Beauty may be easy to see because it’s only skin deep, but looking below the skin’s surface hasn’t always been as easy to do. Even though a variety of optical imaging techniques are available, such as pulse oximeters and simple vision systems, most of these techniques have limitations. They’re limited by their resolution, their ability to penetrate the skin, and the amount of time needed to create an image. However, advances in imaging technology are creating a new generation of tools that will allow you to look below the skin and see detailed images of sub-surface structures generated in real time that you can use to make better diagnoses or to improve surgical outcomes. Confocal Reflectance Scanning Confocal reflectance imaging is a promising approach for examining the morphology of living tissues. (See images above for examples of this technology.) Presently, confocal reflectance imaging can produce good-quality images of nuclear and cellular detail, at morphology in vivo, and in excised tissue ex vivo. “We have quickly moved from trying to understand how normal human skin appears under confocal reflectance scanning to being able to characterize pigmented skin lesions, basal cell cancers, margins of tumors and lesions, and inflammatory skin conditions,” explained Milind Rajadhyaksha, Ph.D., Principal Research Scientist at the Optical Science Laboratory at Northeastern University. “Our goal is to develop criteria that may be useful for clinical screening and diagnostic utility,” Dr. Rajadhyaksha said during a presentation at the American Society for Laser Medicine and Surgery’s annual meeting last April. According to Dr. Rajadhyaksha, one promising area of application is to use confocal reflectance imaging to guide Mohs micrographic surgery. The common features in these cancers can be viewed using a confocal reflective microscope, including the following: • elongated, monomorphic nuclei in the basal layer • increased dilated vascularity • large numbers of inflammatory infiltrates • the dynamic process of leukocytes rolling and sticking to the endothelial wall. “More work is required to understand the morphology of these cancers as seen with the confocal reflectance microscope and to identify other characteristic features, such as parakeratosis, prominent nucleoli, palisading solar elastosis, clefting and mucin,” said Dr. Rajadhyaksha. The goal is to use confocal imaging to detect these cancers pre-operatively or while the patient is undergoing surgery. This would allow the dermatologic surgeon to determine in real time whether the patient needs additional surgery. Dr. Rajadhyaksha and others conducted an ex vivo study on skin excised during Mohs surgeries. Using a combination of aceto-whitening and crossed-polarization to brighten the nuclei, the researchers enhanced the contrast and visibility of nuclear morphology in basal cell cancers. “We have looked at about 200 cases, and the correspondence is very good between the confocal images and the pathology,” explained Dr Rajadhyaksha. “We can detect tumors in about two-thirds of the cases, and we hope to improve on that by improving our instruments, although there may be some tissue chemistry that we need to better understand.” The confocal line scanner consists of a collimated diode laser beam focused by a cylindrical lens and an objective lens to produce a line that is scanned by a prismatic mirror driven by a simple galvanometric scanner. Light that is back scattered from the line is collected and focused onto a linear CMOS Detector through a detection slit. This design incorporates seven main optical complements of which six can be obtained off the shelf. The only non-standard elements are the linear CMOS detector and the video control and timing electronics. Imaging Superficial Tissue Layers with Polarized Light Researchers in Oregon are using polarized lights and filtered cameras to create images of structures just below the surface of the skin. According to Steven L. Jacques, Ph.D., Departments of Dermatology and Biomedical Engineering, Oregon Health & Science University, the process involves illuminating the skin with linearly polarized light and taking two images with a CCD camera, one through a second polarizer aligned parallel to the illumination and one aligned perpendicular to the illumination. The difference in images yields a new image that subtracts the scattered photons from deep in the skin and retains only the photons scattered by the superficial 0.3 mm of skin. “The new image uses only about 10% of the reflected photons, rejecting the other 90% of photons that normally blind the doctor’s view of the skin,” said Dr. Jacques, “unmasking the detail of the superficial but subsurface skin structure where cancer and other pathology occur.” Under polarization, many surface features, such as freckles, simply disappear. A simple nevus, which appears dark under normal white light, appears with greater detail because of the scatter of light from whatever comprises the nevus, which is probably melanosomes. With structures such as a neuro-fibroma, polarized light creates patterns of light and dark, showing where light is being absorbed and reflected. (For an example of images taken with this technology, see below.) “The trick for us now is to learn how to interpret those patterns in these structures to determine what is actually happening in the skin,” explained Dr. Jacques. “Images of actinic keratosis and basal cell carcinoma reveal the structure of the lesions, which is typically poorly visible under normal light.” “We are looking at ways that the imaging system can be used to derive other information about the nature of the structures,” said Dr. Jacques. “We found that we can extrapolate information based on the angle of light scattering, such as the size of the particle that’s doing the reflecting.” For example, a normal cell nucleus reflects at one type of angle, but if it’s abnormal, it reflects at another type of angle. According to Dr. Jacques, all the cellular structures can be mapped — mitochondria, membranes, endoplasmic reticulum and collagen fibers — based on the angular distribution of the light that’s reflected from the tissue. Imaging equipment can capture this information and create images that accent different aspects of cellular structures, for example, the distribution of abnormal, enlarged nuclei. “There is still much work to be done to make a truly functional diagnostic tool, but other researchers using similar tools report that they can see light scattering changes due to nuclear change,” explained Dr. Jacques. “These images may allow detection 1 to 2 weeks prior to the first biochemical marker that can be measured, and many weeks prior to the first morphological marker that can be measured.” Using Spectral Analysis to Encode Nuclear Morphology What if you could image cancerous tissue non-invasively and in real-time, without cutting, removing and staining the tissue? That’s exactly the goal of the research with spectroscopy of refracted light being conducted by Lev T. Perelman, Ph.D., Associate Professor at Harvard Medical School and Director of Biomedical Imaging and Spectroscopy Laboratory at Beth Israel Deaconess Medical Center. “While standard light spectroscopy can give us information about the chemistry and biochemistry of tissue, spectroscopy uses refracted light to produce information about morphology that can be very important for diagnosing various diseases.” According to Dr. Perelman, a refraction or, in other words, light scattering spectrum, generated by an object is dependent upon three primary characteristics of the object: 1. shape 2. size 3. refractive index. The refractive index is a measure of a material’s ability to refract light relative to a known substance. For example, water has a refractive index of 1.33 while that of glass is 1.4. These three parameters define the shape of the spectrum and, conversely, the characteristics of the spectrum, which can be used to derive the shape and size of the object. “When we measured light scattering spectra of T84 cancerous cells and normal cells from the same type of organ, we found significant differences in the spectra,” said Dr. Perelman. (For examples, see the graphic above.) “With subsequent experiments, we began to build spectral models that could be used to extract information about sizes and refractive indexes of specific types of cells.” Dr. Perelman and colleagues found that normal nuclei have a narrow and well-defined distribution, and tumor cells had a much wider distribution and a high refractive index. This difference is caused by the increase in DNA in the nucleus as the cell progresses toward cancer. The resulting increase in the size of the nuclei can also be seen in the changes of the spectra as the nuclei grow. Dr. Perelman and his colleagues conducted a double-blind study involving 16 patients at several hospitals in which an optical fiber probe was inserted in the biopsy channel of an endoscope for spectral data collection and each site was biopsied immediately afterward. “We wanted to detect the increase in size of epithelial nuclei and the increase in concentration of those nuclei,” Dr. Perelman described. “In this study, we were able to quite closely match the spectral results with the results derived from the tissue samples and reliably detect precancerous changes in patients with Barrett’s esophagus.” Multi-Functional Optical Coherence Tomography Johannes de Boer, Ph.D., and his fellow researchers at the Wellman Laboratories for Photomedicine at Massachusetts General Hospital are using in vivo, multi-functional optical coherence tomography (OCT) for simultaneous cross sectional imaging of structure, birefringence and flow in human skin. Dr. de Boer, who is an Assistant Professor in the Department of Dermatology, refers to his approach as multi-functional OCT because he combines multiple detection techniques, such as polarization and Doppler, in a single device to produce images. “Polarization sensitivity in OCT imaging offers promise because there are many tissues in skin that are actually birefringent, including collagen, nerves, muscles and tendons,” said Dr. de Boer. “Doppler is of interest because it allows us to map the vascular structure and qualitatively and quantitatively determine the flow.” OCT uses a broadband light source that is split into a sample arm reflecting from the tissue to be scanned and a reference arm reflecting from a moving mirror. The detector images a series of constructive and destructive combinations of the reflected light from both sources depending on the difference of the path lengths created by the moving mirror, creating an interferogram. “Since the light penetrates beneath the sample’s surface, there are reflections from structures beneath the surface. Each reflecting layer in the sample generates an interferogram,” explained Dr. de Boer. “From the interferograms you can determine the depth of the reflection from the outer surface. By scanning the sample, the detector builds an image of the sample’s interior morphology.” Dr. de Boer hopes to separate and identify various elements of the skin based on the characteristics of the light and produce detailed, fine-grain dimensional images of the sub-surface structure and determine the blood flow through such pathology as burn scars, basal cell carcinoma, sclerotic keratosis and cherry angioma. These images will allow physicians to better determine the extent and organization of skin conditions allowing for better diagnosis and treatment. “Multi-functional OCT creates cross-sectional images 5 mm wide and up to 1.5 mm below the surface on in vivo skin in about 1 second with a resolution of about 10 micron,” says Dr. de Boer. “It gives you quantitative information on layer thickness, structure size and birefringence, and qualitative information on layer, structure, collagen and blood flow.” With this technology you could determine the depth of a burn, you could assess collagen, or you could see more in-depth detail of a basal cell carcinoma. Diffuse Optical Spectroscopy and Diffuse Optical Imaging Diffuse optical spectroscopy and diffuse optical imaging may provide a method for interrogating very large tissue volumes with a moderate resolution up to a few millimeters or several centimeters, 8 or perhaps even 10 centimeters in depth. According to Bruce Tromberg, Ph.D., Professor, Departments of Biomedical Engineering and Surgery and Director, Beckman Laser Institute, University of California, Irvine, these techniques essentially involve shining a light source through tissue and using the diffuse multiple scattered photons that penetrate deeply to obtain a detailed functional and structural image of the tissue. This technique, which Dr. Tromberg estimated will be available within 3 to 5 years, offers the promise of detecting malignant tumors in patients for whom standard detection methods aren’t effective because of factors such as the presence of benign lesions and density of surrounding tissue. “Because of this technology’s ability to detect changes induced by chemo-therapy, we feel that this technique can be used to provide a sensitive, accurate and quantitative measurement of therapeutic efficacy in breast cancer patients,” said Dr. Tromberg. This would allow practitioners to detect changes at an earlier stage to determine if chemotherapy is working for a patient. According to Dr. Tromberg, diffuse photons carry information about the tissue, including cellular proliferation, nucleation, hypoxia and inflammation. Information about the extra-cellular matrix is also transmitted, including insight regarding invasive processes, degradation of the matrix, and necrosis. Various physiological parameters can be quantified, including blood volume and flow, vascular permeability, the concentration of oxy and de-oxy hemoglobin, water and lipid in the tissue. The main challenge is separating light absorption events from light scattering events to make a quantitative measurement. By switching between different semi-conductor diode lasers, different wavelengths of light result. Modulating the amplitude of the light source generates an oscillatory light signal (a photon density wave) whose amplitude and phase depend on the tissue’s optical properties. “The integration of optical methods with conventional imaging techniques and targeted molecular probes is relatively easy to implement. Ultimately, we expect that optical methods will be used routinely to enhance the functional information content of medical images used to diagnose disease and monitor the efficacy of various therapies.” While diffuse optics have a tremendous amount of power on their own, Dr. Tromberg believes that they’re even more powerful when used in combination with other anatomic imaging techniques such as MRI, mammography and ultrasound. “If you combine these approaches with needle localization techniques and with exogenous dyes, molecular dyes that can be targeted or materials like indocyanine green, you can follow their extravasation rates and view the tissues’ biophysical characteristics.” A Small World That’s Getting Bigger Optical techniques are now under development that can provide information about the morphology and architectural structure of in vivo tissue almost instantly. Eventually, you’ll look into the skin at amazing depths and see structures in astonishing detail. The exciting news is that these new developments are not decades in the future. Some may even be available for clinical use before the end of the year.
A Clearer Look at Optical Diagnostic Imaging
Beauty may be easy to see because it’s only skin deep, but looking below the skin’s surface hasn’t always been as easy to do. Even though a variety of optical imaging techniques are available, such as pulse oximeters and simple vision systems, most of these techniques have limitations. They’re limited by their resolution, their ability to penetrate the skin, and the amount of time needed to create an image. However, advances in imaging technology are creating a new generation of tools that will allow you to look below the skin and see detailed images of sub-surface structures generated in real time that you can use to make better diagnoses or to improve surgical outcomes. Confocal Reflectance Scanning Confocal reflectance imaging is a promising approach for examining the morphology of living tissues. (See images above for examples of this technology.) Presently, confocal reflectance imaging can produce good-quality images of nuclear and cellular detail, at morphology in vivo, and in excised tissue ex vivo. “We have quickly moved from trying to understand how normal human skin appears under confocal reflectance scanning to being able to characterize pigmented skin lesions, basal cell cancers, margins of tumors and lesions, and inflammatory skin conditions,” explained Milind Rajadhyaksha, Ph.D., Principal Research Scientist at the Optical Science Laboratory at Northeastern University. “Our goal is to develop criteria that may be useful for clinical screening and diagnostic utility,” Dr. Rajadhyaksha said during a presentation at the American Society for Laser Medicine and Surgery’s annual meeting last April. According to Dr. Rajadhyaksha, one promising area of application is to use confocal reflectance imaging to guide Mohs micrographic surgery. The common features in these cancers can be viewed using a confocal reflective microscope, including the following: • elongated, monomorphic nuclei in the basal layer • increased dilated vascularity • large numbers of inflammatory infiltrates • the dynamic process of leukocytes rolling and sticking to the endothelial wall. “More work is required to understand the morphology of these cancers as seen with the confocal reflectance microscope and to identify other characteristic features, such as parakeratosis, prominent nucleoli, palisading solar elastosis, clefting and mucin,” said Dr. Rajadhyaksha. The goal is to use confocal imaging to detect these cancers pre-operatively or while the patient is undergoing surgery. This would allow the dermatologic surgeon to determine in real time whether the patient needs additional surgery. Dr. Rajadhyaksha and others conducted an ex vivo study on skin excised during Mohs surgeries. Using a combination of aceto-whitening and crossed-polarization to brighten the nuclei, the researchers enhanced the contrast and visibility of nuclear morphology in basal cell cancers. “We have looked at about 200 cases, and the correspondence is very good between the confocal images and the pathology,” explained Dr Rajadhyaksha. “We can detect tumors in about two-thirds of the cases, and we hope to improve on that by improving our instruments, although there may be some tissue chemistry that we need to better understand.” The confocal line scanner consists of a collimated diode laser beam focused by a cylindrical lens and an objective lens to produce a line that is scanned by a prismatic mirror driven by a simple galvanometric scanner. Light that is back scattered from the line is collected and focused onto a linear CMOS Detector through a detection slit. This design incorporates seven main optical complements of which six can be obtained off the shelf. The only non-standard elements are the linear CMOS detector and the video control and timing electronics. Imaging Superficial Tissue Layers with Polarized Light Researchers in Oregon are using polarized lights and filtered cameras to create images of structures just below the surface of the skin. According to Steven L. Jacques, Ph.D., Departments of Dermatology and Biomedical Engineering, Oregon Health & Science University, the process involves illuminating the skin with linearly polarized light and taking two images with a CCD camera, one through a second polarizer aligned parallel to the illumination and one aligned perpendicular to the illumination. The difference in images yields a new image that subtracts the scattered photons from deep in the skin and retains only the photons scattered by the superficial 0.3 mm of skin. “The new image uses only about 10% of the reflected photons, rejecting the other 90% of photons that normally blind the doctor’s view of the skin,” said Dr. Jacques, “unmasking the detail of the superficial but subsurface skin structure where cancer and other pathology occur.” Under polarization, many surface features, such as freckles, simply disappear. A simple nevus, which appears dark under normal white light, appears with greater detail because of the scatter of light from whatever comprises the nevus, which is probably melanosomes. With structures such as a neuro-fibroma, polarized light creates patterns of light and dark, showing where light is being absorbed and reflected. (For an example of images taken with this technology, see below.) “The trick for us now is to learn how to interpret those patterns in these structures to determine what is actually happening in the skin,” explained Dr. Jacques. “Images of actinic keratosis and basal cell carcinoma reveal the structure of the lesions, which is typically poorly visible under normal light.” “We are looking at ways that the imaging system can be used to derive other information about the nature of the structures,” said Dr. Jacques. “We found that we can extrapolate information based on the angle of light scattering, such as the size of the particle that’s doing the reflecting.” For example, a normal cell nucleus reflects at one type of angle, but if it’s abnormal, it reflects at another type of angle. According to Dr. Jacques, all the cellular structures can be mapped — mitochondria, membranes, endoplasmic reticulum and collagen fibers — based on the angular distribution of the light that’s reflected from the tissue. Imaging equipment can capture this information and create images that accent different aspects of cellular structures, for example, the distribution of abnormal, enlarged nuclei. “There is still much work to be done to make a truly functional diagnostic tool, but other researchers using similar tools report that they can see light scattering changes due to nuclear change,” explained Dr. Jacques. “These images may allow detection 1 to 2 weeks prior to the first biochemical marker that can be measured, and many weeks prior to the first morphological marker that can be measured.” Using Spectral Analysis to Encode Nuclear Morphology What if you could image cancerous tissue non-invasively and in real-time, without cutting, removing and staining the tissue? That’s exactly the goal of the research with spectroscopy of refracted light being conducted by Lev T. Perelman, Ph.D., Associate Professor at Harvard Medical School and Director of Biomedical Imaging and Spectroscopy Laboratory at Beth Israel Deaconess Medical Center. “While standard light spectroscopy can give us information about the chemistry and biochemistry of tissue, spectroscopy uses refracted light to produce information about morphology that can be very important for diagnosing various diseases.” According to Dr. Perelman, a refraction or, in other words, light scattering spectrum, generated by an object is dependent upon three primary characteristics of the object: 1. shape 2. size 3. refractive index. The refractive index is a measure of a material’s ability to refract light relative to a known substance. For example, water has a refractive index of 1.33 while that of glass is 1.4. These three parameters define the shape of the spectrum and, conversely, the characteristics of the spectrum, which can be used to derive the shape and size of the object. “When we measured light scattering spectra of T84 cancerous cells and normal cells from the same type of organ, we found significant differences in the spectra,” said Dr. Perelman. (For examples, see the graphic above.) “With subsequent experiments, we began to build spectral models that could be used to extract information about sizes and refractive indexes of specific types of cells.” Dr. Perelman and colleagues found that normal nuclei have a narrow and well-defined distribution, and tumor cells had a much wider distribution and a high refractive index. This difference is caused by the increase in DNA in the nucleus as the cell progresses toward cancer. The resulting increase in the size of the nuclei can also be seen in the changes of the spectra as the nuclei grow. Dr. Perelman and his colleagues conducted a double-blind study involving 16 patients at several hospitals in which an optical fiber probe was inserted in the biopsy channel of an endoscope for spectral data collection and each site was biopsied immediately afterward. “We wanted to detect the increase in size of epithelial nuclei and the increase in concentration of those nuclei,” Dr. Perelman described. “In this study, we were able to quite closely match the spectral results with the results derived from the tissue samples and reliably detect precancerous changes in patients with Barrett’s esophagus.” Multi-Functional Optical Coherence Tomography Johannes de Boer, Ph.D., and his fellow researchers at the Wellman Laboratories for Photomedicine at Massachusetts General Hospital are using in vivo, multi-functional optical coherence tomography (OCT) for simultaneous cross sectional imaging of structure, birefringence and flow in human skin. Dr. de Boer, who is an Assistant Professor in the Department of Dermatology, refers to his approach as multi-functional OCT because he combines multiple detection techniques, such as polarization and Doppler, in a single device to produce images. “Polarization sensitivity in OCT imaging offers promise because there are many tissues in skin that are actually birefringent, including collagen, nerves, muscles and tendons,” said Dr. de Boer. “Doppler is of interest because it allows us to map the vascular structure and qualitatively and quantitatively determine the flow.” OCT uses a broadband light source that is split into a sample arm reflecting from the tissue to be scanned and a reference arm reflecting from a moving mirror. The detector images a series of constructive and destructive combinations of the reflected light from both sources depending on the difference of the path lengths created by the moving mirror, creating an interferogram. “Since the light penetrates beneath the sample’s surface, there are reflections from structures beneath the surface. Each reflecting layer in the sample generates an interferogram,” explained Dr. de Boer. “From the interferograms you can determine the depth of the reflection from the outer surface. By scanning the sample, the detector builds an image of the sample’s interior morphology.” Dr. de Boer hopes to separate and identify various elements of the skin based on the characteristics of the light and produce detailed, fine-grain dimensional images of the sub-surface structure and determine the blood flow through such pathology as burn scars, basal cell carcinoma, sclerotic keratosis and cherry angioma. These images will allow physicians to better determine the extent and organization of skin conditions allowing for better diagnosis and treatment. “Multi-functional OCT creates cross-sectional images 5 mm wide and up to 1.5 mm below the surface on in vivo skin in about 1 second with a resolution of about 10 micron,” says Dr. de Boer. “It gives you quantitative information on layer thickness, structure size and birefringence, and qualitative information on layer, structure, collagen and blood flow.” With this technology you could determine the depth of a burn, you could assess collagen, or you could see more in-depth detail of a basal cell carcinoma. Diffuse Optical Spectroscopy and Diffuse Optical Imaging Diffuse optical spectroscopy and diffuse optical imaging may provide a method for interrogating very large tissue volumes with a moderate resolution up to a few millimeters or several centimeters, 8 or perhaps even 10 centimeters in depth. According to Bruce Tromberg, Ph.D., Professor, Departments of Biomedical Engineering and Surgery and Director, Beckman Laser Institute, University of California, Irvine, these techniques essentially involve shining a light source through tissue and using the diffuse multiple scattered photons that penetrate deeply to obtain a detailed functional and structural image of the tissue. This technique, which Dr. Tromberg estimated will be available within 3 to 5 years, offers the promise of detecting malignant tumors in patients for whom standard detection methods aren’t effective because of factors such as the presence of benign lesions and density of surrounding tissue. “Because of this technology’s ability to detect changes induced by chemo-therapy, we feel that this technique can be used to provide a sensitive, accurate and quantitative measurement of therapeutic efficacy in breast cancer patients,” said Dr. Tromberg. This would allow practitioners to detect changes at an earlier stage to determine if chemotherapy is working for a patient. According to Dr. Tromberg, diffuse photons carry information about the tissue, including cellular proliferation, nucleation, hypoxia and inflammation. Information about the extra-cellular matrix is also transmitted, including insight regarding invasive processes, degradation of the matrix, and necrosis. Various physiological parameters can be quantified, including blood volume and flow, vascular permeability, the concentration of oxy and de-oxy hemoglobin, water and lipid in the tissue. The main challenge is separating light absorption events from light scattering events to make a quantitative measurement. By switching between different semi-conductor diode lasers, different wavelengths of light result. Modulating the amplitude of the light source generates an oscillatory light signal (a photon density wave) whose amplitude and phase depend on the tissue’s optical properties. “The integration of optical methods with conventional imaging techniques and targeted molecular probes is relatively easy to implement. Ultimately, we expect that optical methods will be used routinely to enhance the functional information content of medical images used to diagnose disease and monitor the efficacy of various therapies.” While diffuse optics have a tremendous amount of power on their own, Dr. Tromberg believes that they’re even more powerful when used in combination with other anatomic imaging techniques such as MRI, mammography and ultrasound. “If you combine these approaches with needle localization techniques and with exogenous dyes, molecular dyes that can be targeted or materials like indocyanine green, you can follow their extravasation rates and view the tissues’ biophysical characteristics.” A Small World That’s Getting Bigger Optical techniques are now under development that can provide information about the morphology and architectural structure of in vivo tissue almost instantly. Eventually, you’ll look into the skin at amazing depths and see structures in astonishing detail. The exciting news is that these new developments are not decades in the future. Some may even be available for clinical use before the end of the year.
Beauty may be easy to see because it’s only skin deep, but looking below the skin’s surface hasn’t always been as easy to do. Even though a variety of optical imaging techniques are available, such as pulse oximeters and simple vision systems, most of these techniques have limitations. They’re limited by their resolution, their ability to penetrate the skin, and the amount of time needed to create an image. However, advances in imaging technology are creating a new generation of tools that will allow you to look below the skin and see detailed images of sub-surface structures generated in real time that you can use to make better diagnoses or to improve surgical outcomes. Confocal Reflectance Scanning Confocal reflectance imaging is a promising approach for examining the morphology of living tissues. (See images above for examples of this technology.) Presently, confocal reflectance imaging can produce good-quality images of nuclear and cellular detail, at morphology in vivo, and in excised tissue ex vivo. “We have quickly moved from trying to understand how normal human skin appears under confocal reflectance scanning to being able to characterize pigmented skin lesions, basal cell cancers, margins of tumors and lesions, and inflammatory skin conditions,” explained Milind Rajadhyaksha, Ph.D., Principal Research Scientist at the Optical Science Laboratory at Northeastern University. “Our goal is to develop criteria that may be useful for clinical screening and diagnostic utility,” Dr. Rajadhyaksha said during a presentation at the American Society for Laser Medicine and Surgery’s annual meeting last April. According to Dr. Rajadhyaksha, one promising area of application is to use confocal reflectance imaging to guide Mohs micrographic surgery. The common features in these cancers can be viewed using a confocal reflective microscope, including the following: • elongated, monomorphic nuclei in the basal layer • increased dilated vascularity • large numbers of inflammatory infiltrates • the dynamic process of leukocytes rolling and sticking to the endothelial wall. “More work is required to understand the morphology of these cancers as seen with the confocal reflectance microscope and to identify other characteristic features, such as parakeratosis, prominent nucleoli, palisading solar elastosis, clefting and mucin,” said Dr. Rajadhyaksha. The goal is to use confocal imaging to detect these cancers pre-operatively or while the patient is undergoing surgery. This would allow the dermatologic surgeon to determine in real time whether the patient needs additional surgery. Dr. Rajadhyaksha and others conducted an ex vivo study on skin excised during Mohs surgeries. Using a combination of aceto-whitening and crossed-polarization to brighten the nuclei, the researchers enhanced the contrast and visibility of nuclear morphology in basal cell cancers. “We have looked at about 200 cases, and the correspondence is very good between the confocal images and the pathology,” explained Dr Rajadhyaksha. “We can detect tumors in about two-thirds of the cases, and we hope to improve on that by improving our instruments, although there may be some tissue chemistry that we need to better understand.” The confocal line scanner consists of a collimated diode laser beam focused by a cylindrical lens and an objective lens to produce a line that is scanned by a prismatic mirror driven by a simple galvanometric scanner. Light that is back scattered from the line is collected and focused onto a linear CMOS Detector through a detection slit. This design incorporates seven main optical complements of which six can be obtained off the shelf. The only non-standard elements are the linear CMOS detector and the video control and timing electronics. Imaging Superficial Tissue Layers with Polarized Light Researchers in Oregon are using polarized lights and filtered cameras to create images of structures just below the surface of the skin. According to Steven L. Jacques, Ph.D., Departments of Dermatology and Biomedical Engineering, Oregon Health & Science University, the process involves illuminating the skin with linearly polarized light and taking two images with a CCD camera, one through a second polarizer aligned parallel to the illumination and one aligned perpendicular to the illumination. The difference in images yields a new image that subtracts the scattered photons from deep in the skin and retains only the photons scattered by the superficial 0.3 mm of skin. “The new image uses only about 10% of the reflected photons, rejecting the other 90% of photons that normally blind the doctor’s view of the skin,” said Dr. Jacques, “unmasking the detail of the superficial but subsurface skin structure where cancer and other pathology occur.” Under polarization, many surface features, such as freckles, simply disappear. A simple nevus, which appears dark under normal white light, appears with greater detail because of the scatter of light from whatever comprises the nevus, which is probably melanosomes. With structures such as a neuro-fibroma, polarized light creates patterns of light and dark, showing where light is being absorbed and reflected. (For an example of images taken with this technology, see below.) “The trick for us now is to learn how to interpret those patterns in these structures to determine what is actually happening in the skin,” explained Dr. Jacques. “Images of actinic keratosis and basal cell carcinoma reveal the structure of the lesions, which is typically poorly visible under normal light.” “We are looking at ways that the imaging system can be used to derive other information about the nature of the structures,” said Dr. Jacques. “We found that we can extrapolate information based on the angle of light scattering, such as the size of the particle that’s doing the reflecting.” For example, a normal cell nucleus reflects at one type of angle, but if it’s abnormal, it reflects at another type of angle. According to Dr. Jacques, all the cellular structures can be mapped — mitochondria, membranes, endoplasmic reticulum and collagen fibers — based on the angular distribution of the light that’s reflected from the tissue. Imaging equipment can capture this information and create images that accent different aspects of cellular structures, for example, the distribution of abnormal, enlarged nuclei. “There is still much work to be done to make a truly functional diagnostic tool, but other researchers using similar tools report that they can see light scattering changes due to nuclear change,” explained Dr. Jacques. “These images may allow detection 1 to 2 weeks prior to the first biochemical marker that can be measured, and many weeks prior to the first morphological marker that can be measured.” Using Spectral Analysis to Encode Nuclear Morphology What if you could image cancerous tissue non-invasively and in real-time, without cutting, removing and staining the tissue? That’s exactly the goal of the research with spectroscopy of refracted light being conducted by Lev T. Perelman, Ph.D., Associate Professor at Harvard Medical School and Director of Biomedical Imaging and Spectroscopy Laboratory at Beth Israel Deaconess Medical Center. “While standard light spectroscopy can give us information about the chemistry and biochemistry of tissue, spectroscopy uses refracted light to produce information about morphology that can be very important for diagnosing various diseases.” According to Dr. Perelman, a refraction or, in other words, light scattering spectrum, generated by an object is dependent upon three primary characteristics of the object: 1. shape 2. size 3. refractive index. The refractive index is a measure of a material’s ability to refract light relative to a known substance. For example, water has a refractive index of 1.33 while that of glass is 1.4. These three parameters define the shape of the spectrum and, conversely, the characteristics of the spectrum, which can be used to derive the shape and size of the object. “When we measured light scattering spectra of T84 cancerous cells and normal cells from the same type of organ, we found significant differences in the spectra,” said Dr. Perelman. (For examples, see the graphic above.) “With subsequent experiments, we began to build spectral models that could be used to extract information about sizes and refractive indexes of specific types of cells.” Dr. Perelman and colleagues found that normal nuclei have a narrow and well-defined distribution, and tumor cells had a much wider distribution and a high refractive index. This difference is caused by the increase in DNA in the nucleus as the cell progresses toward cancer. The resulting increase in the size of the nuclei can also be seen in the changes of the spectra as the nuclei grow. Dr. Perelman and his colleagues conducted a double-blind study involving 16 patients at several hospitals in which an optical fiber probe was inserted in the biopsy channel of an endoscope for spectral data collection and each site was biopsied immediately afterward. “We wanted to detect the increase in size of epithelial nuclei and the increase in concentration of those nuclei,” Dr. Perelman described. “In this study, we were able to quite closely match the spectral results with the results derived from the tissue samples and reliably detect precancerous changes in patients with Barrett’s esophagus.” Multi-Functional Optical Coherence Tomography Johannes de Boer, Ph.D., and his fellow researchers at the Wellman Laboratories for Photomedicine at Massachusetts General Hospital are using in vivo, multi-functional optical coherence tomography (OCT) for simultaneous cross sectional imaging of structure, birefringence and flow in human skin. Dr. de Boer, who is an Assistant Professor in the Department of Dermatology, refers to his approach as multi-functional OCT because he combines multiple detection techniques, such as polarization and Doppler, in a single device to produce images. “Polarization sensitivity in OCT imaging offers promise because there are many tissues in skin that are actually birefringent, including collagen, nerves, muscles and tendons,” said Dr. de Boer. “Doppler is of interest because it allows us to map the vascular structure and qualitatively and quantitatively determine the flow.” OCT uses a broadband light source that is split into a sample arm reflecting from the tissue to be scanned and a reference arm reflecting from a moving mirror. The detector images a series of constructive and destructive combinations of the reflected light from both sources depending on the difference of the path lengths created by the moving mirror, creating an interferogram. “Since the light penetrates beneath the sample’s surface, there are reflections from structures beneath the surface. Each reflecting layer in the sample generates an interferogram,” explained Dr. de Boer. “From the interferograms you can determine the depth of the reflection from the outer surface. By scanning the sample, the detector builds an image of the sample’s interior morphology.” Dr. de Boer hopes to separate and identify various elements of the skin based on the characteristics of the light and produce detailed, fine-grain dimensional images of the sub-surface structure and determine the blood flow through such pathology as burn scars, basal cell carcinoma, sclerotic keratosis and cherry angioma. These images will allow physicians to better determine the extent and organization of skin conditions allowing for better diagnosis and treatment. “Multi-functional OCT creates cross-sectional images 5 mm wide and up to 1.5 mm below the surface on in vivo skin in about 1 second with a resolution of about 10 micron,” says Dr. de Boer. “It gives you quantitative information on layer thickness, structure size and birefringence, and qualitative information on layer, structure, collagen and blood flow.” With this technology you could determine the depth of a burn, you could assess collagen, or you could see more in-depth detail of a basal cell carcinoma. Diffuse Optical Spectroscopy and Diffuse Optical Imaging Diffuse optical spectroscopy and diffuse optical imaging may provide a method for interrogating very large tissue volumes with a moderate resolution up to a few millimeters or several centimeters, 8 or perhaps even 10 centimeters in depth. According to Bruce Tromberg, Ph.D., Professor, Departments of Biomedical Engineering and Surgery and Director, Beckman Laser Institute, University of California, Irvine, these techniques essentially involve shining a light source through tissue and using the diffuse multiple scattered photons that penetrate deeply to obtain a detailed functional and structural image of the tissue. This technique, which Dr. Tromberg estimated will be available within 3 to 5 years, offers the promise of detecting malignant tumors in patients for whom standard detection methods aren’t effective because of factors such as the presence of benign lesions and density of surrounding tissue. “Because of this technology’s ability to detect changes induced by chemo-therapy, we feel that this technique can be used to provide a sensitive, accurate and quantitative measurement of therapeutic efficacy in breast cancer patients,” said Dr. Tromberg. This would allow practitioners to detect changes at an earlier stage to determine if chemotherapy is working for a patient. According to Dr. Tromberg, diffuse photons carry information about the tissue, including cellular proliferation, nucleation, hypoxia and inflammation. Information about the extra-cellular matrix is also transmitted, including insight regarding invasive processes, degradation of the matrix, and necrosis. Various physiological parameters can be quantified, including blood volume and flow, vascular permeability, the concentration of oxy and de-oxy hemoglobin, water and lipid in the tissue. The main challenge is separating light absorption events from light scattering events to make a quantitative measurement. By switching between different semi-conductor diode lasers, different wavelengths of light result. Modulating the amplitude of the light source generates an oscillatory light signal (a photon density wave) whose amplitude and phase depend on the tissue’s optical properties. “The integration of optical methods with conventional imaging techniques and targeted molecular probes is relatively easy to implement. Ultimately, we expect that optical methods will be used routinely to enhance the functional information content of medical images used to diagnose disease and monitor the efficacy of various therapies.” While diffuse optics have a tremendous amount of power on their own, Dr. Tromberg believes that they’re even more powerful when used in combination with other anatomic imaging techniques such as MRI, mammography and ultrasound. “If you combine these approaches with needle localization techniques and with exogenous dyes, molecular dyes that can be targeted or materials like indocyanine green, you can follow their extravasation rates and view the tissues’ biophysical characteristics.” A Small World That’s Getting Bigger Optical techniques are now under development that can provide information about the morphology and architectural structure of in vivo tissue almost instantly. Eventually, you’ll look into the skin at amazing depths and see structures in astonishing detail. The exciting news is that these new developments are not decades in the future. Some may even be available for clinical use before the end of the year.
