A transmissive laser speckle imaging technique for measuring deep tissue blood flow: an example application in finger joints.

BACKGROUND AND OBJECTIVE: Laser speckle perfusion imaging (LSPI) is a minimally invasive optical measure of relative changes in blood flow, providing real-time, high resolution, two-dimensional maps of vascular structure. Standard LSI imaging uses a light-reflective geometry that limits the measurement to a thin surface layer of 0.2-1 mm. The objective of this study was to test a new LSI instrument geometry with the laser source opposed to the image capture plane (light transmissive). Captured light then travels the entire tissue thickness (10-15 mm), sampling much deeper regions of interest than conventional optical imaging techniques. STUDY DESIGN: Reflective-light (conventional) and transmissive-light LSI modes were used to measure finger joint blood flow during a timed tourniquet occlusion of the brachial artery in volunteer participants. RESULTS: There was greatly increased visibility of vessels underlying the skin in the light-transmissive mode LSI mode. Established LSI algorithms were shown to still work in the light-transmissive mode, despite decorrelation due to finite laser coherence length and the light passing through a tissue thickness of 10-15 mm. CONCLUSION: Transmissive LSI can be used to measure blood flow deep (10-15 mm) into tissues. This could be useful for non-invasive measurements of finger joint synovial blood flow in diagnosing and treating peripheral vascular disorders, such as rheumatoid arthritis.

Kinetics of blood flow during healing of excisional full-thickness skin wounds in pigs as monitored by laser speckle perfusion imaging.

BACKGROUND/PURPOSE: The laser speckle perfusion imaging (LSPI) system is a new, non-invasive technique for rapidly and reproducibly measuring tissue perfusion. The high resolution and frame rate of the LSPI overcome many of the limitations of traditional laser Doppler imaging techniques. Therefore, LSPI is a useful means for evaluating blood flow in a variety of situations. The present study investigates the ability of the LSPI system to detect temporal changes in blood flow during the healing of cutaneous wounds in a well-characterized animal model. METHODS: Full-thickness excisional skin wounds (2 x 2 cm) were created on the backs of juvenile female red Duroc pigs. Every week post-injury, the wounds were measured and photographed, and normalized blood flow values were determined using the LSPI system. RESULTS: Tissue perfusion values were available after complete re-epithelialization and removal of the eschar, at day 21. At this point, wound blood flow was significantly elevated as compared with the surrounding, uninvolved skin. Wound blood flow declined steadily during healing, and approached normal values by day 35 post-injury. CONCLUSION: The kinetics of blood flow during excisional wound healing in the red Duroc model are comparable with that previously observed in laser Doppler imaging of healing human skin wounds and hypertrophic scars. These results therefore confirm that the red Duroc is a good model of human wound healing, and further indicates that the LSPI is an excellent technique for evaluating angiogenesis and neovascularization during healing in this and other models.

Endoscopic laser speckle imaging of tissue blood flow: applications in the human knee.

This work represents the first clinical data acquired with the endoscopic laser speckle imaging (eLSPI) system, a new diagnostic tool developed for real-time imaging of tissue blood flow during endoscopic surgical procedures. eLSPI was used to image tissue perfusion in the medial compartment of the knee of five patients requiring arthroscopic knee surgery. The effectiveness of eLSPI as a diagnostic tool was tested by measuring changes in tissue perfusion resultant from tourniquet application, and intra-articular epinephrine. eLSPI produced real-time perfusion video images of tissue blood flow in the knee joint. Tourniquet applications produced consistent decreases in mean perfusion index measurements (29.3% +/- 5.1% in meniscus; 39.5% +/- 8.2% in synovium with an intra patient variability of 6%-9%). A dose-dependent vasoconstrictive response to the administration of intra-articular epinephrine was measured, with maximum dose producing a mean decrease in perfusion of 31.0%-9.3% in meniscus and 41.2%-10.9% in synovium. eLSPI consistently detects decreases in articular tissue blood flow resultant from tourniquet inflation or from the administration of increasing concentrations of epinephrine. These are the first in vivo results indicating physiologic changes in articular tissue as a function of two commonly applied practices in endoscopic joint surgery.

Characterization of a 10.3-microm pulsed DFB quantum cascade laser.

We have measured the output parameters of a 10.3-microm pulsed distributed-feedback (DFB) quantum cascade (QC) laser manufactured by Alpes Lasers and intended for high-sensitivity detection of ammonia and ethylene. The laser beam was collimated with an AR-coated aspheric ZnSe lens with focal length of 11.6mm and clear aperture of 16.5mm. Near- and far-field distributions of the laser emission were recorded with an infrared imaging camera. The fast-and slow-axis laser beam divergences were measured to be 1.2 and 1.4 mrad (FWHM), respectively. The divergence was found to be increasing with injection current. An air-spaced Fabry-Perot interferometer with free spectral range of 0.05 cm(-1) was used to measure the frequency tuning rates of the laser. The laser was tuned by either heat sink temperature, injection current or pulse repetition rate with rates of approximately -8 x 10(-2)cm(-1)K(-1), -7 x 10(-2)cm(-1)A(-1) and -9 x 10(-4)cm(-1)kHz(-1), respectively. The laser frequency decreased linearly with a rate of 10(-2)cm(-1)ns(-1) ( approximately 300 MHzns(-1)) for laser pulses varied from 10 to 50 ns, and the frequency chirp rate was found to decrease for longer laser pulses.

Frequency tuning of long-wavelength VCSELs.

Tuning properties of long-wavelength VCSELs have been studied experimentally, for the first time to our knowledge. Injection current and temperature tuning rates of two VCSELs operating near 1,512 and 1,577 nm have been measured using a Fabry-Perot etalon with free spectral range 0.056 cm(-1). A 100-Hz saw-tooth modulation with depths of modulation of approximately 10% or less was superimposed on a direct injection current (dc bias) to tune lasers in narrow spectral intervals (0.3-1.2 cm(-1)) around a central frequency set by the dc bias. The lasers have been found to be capable of being tuned faster at higher levels of dc bias. The enhancement factors were up to approximately 2 and approximately 3 for the 1,512- and 1,577-nm lasers, respectively, as compared with their tuning rates measured at the levels of the dc bias close to the threshold of lasing. A linear dependence between injection current tuning rates and the levels of dc bias has been observed. Temperature tuning coefficients have been proved to be independent of the laser heat sink temperature and of the dc bias. Frequency tuning curves were approximated with a second-order polynomial. The frequencies of more than 40 absorption lines of CO, CO(2), H(2)O and NH(3) known from spectral databases were compared with the calculated frequencies. The accuracy of the approximation was found to be within 0.2 cm(-1) for spectral intervals up to 38 cm(-1). The dependence of current tuning rates of the VCSELs on dc bias was shown to be taken into account for accurate analysis of absorption line profiles. The results obtained can be used for precise spectroscopic measurements with long-wavelength VCSELs.

A comparison of two laser-based methods for determination of burn scar perfusion: laser Doppler versus laser speckle imaging.

UNLABELLED: Laser Doppler perfusion imaging (LDI) is an established technique for early assessment of burn depth to help determine a course of treatment. Laser speckle perfusion imaging (LSPI) is an alternative laser based, non-invasive perfusion monitoring technique that offers rapid and high resolution images of tissue. We have evaluated the ability of the LSPI instrument in determining and monitoring burn scar perfusion over time and compared it with the LDI instrument as a standard. METHODS: Ten patients with hypertrophic burn scars (time since injury: 1-8 months) were recruited. Burn scars were scanned with both instruments (LSPI and LDI) monthly over a period of 11 months. Clinical grading of the burn scars was assessed on every scan date using the Vancouver burn scar scale. RESULTS: Comparison of the perfusion values determined by each instrument shows a strong positive correlation, r2=0.86 (n=63). Each instrument's output also correlated significantly with the clinical grading of the scar, indicating the expected decrease in perfusion as the clinical condition of the scars improved with time. SIGNIFICANCE: The new LSPI instrument compared favorably with the established LDI instrument, yielding similar results. The considerably faster scan time and higher resolution of the LSPI method provides a distinct clinical advantage, both in terms of patient comfort and for reliably matching perfusion characteristics to their associated anatomical features. The fast temporal response of the LSPI instrument could be used to monitor near real-time responses to mechanical or pharmacological interventions to study dynamic vascular changes to burn damaged tissues.

A laser speckle imaging technique for measuring tissue perfusion.

Laser Doppler imaging (LDI) has become a standard method for optical measurement of tissue perfusion, but is limited by low resolution and long measurement times. We have developed an analysis technique based on a laser speckle imaging method that generates rapid, high-resolution perfusion images. We have called it laser speckle perfusion imaging (LSPI). This paper investigates LSPI output and compares it to LDI using blood flow models designed to simulate human skin at various levels of pigmentation. Results show that LSPI parameters can be chosen such that the instrumentation exhibits a similar response to changes in red blood cell concentration (0.1%-5%, 200 microL/min) and velocity (0-800 microL/min, 1% concentration) and, given its higher resolution and quicker response time, could provide a significant advantage over LDI for some applications. Differences were observed in the LDI and LSPI response to tissue optical properties. LDI perfusion values increased with increasing tissue absorption, while LSPI perfusion values showed a slight decrease. This dependence is predictable, owing to the perfusion algorithms specific to each instrument, and, if properly compensated for, should not influence each instrument's ability to measure relative changes in tissue perfusion.

Endoscopic laser imaging of tissue perfusion: new instrumentation and technique.

BACKGROUND AND OBJECTIVES: New instrumentation, based on a previously established laser speckle perfusion imaging (LSI) technique is evaluated for its ability to capture and generate blood flow images during endoscopic surgery. STUDY DESIGN/MATERIALS AND METHODS: Investigations are detailed in an in-vitro blood flow model simulating physiological properties of vascularized tissue, and in-vivo in rabbit joint capsule tissue. RESULTS: In-vitro measurements showed a linear response of the instrument to blood flow in the range of 0-800 microl/minute, where data points were significantly correlated with an r(2) value of 0.96. In-vivo measurements showed a 58.7% decrease to the medial collateral ligament during occlusion of the femoral artery. CONCLUSIONS: Blood flow images demonstrate that the endoscopic LSI technique is capable of measuring relative tissue blood flow changes at high resolutions and rapid response times and incorporates well with endoscopic surgeries.

Comparison of laser speckle and laser Doppler perfusion imaging: measurement in human skin and rabbit articular tissue.

Laser Doppler perfusion imaging (LDI) is currently used in a variety of clinical applications, however, LDI instruments produce images of low resolution and have long scan times. A new optical perfusion imager using a laser speckle measurement technique and its use for in vivo blood flow measurements are described. Measurements of human skin and surgically exposed rabbit tissue made using this instrument were compared with a commercial laser Doppler perfusion imaging instrument. Results from blood flow measurements showed that the laser speckle imager measured an 11-67% decrease in blood flow under arterial occlusion. Under similar conditions, the laser Doppler imager measured blood flow decreases of 21-63%. In comparison with LDI, it was observed that the higher temporal resolution of the laser speckle imager was more sensitive to measuring the hyperaemic response immediately following occlusion. This in vivo study demonstrated some of the several advantages laser speckle imaging has over conventional LDI, making the new instrument more versatile in a clinical environment.

Light dosimetry using the P3 approximation.

In earlier work, we demonstrated that radiance, calculated using the P3 approximation in a plane wave geometry, could be used to accurately predict the optical parameters of an Intralipid/methylene blue phantom. Plane wave geometry is impractical for clinical use but the results of this work encouraged us to further develop the P3 approximation for a spherical geometry, described in this paper. Radiance predicted by this model for a defined Intralipid/methylene blue phantom was compared with radiance measured in this phantom. The results demonstrate that the spherical derivation of the P3 approximation will reproducibly predict optical parameters of a tissue phantom as effectively as the slab geometry derivation of the P3 approximation. In a similar protocol, the P3 approximation was used to estimate the optical parameters of ex vivo human prostate. Radiance in this case was measured in the prostate samples using an after loading technique. Three prostate samples tested were found to be surprisingly optically homogeneous. The after loading protocol described in this paper could form the basis of a minimally invasive and effective clinical method to optically characterize human prostate.

Detection of transplant vasculopathy in a rat aortic allograft model by fluorescence spectroscopic optical analysis.

BACKGROUND AND OBJECTIVE: Transplant vasculopathy is a leading cause of late cardiac graft loss. We have examined laser-induced fluorescence (LIF) spectroscopy as an optical diagnostic tool for detection of intimal plaque development and inflammatory cellular invasion in a rat model of aortic allograft transplant. STUDY DESIGN/MATERIALS AND METHODS: Infrarenal aortic segments were transplanted from Lewis to Sprague Dawley rats. A range of vasculopathy development was produced by treatment with a viral anti-inflammatory protein. LIF spectra were recorded from the intima of aortic implants at 28 days. Fluorescence intensity was analyzed for correlation with vasculopathy development. RESULTS: Significant differences in LIF intensity at 400-450 nm (P < or = 0.05 by ANOVA) were detected. LIF emission was correlated with plaque growth (R2 = 0.980), vessel narrowing (R2 = 0.964), and cellular invasion (R2 = 0.971) by regression analysis. CONCLUSION: LIF optical analysis provides a nontraumatic diagnostic approach for detection of atherosclerosis prior to cardiac transplant or during development of vasculopathy after transplant.

Radiance modelling using the P3 approximation.

Light dosimetry is an essential component of effective photodynamic therapy (PDT) of tumours. Present PDT light dosimetry techniques rely on fluence-based models and measurements. However, in a previous paper by Barajas et al, radiance-based light dosimetry was explored as an alternative approach. Although successful in demonstrating the use of Monte Carlo (MC) simulations of radiance in tissue optical characterization, the MC proved time consuming and impractical for clinical applications. It was proposed that an analytical solution to the transport equation for radiance would be desirable as this would facilitate and increase the speed of tissue characterization. It has been found that the P3 approximation is one such potential solution. Radiance and fluence expressions based on the P3 approximation were used to optically characterize an Intralipid-based tissue phantom of varying concentration of scatterer (Intralipid) and absorber (methylene blue) using a plane wave illuminated, semi-infinite medium geometry. The results obtained compare favourably with the Grosjean approximation of fluence (a modified diffusion theory) using the same optical parameters (mu(a), mu(s), g). The results illustrate that radiance-based light dosimetry is a viable alternative approach to tissue characterization and dosimetry. It is potentially useful for clinical applications because of the limited number of invasive measurements needed and the speed at which the tissue can be characterized.

Biodistribution of Photofrin II and 5-aminolevulinic acid-induced protoporphyrin IX in normal rat bladder and bladder tumor models: implications for photodynamic therapy.

Photodynamic therapy (PDT) has been considered as a potential therapy for superficial bladder carcinomas. Cutaneous photosensitivity and reduction of bladder capacity are the two well-known complications following systemic administration of the commonly used photosensitizer, Photofrin II (PII). The objective of the present study was to evaluate whether intravesical (i.b.) instillation of photosensitizers for PDT of bladder cancer might be a more suitable treatment method. Female Fischer rats were utilized to develop orthotopic and heterotopic bladder tumor models. Rats bearing orthotopic bladder tumors were treated either intravesically or intravenously with graded doses of 5-aminolevulinic acid (ALA) or PII. Normal rats received the same doses of ALA or PII. As well, rats bearing heterotopic tumor were studied for comparison. The biodistribution times (times allowed for tissue uptake and bioconversion following drug administration) were 2, 4 or 6 h. Porphyrin fluorescence intensities within tumor, urothelium, submucosa, bladder muscularis and abdominal muscle were quantitated by confocal laser scanning microscopy. Following intravenous (i.v.) injection of ALA, tumor protoporphyrin IX (PpIX) levels peaked at 4 h and diminished by 6 h. The PpIX ratios of tumor-to-bladder mucosa, submucosa and muscle layers were 3:1, 5:1 and 8:1, respectively, 4 h following 1000 mg/kg ALA injection. After ALA instillation, the optimal biodistribution time appeared to be 4 h. Bladder instillation provided comparable tumor labeling with the i.v. route, but lost selectivity of PpIX accumulation between tumor and normal urothelium. The PpIX ratio of tumor-to-bladder muscularis was 5:1. After i.b. instillation of PII, porphyrin fluorescence was detected only within tumor and urothelium, while porphyrin fluorescence was mainly located in bladder submucosa following i.v. injection. Intravesical administration of ALA or PII might be feasible for PDT of superficial bladder cancers.

Monte Carlo modelling of angular radiance in tissue phantoms and human prostate: PDT light dosimetry.

Photodynamic therapy (PDT) is a promising technique for destroying tumours. Photosensitizing drugs presently available are not sufficiently tumour specific; hence, light dosimetry is required in order to control light exposure and thereby restrict cell kill to the target tissue to avoid damage to healthy tissue. Current light dosimetry methods rely on tissue optical characterization by fluence measurements at several points. Fluence-based tissue characterization is impractical for tumours in organs such as prostate where access by optical probes is limited and the tumours are highly optically inhomogeneous. This paper explores the potential of radiance-based light dosimetry as an alternative. Correlation is found between Monte Carlo simulation of radiance in a tissue phantom and radiance measurements made using a new radiance probe. Radiance is sensitive to variations in the tissue optical parameters, absorption coefficient mu(a), scattering coefficient mu(s), and anisotropy factor g, and therefore is potentially useful for tissue characterization. Radiance measurements have several advantages over fluence measurements. Radiance measurements provide more information from a single location, better spatial resolution of the tissue optical parameters, and higher sensitivity in discriminating between different media. However, the Monte Carlo method is too slow to be of practical value for tissue characterization by correlation of measured and simulated radiance. An analytical solution to the transport equation for radiance would be desirable as this would facilitate and increase the speed of tissue characterization.

Preclinical assessment of hypocrellin B and hypocrellin B derivatives as sensitizers for photodynamic therapy of cancer: progress update.

Hypocrellins are perylenequinone pigments with substantial absorption in the red spectral region and high singlet oxygen yield. They are available in pure monomeric form and may be derivatized to optimize properties of red light absorption, tissue biodistribution and toxicity. In vitro screening of synthetic derivatives of the naturally occurring compound, hypocrellin B (HB), for optimal properties of cyto-(dark) toxicity and phototoxicity resulted in selection of three compounds for preclinical evaluation: HBEA-R1 (ethanolaminated HB), HBBA-R2 (butylaminated HB) and HBDP-R1 [2-(N,N-dimethylamino)-propylamine-HB]. Extinction coefficients at 630 nm (epsilon 630) are 6230, 6190 and 4800, respectively; and 1O2 quantum yields, phi, 0.60, 0.32 and 0.42. Intracellular uptake is essentially complete within 2 h (HBEA-R1, HBBA-R2) and 20 h (HBDP-R1). Greatest uptake is associated with lysosomes and Golgi. The HBEA-R1 and HBBA-R2 elicit phototoxicity in vitro primarily via the type II mechanism, with some type I activity under stringently hypoxic conditions. Transcutaneous phototherapy with HBEA-R1 permanently ablates EMT6/Ed tumors growing in the flanks of Balb/c mice, with minimal cutaneous effects. The HBBA-R2 does not elicit mutagenic activity in strains TA98 and TA100 of Salmonella typhimurium. Further development of selected hypocrellin derivatives as photosensitizers for photodynamic therapy is warranted.

In vivo light transmission spectra in EMT6/Ed murine tumors and Dunning R3327 rat prostate tumors during photodynamic therapy.

BACKGROUND AND OBJECTIVE: Variations in the optical coefficients in tissue and the photosensitizer during photodynamic therapy (PDT) will require adjustment of the light dose during the course of therapy. We have studied the dynamics using light transmission spectra for two different tumor models when tetrasulfonated aluminum phthalocyanine (AlPcS4) was used as photosensitizer. STUDY DESIGN/MATERIALS AND METHODS: Spectra were measured noninvasively in the EMT6/Ed murine tumor model, and with interstitially implanted source and probe fibers in the Dunning R3327-AT rat tumor model. Measurements were performed in the range 600-840 nm, using a tunable dye laser, a diode laser, and a Ti:Sapphire laser. AlPcS4 has absorption in the range 600-700 nm with an absorption peak at 670 nm in saline. RESULTS: The in vivo spectrum of AlPcS4 both in the EMT6/Ed tumor model and the Dunning R3327-AT tumor model differs from the spectrum of AlPcS4 in saline. The absorption at 670 nm was reduced, whereas the absorption at 640 nm increased. Exposure of phototherapeutic levels of light caused reduced light absorption by the photosensitizer and further spectral shift. CONCLUSION: We found that the AIPcS4 absorption spectrum changes in a biological environment, and we also observed increased light transmission at the treatment wavelength during PDT in both tumor models. Instability in the absorption spectrum of the photosensitizer may influence the effectiveness of PDT.

Temporal and illumination-induced variations in the in vivo light transmission spectra of four photosensitizers in EMT6/Ed murine tumours.

Temporal and illumination-induced variations in the in vivo light transmission spectrum of the photosensitizer will influence light dosimetry for photodynamic therapy (PDT). The present authors have studied the in vivo spectra of four photosensitizers in the EMT6/Ed murine tumour model in Balb/c mice. The following photosensitizers were used: bis(dimethylthexylsiloxy)silicon 2,3-naphthalocyanine (SiNc 8), benzoporphyrin-derivative monoacid ring A (BPD Verteporfin), Photofrin and ethanolamined hypocrellin B (HBEA-R2). Spectra were measured non-invasively in the EMT6/Ed murine tumour model in the spectral range 600-840 nm, using a diode laser, a dye laser and a Ti:sapphire laser. Red-shift and broadening of the SiNc 8 absorption band was observed at 790 nm, and a slight red-shift was observed in the BPD, HBEA-R2 and Photofrin in vivo absorption spectrum. Exposure to 300 J of light at the peak absorption wavelength caused complete photobleaching of BPD at 690 nm, and a reduced absorption by SiNc 8 at 780 nm, Photofrin at 626 nm, and HBEA-R2 at 656 nm.

Hypocrellins as photosensitizers for photodynamic therapy: a screening evaluation and pharmacokinetic study.

Hypocrellin compounds were selected as potential photosensitizers for photodynamic therapy (PDT) owing to their high quantum yields of singlet oxygen (1O2), and facility for site-directed chemical modification to enhance phototoxicity, pharmacokinetics, solubility, and light absorption in the red spectral region, among other properties. Parent hypocrellins A and B share an absorption peak at 658 nm. These molecules may therefore be considered useful progenitors of derivatives which absorb more strongly in the red, considering that the ideal sensitizer should absorb in the 650-800 nm range, beyond the absorption range of hemoglobin and melanin, and where light penetration in tissues is maximized through reduced scattering. A series of pure, monomeric hypocrellin derivatives was tested for properties of dark cytotoxicity and photosensitizing potential by clonogenic assay in monolayer cultures of EMT6/Ed murine tumor cells. Their respective toxicities are reported on a molar basis. The in vitro screening assay has, to date, resulted in the selection of four hypocrellin derivatives for further development as photosensitizers for PDT. Cellular uptake for photosensitizing doses of selected compounds was determined by fluorimetry. Dose escalation studies in rodents indicate that potentially photosensitizing doses promote no demonstrable systemic toxicity.

Anisotropy of radiance in tissue phantoms and Dunning R3327 rat tumors: radiance measurements with flat cleaved fiber probes.

BACKGROUND AND OBJECTIVE: The goal of this study is to determine if flat cleaved fiber probes are appropriate for interstitial measurements of radiance in tissue. Flat cleaved probes have the advantage of high responsivity, and they are easy to insert into tissue. Owing to the non-isotropic response of flat cleaved probes, a calibration function is required, taking the anisotropy in the radiance in tissue into account. STUDY DESIGN, MATERIALS AND METHODS: The method used to determine this function consists of radiance measurements in tissue, performed with a flat cleaved fiber probe mounted on a stereotactic stage for insertion into the tissue from different directions. Interstitial irradiation at 630 nm was delivered by a spherical source. RESULTS: We found that the degree of anisotropy in the radiance decreases with increasing distance from the interstitially implanted source in two different tissue phantoms and in the Dunning R3327-AT and R3327-H rat tumor models. CONCLUSION: A position-dependent calibration function is required for interstitially implanted flat cleaved fiber probes. An anisotropy function is presented, which modifies the measurements of radiance with a flat cleaved probe, to account for the change in anisotropy in the radiance. The anisotropy functions for the two tumor models differ substantially.

Photosensitization by anticancer agents 21: new perylene- and aminonaphthoquinones.

Hypocrellins are under intensive investigation as photosensitizing agents for photodynamic therapy (PDT). A recent advance in the synthesis of hypocrellin congeners resulted in the production of an amino-substituted hypocrellin-B, and its "half chromophore." Both compounds exhibit stronger red light absorption than previously reported hypocrellins, and, therefore, merit investigation as photosensitizers.