Hult, K. Tenland, B. Gesslein, N. Reistad, and M. Malmsjö, “Hypoperfusion following the injection of epinephrine in human forearm skin can be measured by RGB analysis but not with laser speckle contrast imaging,” Microvasc. Res. 121, 7–13 (2019). [CrossRef]
35. S. Ghali, K. R. Knox, J. Verbesey, U. Scarpidis, K. Izadi, and P. A. Ganchi, “Effects of lidocaine and epinephrine on cutaneous blood flow,” J. Plast. Reconstr. Aesthetic Surg. 61(10), 1226–1231 (2008). [CrossRef]
36. T. P. O’Malley, G. N. Postma, M. Holtel, and D. A. Girod, “Effect of local epinephrine on cutaneous bloodflow in the human neck,” Laryngoscope 105(2), 140–143 (1995). [CrossRef]
37. R. Sheikh, J. Hult, J. Bunke, U. Dahlstrand, C. Ansson, K. Memarzadeh, and M. Malmsjö, “Maximal haemostatic effect is attained in porcine skin within 7 min of the administration of a local anaesthetic together with epinephrine, refuting the need for a 30 min waiting time,” JPRAS Open 19, 77–81 (2019). [CrossRef]
38. I. Fredriksson, M. Larsson, and T. Strömberg, “Optical microcirculatory skin model: assessed by Monte Carlo simulations paired with in vivo laser Doppler flowmetry,” Microvasc. Res. 78(1), 4–13 (2009). [CrossRef]
39. J. Wang, D. Zhu, M. Chen, and X. Liu, “Assessment of optical clearing induced improvement of laser speckle contrast imaging,” J. Innov. Opt. Health Sci. 03(03), 159–167 (2010). [CrossRef]
40. J. Bunke, R. Sheikh, N. Reistad, and M. Malmsjö, “Extended-wavelength diffuse reflectance spectroscopy for a comprehensive view of blood perfusion and tissue response in human forearm skin,” Microvasc. Res. 124, 1–5 (2019). [CrossRef]
41. A. L. Cole, R. A. Herman, J. B. Heimlich, S. Ahsan, B. A. Freedman, and M. S. Shuler, “Ability of near infrared spectroscopy to measure oxygenation in isolated upper extremity muscle compartments,” The Journal of Hand Surgery 37(2), 297–302 (2012). [CrossRef]
42. K. R. Ward, R. R. Ivatury, R. W. Barbee, J. Terner, R. Pittman, I. P. Torres Filho, and B. Spiess, “Near infrared spectroscopy for evaluation of the trauma patient: a technology review,” Resuscitation 68(1), 27–44 (2006). [CrossRef]