Eric J. Seibel - US grants

This project will develop engineering designs for an appliance that will help people with certain types of low, or impaired, vision to see. The designs will be based on the principle that engineering design is a decision-making process, not merely a problem-solving one. Thus, as an aid for the decision maker, a model will be built that defines concepts such as value, attributes, demand, time, price, and other factors, and utilizes a rational for optimization. The project will have two main parts: (1) the development of the decision-making model and its application to the design of the Low-Vision Aid (LVA) and (2) the development of LVA demonstrations that will illustrate the feasibility and importance of such a product for people with low vision. Vision enhancement will be achieved by exploiting the unique attributes of the Virtual Retinal Display (VRD), a means of safely scanning image data directly onto the retina without intermediate image formation surfaces. The VRD projects an image on the retina that is very bright, has high contrast, has highly saturated colors, and is projected into the eye via a very small exit pupil. If successful, this research will lead to the development of easy-to-use vision enhancement aids, devices that will enable the visually impaired to read printed material (e.g., books, magazines, newspapers, etc.), to watch TV, to interact with a computer monitor, and to navigate in the physical world. The impairments that this instrument will help overcome include corneal scars, keratoconus, presbyopia, and cataracts. The optical characteristics of the VRD may also prove to be an aid for macular degeneration and amblyopia. The development of engineering designs is the first step in the process of making the LVA available to people subject to these diseases. The project will also provide an example application of decision-based engineering design.

9978888
Seibel
The problem of applying advanced electronic imaging and computer display technologies to aid persons with visual disabilities has resulted in bulky and obtrusive eyewear that has been shown in the marketplace not to be worth the high price. Nonetheless, with US citizens both living and staying active longer, there is even greater need for effective, wearable, and desirable low vision aids. There are 14 million Americans that are hampered in their daily lives by partial loss of visual acuity and/or insufficient visual field (National Advisory Eye Council 1998), and that number is expected to rise much faster than the rate of population growth. A new approach is being undertaken in this study, hinging upon a new display technology, called retinal light scanning. By scanning a low power beam of light from a laser or light-emitting diode directly into the eye, a bright image is generated as the beam is modulated in power and scanned across the retina. This 'laser light show on the retina' provides not only brighter display illumination than standard displays, but the narrow beam provides unprecedented depth of focus. Reading speed studies conducted in our laboratory have shown that retinal light scanning has great potential for transferring images to the retina with minimal distortion. In concert with the recent development of retinal light scanning, technological improvements have been made in micro-cameras, wearable computers, real-time image processing, and spectacle-mounted hardware. Our approach is to reduce the size and cost associated with retinal light scanning which will be redesigned specifically as a wearable low vision aid. We will incorporate the latest advancements by our collaborators in the areas of cameras, wearable computers, and image processing. These technologies will be integrated into prototype vision aids that will be a testbed for our newest theories of human-computer interface science, such as spectacles that have an augmented central view from retinal light scanning while allowing natural viewing of the surroundings to maintain situational awareness. In collaboration with the Department of Services for the Blind in Seattle, WA and the Schepens Eye Research Institute in Boston, MA, we will test the prototype low vision aids and their performance in the hands of low vision volunteers doing everyday tasks. Thus, we believe the same problems that have led to a lack of acceptance of previous high-tech vision aids can be overcome by using retinal light scanning technology as a basis for our research effort. From our experience, the goals of designing and testing novel vision aids for the partially sighted in our community is highly motivating and rewarding work for both undergraduate and graduate science and engineering students.

DESCRIPTION (provided by applicant): Catheter Scope for Intraluminal Imaging of Early Neoplasia A micro-optical scanner has been demonstrated to efficiently transmit high-quality laser illumination at high resolution across a wide and variable field-of-view. The scanner is a microfabricated optical fiber that is driven in vibratory resonance. Since the laser light is scanned, images are acquired one pixel at a time, and most importantly, only a single optical fiber is required for illumination. This single optical fiber and a few collection fibers (either optical fibers or electrical wires from optical detectors) can be contained in an ultrathin catheter-style package of less than 2.5 mm in diameter. Therefore, high quality images can be obtained in regions of the body that were previously inaccessible. We propose to fabricate, test, and develop a proof-of-concept catheter scope in the first year (R21 phase), and then to build and test in vivo more advanced prototypes in the subsequent 3 years (R33 phase). The ultrathin catheter scope will be designed to fit within the 2.8 mm diameter working channel of a standard GI endoscope or standard flexible bronchoscope. Immediate goals are to determine feasibility of using this new in vivo imaging technology for the early detection of cancer in the pancreas and peripheral lung. The unique features of the prototype catheter scope are its high flexibility, small diameter, variable resolution imaging, enhanced depth perception using stereo-pair detectors, and enhanced image contrast using laser-induced fluorescence, and polarization contrast of the epithelial tissue layers. The long-term goals are to bring laser scanning endoscopy and bronchoscopy to the forefront of minimally-invasive medical practice, as a tool to image remote locations in the body, screen and diagnose for early detection of cancer, and in the future deliver optical therapies with pixel accuracy.

0421579
Seibel
The investigators are planning to build a new research instruments for a true 3D display to view high resolution 3D data with all depth cues in agreement. Current imaging systems suffer from the accommodation-vergence mismatch limitation. The proposed system will provide variable ocular accommodation cues that match vergence and stereoscopic retinal disparity demands for objects at different distances in a 3D scene. The proposed instrument promises to reduce eye fatigue and improve the quality of the 3-D images.

DESCRIPTION (provided by applicant) Transbronchial needle aspiration and cytological brush sampling of the bronchi are the methods of choice for early diagnosis of lung cancer. However, sampling lesions in the more peripheral lung have low probability of diagnostic yield (sensitivity) because the tissue cannot be clearly imaged during the biopsy procedure. The major problem is the lack of an ultrathin and flexible bronchoscope that maintains high-resolution imaging while providing a biopsy channel. A new microfabrication technique has been developed that micromachines a single, commercial optical fiber into a resonant cantilever that produces wide a field-of-view of directed laser illumination for low-cost in vivo imaging. The fiberoptic scanner can be fit within a 2 mm diameter, so that a flexible scope resembling a catheter can be constructed. This catheter-like scope can be used for imaging regions of the human body that have been inaccessible to larger, less flexible, and more expensive scopes based on coherent fiberoptic bundles that do not have room for a biopsy channel. The Pentax Corporation has an option to license this fiber-scanning catheterscope technology from the University of Washington, and a joint partnership has been formed to develop this technology to a prototype the device and also to prototype manufacturing processes that will lead to an in vivo imaging instrument with a disposable distal end. The goals of this project are to (I) construct a 2.0 mm diameter prototype catheterscope with a biopsy channel that can pass a 22-gauge needle, (2) use low-cost components and develop high-volume manufacturing processes so the total cost of the distal end is <$100, (3) test the prototype for image-guided biopsy using an in vitro model and an appropriate animal model, and (4) facilitate technology transfer from the academic inventors to the medical device manufacturer so that the successful diagnosis of cancer in the peripheral lung can be significantly improved while healthcare costs are reduced. In future clinical practice, the catheterscope will be introduced into the peripheral lung through the biopsy channel of a larger bronchoscope, and possibly used without the aid of fluoroscopy, further reducing the cost of lung cancer diagnosis. Due to the directed laser illumination, the fiber-scanning catheterscope can be used in conjunction with laser-based diagnoses and therapies, such as laser-induced intrinsic fluorescence at 405 nm for improved selection of biopsy sites, Applications in the future are the diagnosis of pancreatic, breast, bladder, and other lumenal cancers.

The main goal of this project is to extend the capabilities of ERCP with high-resolution optical imaging while making the insertion and examination technique as minimally invasive as possible. In addition to high-quality and high-resolution imaging in a miniature instrument, these advances will be engineered into a single-operator system with 4-way tip deflection and new accessory devices. This project will develop the tip bending mechanism without sacrificing size, cost, and performance of the proposed mini-pancreatoscope (MP). The new MP will be tested against current ERCP procedures by gastroenterologists using a synthetic model for ERCP training that contains force sensors.

Because a majority of cancers originate in the epithelium, the development of a minimally invasive laser therapy endoscope is proposed for the treatment of early cancer and precancerous lesions. During treatment for bladder cancer, many early cancers go undiagnosed, resulting in the highest recurrence rate of any cancer. A new scanning fiber endoscope will be developed for integrated laser imaging, early tumor identification, staging, and treatment, using topically applied photosensitizer dyes. To provide accurate control of the laser treatment, the same micro-optical fiber scanner is used for both in vivo imaging and laser therapy. This dual functionality will insure pixel-accurate delivery of the high-intensity laser light. The fiber scanner is located at the distal tip of an ultrathin (1.2 mm outer diameter) and flexible endoscope. Initial testing will be conducted on living artificial tissue models and a rat bladder cancer model that have been seeded with cancer cells from culture. The resulting superficial tumors within the epithelium will be destroyed using fluorescence image-guided laser therapy at single pixel accuracy. Performance evaluations of two therapeutic laser wavelengths and optical cancer indicators will be used to choose the most effective system, taking into account efficiency and safety. All systems under testing will have a high performance versus low clinical cost in practice. Furthermore, there is broad application of this technology to earlier treatments among lung, colorectal, esophageal, pancreatic, and bile duct cancers. Due to the reduced length of this project there will be no development of an interactive computer interface that estimates dosimetry of the image-guided intervention under real-time feedback control for the purpose of minimizing collateral damage to the tissue.

DESCRIPTION (provided by applicant): This application addresses broad Challenge Area (06) Enabling Technologies and specific Challenge Topic, 06-DK-103: Enabling Technologies in Imaging. Determination between benign and malignant strictures in the bile ducts is difficult. Although non-invasive imaging modalities can locate the stricture, a definitive diagnosis cannot be made without an optical measurement of cells and tissue as no biomarker exists. Peroral cholangioscopy is the least invasive method to optically image the bile ducts but to reach a stricture requires a long, flexible, and small-diameter endoscope. The current technology of an imaging sensor or optical fiber for each pixel in the acquired image does not allow both ultrathin and high-resolution endoscopy, making the clinician legally blind when imaging the bile duct. A new technology is proposed that puts a small microscanner at the tip of a catheter, making a rigid tip of only 9-mm and a rugged flexible shaft with a 6-mm bending radius. Previously this catheterscope with distal-tip microscanner can acquire 600-line color images at 30 Hz (video rate) and bile duct imaging has been performed in a live pig. In this proposal, the addition of tip bending and a doubling of the optical resolution to HDTV (1024 lines) will be developed for a new steerable "guidewire with eyes". The proposed new guidewire with eyes will be 1.2 to 1.6 mm in diameter, approximately twice the diameter of a standard guidewire. New cannula-style brush and forceps tools will be developed for image-guided biopsy and tested first in pig bile ducts and then in at least 20 human subjects, split between having benign and malignant causes of the stricture. Since the 1-mm catheterscope may be able to traverse the stricture, staging or lateral extent of any suspected carcinoma will be measured. Additional clinical advantages of this scanning catheterscope are that the low-power, red, green, and blue laser light is inherently narrow band and both enhanced spectral imaging and zoom magnification are user-controlled real-time features. By enhancing image contrast and feature size, biopsy site selection can be improved and no longer does the clinician have to be legally blind during an endoscopic retrograde cholangiography procedure. In the future, when bile duct surveillance is more common, the laser illumination can be changed to measure fluorescence biomarker signatures of early cancer and pre-cancer without adding size, cost, or complexity to the probe. PUBLIC HEALTH RELEVANCE: Determining diagnosis and extent of disease in the bile duct is extremely difficult, and with current endoscope technology these decisions are being made with clinicians legally blind. Therefore, a new laser scanning technology for image-guided diagnosis of disease in the bile duct is proposed that affords high quality video imaging. This new technology will be developed and tested for the purpose of improving the efficiency and accuracy of diagnosing bile duct diseases, especially cancer from benign complications that can mimic this deadly disease. The goal of this project is to provide biopsies with higher accuracy and diagnostic yields, while introducing a less invasive clinical instrument. Having a more effective and less invasive clinical instrument is expected to lower overall cost for managing patients with indeterminate biliary strictures.

DESCRIPTION (provided by applicant): The diagnosis of disease and surgical removal of diseased tissue relies on optical imaging that the trained human observer can clearly distinguish between healthy and normal tissue. Currently stains and biomarkers are used that help differentiate both the chemistry and morphology of cellular and tissue structures. However, in vivo use of these stains, biomarkers, and nanoparticles have safety and regulatory concerns for routine clinical practice and disease diagnosis is performed only outside the body. Nonetheless, there is medical need to obtain this molecularly- specific high-resolution imaging inside the living human body. Coherent Raman scattering is a promising technology that offers label-free chemical contrast with sub-cellular image resolution at video rates. This Raman signal is generated by near-infrared laser light that probes the chemical bonds within cells while not appearing to damage living tissue. However, traditional endoscopes are not based on laser scanning and thus not suitable for taking this advanced imaging into the clinic. An alternative endoscope technology, compatible with coherent Raman scattering, such as coherent anti-Stokes Raman scattering (CARS), uses a scanning optical fiber that scans laser light to form images at video rates. This technology produces high-quality laser-based images from an ultrathin and flexible endoscope. In this project, specific design modifications are being made to produce the first ultrathin and flexible endoscope for coherent Raman scattering at image acquisition rates that are clinically acceptable for endoscopic tissue diagnosis and surgery. Specialty illumination optical fiber and micro-optical lens assemblies will be custom designed, fabricated, and tested on tissue. The performance of this new endoscope will be compared against a set of quantitative milestones, and for direct comparison to histology. By the end of this two year project, the ability to image living tissue with sub-cellular spatial and chemical resolution without biomarkers will be established. Future uses of such technology will be the direct and rapid assessment of tumor margins and other diseased tissue without having to take repeated biopsies during surgery and endoscopic examination. PUBLIC HEALTH RELEVANCE: A new type of mini-endoscope will be designed and developed that generated video images with specific molecules highlighted without adding any drug, chemical, or particle into the living system. The technique uses scanned laser light that is not absorbed by the cells, but can help determine the disease state of tissue without taking out any tissue for pathology.

1014976
Seibel

A microscopic tool being developed by this team will allow quantitative absorption and fluorescence 3D imaging of cells and their organelles which can be correlated to the clinical standards of disease diagnosis for the first time. This project develops new fields of 3D quantitative microscopy in two important contrast modes that bridge a gap between clinical diagnosis and medical research. In addition, the technology of 3D image processing from quantitative CT imaging for disease diagnosis is being applied to microscopy in a unique collaboration.

Textbooks of cell biology depict a cell as flat with two-dimensional features. If disease diagnosis is 3D and more quantitative and sensitive to disease onset, then these textbooks will likely change to more accurately represent the cell anatomy. Teaching students on cellular structures and functions may be significantly changed due to advancements of true 3D microscopy. Providing cytologists and molecular biologists with quantification tools will result in earlier diagnosis of lethal diseases like lung cancer and improved understanding of the disease mechanisms.

DESCRIPTION (provided by applicant): The broad, long-term objective of this project is to develop a multi-institutional, multi-disciplinary translational Research Center to accelerate the clinical use of targeted imaging methodologies for the early detection and prevention of esophageal adenocarcinoma through the Barrett's Esophagus Translational Research Network (BETRNet). 3 inter-related Primary Research Projects and Shared Research Resources (Cores) will be established to develop advanced endoscopic imaging as an enabling technology for management of patients with Barrett's esophagus by visualizing molecular targets from amplified and overexpressed genes. In Project 1, genomic data will be analyzed to identify molecular targets based on properties of gene amplification and/or overexpression. These targets will be used to select highly specific peptides in Project 2 that will be fluorescent-labeled for detection on imaging. A multi-spectral scanning fiber endoscope will be developed in Project 3 to perform real time visualization of a panel of peptides in 3 fluorescence channels. The unique instrument design can pass through the working channel of a standard medical endoscope to allow for concurrent white light imaging. Future versions of this instrument may have 12 or more channels. Image processing algorithms will developed to red-flag diseased regions on a panoramic view of the distal esophagus. This integrated imaging strategy aims to assist the physician in guiding tissue biopsy of high grade dysplasia and early adenocarcinoma that can not been seen otherwise. Phase 1 studies will be performed to prepare for a future multi-center clinical validation study. Future applications will allow for real time spatial visualization and monitoring of other targets, including those associated with stem cells or with biomarkers of cancer risk. Moreover, imaging of molecular targets can be performed in patients with Barrett's esophagus over time to better understand the molecular mechanisms of this disease. The 3 Cores will support the Primary Research Projects, Pilot Projects, and Cross-BETRNet Projects. The Administrative Core will provide support for grants and regulatory management, including budgets, financial reporting, progress reports, human subjects protocols, investigation of new drug (IND) applications, and Steering Committee representation. The Bioinformatics Core will provide support the Primary Projects for statistical evaluation of genomic, imaging, and clinical data and the BETRNet Network for data sharing and archiving of tissue specimens. The Validation & Pathology Core will provide support for validation of peptide binding activity to amplified and overexpressed gene targets in high-grade dysplasia and early adenocarcinoma.

A multi-spectral scanning fiber endoscope (SFE) will be developed to image the panel of fluorescent-labeled peptides developed in Project 2 that have specific binding activity to amplified and/or over expressed gene targets identified in Project 1. Image processing algorithms will be developed to stitch together individual images to form a panorama of the esophageal mucosal surface, and red-flag idenfificafion of highgrade dysplasia and early adenocarcinoma will be performed in Core B to assist the physician by guiding biopsy. The performance of this integrated imaging strategy will be validated in Core C. This instrument will pass through the working channel of a medical endoscope in Phase 1 clinical studies of the individual pepfides assigned to three fiuorescent-labels in Project 2. The expected establishment of safety will prepare this system for a future multi-center clinical validation study.

DESCRIPTION (provided by applicant): Complete resection of tumor tissue remains one of the most important factors for survival in patients with cancer. Surgical removal is the most common front-line cancer therapy. Tumor resection in the brain is exceptionally difficult because leaving residual tumor tissue leads to decreased survival and removing normal healthy brain tissue leads to life-long neurological deficits. Brain surgery requires a very high degree of dexterity, accurate navigation, and niicro-precision cutting over long durations; thus it is an idel candidate for robotically assisted surgery. However, tumor resection is compounded by the need to make a small opening (keyhole) in the skull, and the difficulty of distinguishing normal from diseased tissue in an intraoperative setting. A minimally-invasive robotic system that allows surgeons to directly visualize and accurately discriminate neoplastic (cancer) from non-neoplastic tissue in a real-time intra-operative setting is currently not available, but an ideal gal for NRI. We propose to overcome two major limitations affecting robotically-assisted surgery in a team approach: inability to (1) automatically and (2) optically guide treatments in a miriimally-invasive intraoperative environment with advanced photonics and new cancer biomarkers. Dr. Hannaford will lead the integration of the RAVEN II open hardware and softvvare robotic system with laser-based endoscopic imaging. A single robot arm will hold a standard surgical tool for resecting/removing tumor, and a novel scanning fluorescence and reflectance imaging system to provide the advanced photonics in an ultra-small size. A team of three research neurosurgeons (Drs. Olson, Ellenbogen, Sekhar) will help develop clinically relevant phantoms and biological models of future image-guided brain surgery. PI, Dr. Seibel will provide a new multi-modal scanning fiber endoscope (mmSFE) technology that allows advanced laser imaging, diagnostic, and therapeutic biophotonics approaches to intra-operative keyhole surgery for improved performance and safety.

1212540
Seibel

A new method for processing and imaging of thin-needle core biopsy (TNCB) of resected tissue containing pancreatic cancer (animal and human) is proposed. TNCB specimens of 0.25mm diameter will be obtained by a custom-designed coring device, while larger 0.40mm cores will be taken with a new commercial biopsy needle. Corresponding fine needle aspirate (FNA) specimens will be taken at or alongside the TNCB sample sites. While maintaining TNCB tissue within a tube or microfluidic chamber, the specimen will be fixed, stained, and optically cleared. Absorption dyes of hematoxylin and eosin (H&E) will provide transmission image contrast for the morphological structure of cells and tissue.

Fluorescence dye will be imaged from various immunohistological targets, such as proteins expressed at the base epithelial membrane to help determine invasion and protein expression. Optical projection tomographic microscopy (OPTM) will be used to generate 3D multimodal images of biopsy tissue. Although the length of the tissue core will be longer than the microscope field of view (FOV), different regions of the tissue will be imaged and stitched together by image processing. Both 2D and 3D visualization of the entire TNCB specimen will be provided to a pathologist using a custom computer interface to manipulate the 3D dataset. Sensitivity and specificity of cancer diagnosis will be made using tissue-slice pathology as the gold standard. Comparative evaluation of pancreatic cancer diagnosis will be made between TNCB imaged with the OPTM versus conventional FNA. This new technique is expected not only to detect the presence of neoplastic cells but detect invasion of tissue by cancer cells.

Abstract

#1351110
Eric J. Seibel


In this EAGER project a new engineering design for an endoscopic medical device is proposed. This new approach will allow a working channel to be included within an ultrathin and flexible endoscope of 1.7-mm diameter. Instead of shrinking the conventional endoscope of separate channels for illumination, imaging, flushing, and biopsy; the new approach will place an ultrasmall imager at the tip of a working channel that provides flushing and biopsy. This new approach is transformative for clinical imaging and critically helpful for accelerating the development of optical biopsy. The working channel will provide the key requirements of fluid exchange and tissue vacuum, biopsy, and retrieval. Testing will involve the biopsy apparatus, which is suction of the epithelial mucosa into a cutting chamber that is connected to the proximal end for biopsy retrieval. The biopsy apparatus will be modified in an iterative manner using test results from a synthetic phantom and ex vivo pig lung. The integrated microendoscope and biopsy apparatus will be fabricated and tested for endoscopic guidance to specific sites for biopsy. Fluorescently labeled cells and microspheres will be injected into the luminal wall so that accuracy as well as adequacy of the biopsy will be the quantitative milestones.

DESCRIPTION (provided by applicant): There are over three million needle biopsies performed every year in the USA, typically for the diagnosis of breast, prostate, thyroid, lung, liver, and pancreatic cancers. Because needle biopsies are less invasive than surgical biopsies, they are much more cost-effective. However, a major limitation for all needle biopsies are that the pathologist or cytologist views the tissue or cells in only two dimensions, while the important structures for determining cancer and invasiveness of the disease is more clearly seen in threes dimensions (3D) versus two dimensions. What is true in radiology is expected in pathology, 3D imaging will give a quantum increase in the information content, expected to increase sensitivity and specificity of early cancer diagnosis. For many cases such as suspected pancreatic cancer, the small needle biopsy is all the tissue that the clinical team has to test before making a lifesaving decision. In this proposal, the preferred specimen for pancreatic cancer diagnosis is a thin needle core biopsy, which the entire 1-2cm long specimen is visualized in 3D using advanced optical imaging. After needle procurement from ex vivo pancreas, all handling and preparation for highest quality 3D imaging will be prototyped and validated within a novel microfluidics device. Every tissue preparatory step of fixing, staining, clearing, and transporting to 3D imaging is conducted within the microfluidics channel, which can be automated. The goal of this project is to reduce the time for biopsy tissue preparation to less than half the standard time in a pathology lab. Furthermore, the 3D architecture of the tissue will be preserved for the first opto-mechanical measures of biopsy intactness. The diagnostic value of the needle biopsy specimens after automated preparation will be determined by experienced pathologists. The ability to visualize 3D morphology within tissue biopsies will also allow the pathologists to bette compare ex vivo diagnosis with new in vivo 3D imaging. These ex vivo technologies that rely on optically cleared tissue and traditional bright-field imaging (Z-stack imaging and optical projection tomographic microscopy) can become the pathologists' bridge for better understanding in vivo 3D imaging technologies, such as confocal, optical coherence tomography, and photo-acoustic imaging. This project is expected to lead to a transformation in pathology from 2D to 3D, and in the ability to provide less invasive, low-cost and rapid cancer diagnosis, directly affecting several millions of US citizens per year.

? DESCRIPTION (provided by applicant): Cholangiocarcinoma (bile duct cancer) is the most common malignancy of the biliary tract, and is increasing in incidence and mortality worldwide. It is usually diagnosed at an advanced stage, at which point the overall prognosis is poor. However, patients with early stage tumors who undergo resection have excellent outcomes. Conversely, lesions of biliary intraepithelial neoplasia (BilIN), a type of cholangiocarcinoma, frequently appear as indeterminate biliary strictures, and because of the difficulty obtaining enough human tissue for histological analysis, clinical diagnosis is often inconclusive. As a result, a large fraction of patients undergo major surgery only to find that the strictures are benign, or wait too long to have curative surgery. The goal of the collaboration is to develop an image-based diagnostic method that uses fluorescently stained probes and a novel endoscopy device to map the probes. With this method, the time to cancer detection could be reduced and the accuracy of diagnosis improved in the following ways: Quantitative mapping of biomarkers of the biliary epithelium is expected to enable detection of molecular changes before structural changes are apparent - it should improve the ability to distinguish cancer from nonlethal disease, and it may be able to identify molecular targets for cancer therapy. High-contrast fluorescence molecular probes can also be used to image the entire indeterminate biliary stricture and can be used to guide biopsy. Because the molecular biology of cholangiocarcinoma is highly heterogeneous, a method of detecting multiple molecular targets is needed. The researchers will use multiple probes known to bind with different cancer targets and, from an existing prototype, will develop a multimodal multispectral scanning fiber endoscope (mmSFE) capable of wide-field imaging of the fluorescence probes. The probes will be optimized and validated, and the mmSFE will be tested with the validated probes in vivo in mice. After the mouse studies, the mmSFE will be tested in pilot human subject trials using topical application of the fluorescence molecular probes and minimally invasive endoscopy. Quantitative tests will also be performed with ex vivo human tissue and phantom pancreatobiliary organs. In vivo and ex vivo images will be compared with pathology findings in order to develop image-based diagnostic tools that can be used alone or combined with tissue and cell samples. Recent research indicates that methods developed for early detection of biliary cancer will be applicable to cancers of other pancreatobiliary organs, specifically cancer of the pancreas, and the fourth leading cause of cancer deaths in the United States. This project has the long-term potential to reduce late-stage diagnoses of pancreatobiliary cancer, thereby improving prognoses, as well as to reduce overdiagnosis of indeterminate biliary strictures, thereby preventing unnecessary major surgery and trauma.

? DESCRIPTION (provided by applicant): Dental caries is one of the most common chronic infectious diseases in the world. The occlusal (biting surface) and interproximal (between teeth) areas are where there need for quantitative imaging of tooth decay and depth of cracks in the USA. In addition a majority of the caries load in USA is carried by different groups, such as Alaska Native people who have many times the national rate of tooth decay. Health care disparities in this population are exacerbated by rural remoteness. The average number of dental fillings per Alaskan Native person is double that of urban dwellers. To provide safe, high-quality dental care in these remote areas, an x- ray replacement is desired which provides high-resolution optical images of cracks and decay deep within teeth that does not require ionizing x-ray radiation. The current problem is that the optimal wavelengths of infrared light (1310 and 1460 nm) that can be used to see decay in teeth, do not have small high-resolution cameras for pediatric clinical use. Thus, an alternative method of forming infrared images will be developed and tested that produces the highest contrast of the lesion, which is scanning laser light either i transmission, reflection, and in a right-angle configuration (for example, scan the occlusal surface and detect above the gum line). The smallest laser-based imaging device has been developed at the University of Washington (UW), being the diameter of a round toothpick (1.2-mm) while producing high-quality video images at multiple wavelengths. The scanning fiber endoscope (SFE) technology has been tested by Dr. Joel Berg in children at The Center for Pediatric Dentistry with shared administration from UW Pediatric Dentistry and Children's Hospital, Seattle, WA. Although this device used visible light for testing the health of the enamel the change to infrared is straightforward, allowing much deeper imaging of cracks and carious lesions using optimal laser wavelengths for high-contrast transmittance and reflectance imaging. The small size of the scanner allows the two imaging modes to run concurrently if needed to find the extent of a crack, maximum depth of caries, and even total volume of caries under surveillance with non-surgical medicinal therapies. The hand piece will be smaller than any bitewing radiographic sensor used in children now, while providing new features of imaging during restorative work, possibly under tele-medicine conditions. The SFE technology is fundamentally low in cost because it uses laser and detectors that are fully developed in the telecommunications industry. The low- power scanned laser light imaging can be made portable and carries no significant risks. This project will fabricate a prototype device, determine its optimal imaging modes (wavelength &configuration), develop new 3D imaging algorithms, test with extracted teeth and verify with a pilot study with children 6 to 12 years of age.

Caries and gum disease are especially prevalent and severe in low-income and rural communities, which often lack access to convenient and affordable dental care. Members of these communities are more likely to visit hospital emergency departments with advanced stages of oral disease that need surgical treatment; this increases the cost of dental care for families and burdens hospital resources. Tooth decay is the most common chronic disease in American children and adolescents age 6-19 despite being a preventable disease. In 2010, untreated tooth decay, or caries, affected 2.4 billion people worldwide. Early childhood caries leads to pain, infection, and discomfort. And in adults, chronic oral infections may increase the risk of preterm birth and diseases like diabetes and atherosclerosis (hardening of the arteries). This project will develop a hand-held electric-toothbrush-like device with a user-friendly interface that can be used in the home or by community health workers in schools or rural clinics. The wand, which uses a safe optical scanning method, will send data and images to a dentist, who will be able to monitor how well dental treatments are working. The data and images collected by the device can be analyzed to discover where plaque and bacteria are present, providing a way to predict, and then prevent, disease. The wand will be able to prompt the user to address problem areas. This system, which will connect remote users to a dental provider, has the potential to improve the prevention and treatment of oral disease and the quality of life for people who do not have convenient access to regular, in-person dental care.

To enable this vision, this project will develop a smart system that offers screening, surveillance, and prediction for people to improve their oral health and prevent disease. A mature technology of laser-based imaging that is spatially registered with fluorescence spectral analysis will be used to study the complexities of the oral biofilm and dental demineralization. The bacterial measurement is performed with a hand-held single wavelength optical scanning probe that forms images from both reflectance and fluorescence contrast, with the option of taking laser-induced fluorescence spectra for diagnosis of enamel demineralization. To provide a smart interface, the optical information will be analyzed to identify trends, a new modality for the dental field. With analyses of the fluorescence signal from the plaque deposits displayed as a trend, the user, in collaboration with clinician, can monitor variations in oral health and the effectiveness of treatments. The research team will use an iterative process to develop, design, evaluate, and refine the tool. The tool will be tested for three scenarios: (1) a plaque- and caries-screening program for a trained lay user on an untrained "patient" (e.g., parent for a child, school nurse for pupils, pediatrician for young children, or staff at a remote rural clinic for patients); (2) a caries surveillance program for trained lay users electronically connected to a dentist who can guide the use of the tool; and (3) a caries prediction program, initially for clinical users.

This project brings together a multidisciplinary team from the University of Washington (UW), in Seattle, Washington, with expertise in human centered design, engineering, sensing and machine learning, oral biology, and dentistry: Eric J. Seibel (Principal Investigator, PI), Mechanical Engineering; Sean Munson (co-PI), Human Centered Design & Engineering; Shwetak Patel, Computer Science & Engineering; Zheng Xu, DDS, School of Dentistry, Pediatric Dentistry; Jeff McLean, School of Dentistry, Periodontics. Lead industry partner Water Pik, Inc. (Fort Collins, CO), brings experience and knowledge about building, marketing, and distributing advanced dental devices to consumers: Deborah Lyle, Director of Professional & Clinical Affairs, and Jay McCulloch, Vice President, Global Marketing, Oral Care. Broader context partners are UW CoMotion (Seattle, WA), a technology transfer partner; QualComm Tech, Inc. (San Diego, CA), a wireless technology leader; Open Photonics, Inc. (Winter Park, FL), a business partner; and the Alaska Native Tribal Health Consortium (Anchorage, AK), a non-profit community partner interested in the development of an optical diagnosis device for use with its rural beneficiaries.

This award is partially supported by funds from the Directorate for Computer and Information Science and Engineering (CISE), Division of Computer and Network Systems (CNS).

- Project Summary/Abstract Recently, the incidence of esophageal (EAC) and gastro-esophageal junction (GEJAC) adenocarcinoma has increased dramatically, and have a poor 5-year survival rate of less than 15%. When detected early, these patients can have a good clinical outcome following surgery. These observations underscore the importance of early cancer detection. Patients with Barrett's esophagus (BE) are known to be at increased risk. Our overarching goal is to advance new methods of imaging to visualize the effects of spatial distribution of genetic alterations in BE by using novel imaging methods to evaluate tumor heterogeneity on the progression toward EAC. We propose a multi-institutional, trans-disciplinary, translational Research Center in the Barrett's Esophagus Translational Research Network (BETRNet). Our mission is to build on our expertise in genomic characterization, peptide biochemistry, and clinical translation to achieve our ultimate goal to perform early cancer detection at an early stage where therapeutic intervention can be most effective. We will identify a complementary panel of genes that are overexpressed on the cell surface and will be used to develop and validate new peptide imaging agents. The targets chosen will address 3 important clinical needs: 1) Real-time endoscopic identification of pre-malignant lesions and early stage cancer to guide endoscopic resection; 2) Risk stratification of BE patients for timing of endoscopic surveillance; and 3) Detection of gastro- esophageal junction adenocarcinomas in patients without endoscopic appearance of BE. We will use state-of-the-art genomic tools to to identify early overexpressed gene targets that arise in progression of BE to EAC by providing comprehensive analyses of gene expression alterations, DNA copy number variation, and genetic mutations. We will select candidate genes that are expressed on the cell surface where they can be endoscopically imaged in vivo. We will rigorously validate the panel of candidate targets with quantitative RT-PCR and immunohistochemistry on tissue microarrays using an independent cohort of human esophagus specimens. We will use these targets to first identify and validate monomer peptides that are highly specific. We will then arrange monomer peptides in a dimer configuration to produce multivalent ligand target interactions to improve binding performance and allow for early targets to be detected at low levels of expression. We will use a flexible fiber multi-spectral endoscope that can pass through the working channel of a standard medical endoscope to detect multiple targets at the same time. Successful completion of these aims will provide an integrated multi-spectral imaging methodology to longitudinally visualize overexpressed molecular targets that drive progression of Barrett's esophagus to esophageal adenocarcinoma. This innovative approach can serve as the foundation for validated preventive measures to improve patient management.

Abstract ? The Pilot Projects Core will support Pilot Projects to be performed within the individual institutions of our Research Center and Cross-BETRNet Projects with investigators at the other Research Center. Outreach to outside investigators will be performed to increase awareness of the program. This Core will support innovative ideas that advance the overall mission of the Research Center through Pilot Projects to be performed at our own institutions. This Core will also establish collaborative interactions with investigators from other BETRNet Research Centers to achieve synergy and identify new directions for research in BE and EAC. This Core will also reach out to outside investigators and leverage additional expertise and resources to advance the mission of the BETRNet program.

Project Summary Currently there are no standard guidelines followed for core needle biopsy (CNB) acquisition and preparation. Millions of CNBs are procured annually as the preferred minimally invasive procedure for diagnosing breast cancer and assessing therapeutic strategies based on biomarkers from these small biospecimens. However, the downstream steps in pathology were originally not designed for CNBs ? a series of manual procedures that are not standardized. These are: adequacy testing, fixation & dehydration, paraffin embedding, thin-sectioning, staining for image-based morphological analysis, immuno-histochemical (IHC) analysis, and subsequent microdissection to enrich for RNA extraction. Our project is designed to rapidly evaluate the CNB for adequacy, the presence of cancer cells to make a diagnosis, using CoreView millifluidic device from the University of Washington, Seattle. Automated computer-controlled fluidic pumps help to remove the CNB from the needle, transport, clean and stain the tissue surface in a repeatable system designed for CNBs. In partnership with the University of California at Davis, the new method of microscopy with ultraviolet surface excitation will be used which rapidly produces images appear to match the standard H&E images used for adequacy testing. In addition, MUSE will be expanded to include multiplexed molecular of the same CNB. High clinical impact for personalized medicine is provided by this multiplexed fluorescence immunostaining of the biopsy to determine ER/PR and HER2 expression levels that are co-registered with the virtual H&E. The entire process of multiplexed MUSE imaging at the point of care allows the specimen to remain unfixed during this process, which triages one of the CNBs into genomic and other ?omics analyses in the future that require high yield and quality of biomarkers. The other 1 or 2 CNBs can continue into formalin fixed paraffin embedded conventional histopathology work flow for the future validation of this rapid assessment system. In this feasibility study, the first molecular and morphological CNB instrumentation will be designed and fabricated based on CoreView-MUSE patented technologies. Testing will compare the CNB surface images compared to conventional thin section H&E and IHC as well as the quality and quantity of RNA from these two approaches. Pilot testing of full automation will determine the ultimate speed of processing from needle to multimodal adequacy determination that should range from 5 to 15 minutes at the point of care. Once developed and validated for breast cancer, the CoreView-MUSE system will be expanded to lung, pancreas, and other cancers that rely on CNB accurately sampling cancer from the lesion and providing phenotype classifications for precision therapies.

Intellectual Merit:
There is an increased need for ventilators for patients suffering from the current SARS-CoV-2 pandemic. This project aims to reduce the burden of ventilator associated pneumonia on hospitals that will lead to an overall reduction in the length of required mechanical ventilation. VAP is a worldwide problem that penetrates all echelons of intensive care units throughout the world, and biofilms are regarded as a significant source of this persistent problem. This research project proposes to build and investigate a novel device for the prevention of VAP by reducing or eliminating the burden of biofilm formation on the inside of endotracheal tubes (ETT). The proposed device has the potential to destroy bacteria before they have a chance to form biofilms on the inner lumen of endotracheal tubes), reducing the incidence of this significant burden on healthcare systems across the globe in a practical and cost efficient manner. Specifically, the device will be used to reduce colonization and biofilm formation on the inner lumen of ETTs by regular application of deep ultraviolet (UVC) radiation via a thin catheter to be fed down the ETT. This portable handheld device would be used by respiratory therapists and/or bedside nurses as part of routine care of mechanically ventilated patients. Short wavelength UVC would not penetrate out and through the ETT, thus sparing the oropharyngeal and respiratory mucosa from radiation exposure.

Broader Impact:
The risk of developing VAP is 2-16 per 1,000 ventilator-days, and a diagnosis of VAP has been associated with an additional cost of $40,000 per patient. If the proposed device reduces the burden of VAP, its adoption into regular ICU care could save thousands of lives and billions of dollars per year. The proposed research will further the body of knowledge on the timeline by which biofilms form inside ETTs and will also contribute to our knowledge of antibiotic-free decontamination strategies, especially those for use at the bedside of critically ill patients. Additionally, the research will contribute strongly to the present body of knowledge on diffuse UVC radiation and its safety for use inside standard ETTs in healthy humans. This work will also contribute to the knowledge of antibiotic-free decontamination strategies, especially those for use at the bedside of critically ill patients.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Angioscopes are ultrathin and flexible forward-viewing optical imaging devices that guide clinical procedures in the cardiovascular system. Cardiovascular disease, led by heart attack and stroke, are the leading cause of death in the US and globally. Due to basic limitations of conventional optics, these angioscopes are currently made with a bundle of over a thousand glass optical fibers, a 50-year-old technology that provides resolution that is too low and a stiffness that is too high for important potential applications. To reach clinically significant targets in the brain and heart, the angioscope needs to be more flexible and the rigid tip length must be reduced to only a few times the width of a human hair. Such an incredibly agile angioscope in the hands of a neurosurgeon could snake its way deep into the brain to remove blood clots, which can help a stroke patient. Further, a cardiologist could use this device to pass vessel-clogging plaque deposits and accurately apply a range of therapies in coronary arteries in response to heart attacks. The potential to reduce morbidity and mortality from stroke and heart attacks could benefit many individuals. This research project at the interface between nanophotonics and bioengineering aims to develop the technology that could enable such ultra-miniature agile angioscopes by using emerging optical hardware and artificial intelligence-enabled software image reconstruction. The project brings together scientists and engineers from academia and startup companies with medical professionals to solve this high-impact problem.

Ultrathin and flexible forward-viewing endoscopes, also known as angioscopes, are of critical importance for treating many cardiovascular diseases, including stroke and heart attacks, both of which are among the leading causes of death in the United States. Current medical instruments based on traditional refractive optics are too bulky to be used deep in the brain and in diseased coronary arteries. To reach locations of stroke in the brain, the rigid tip length in an angioscope must be reduced to sub-millimeter length scale. Emerging nanophotonics and metamaterial technology have the potential to achieve such clinically significant miniaturization. Meta-optics provide many degrees of freedom to design completely new types of optical elements. Multi-scale electromagnetic simulation coupled with optimization techniques have already enabled design of a meta-optic combining functionalities of multiple optical elements. In conjunction with a computational backend, meta-optics that also capture aberration-free images in full color should be possible. Combining computational inverse methods based on machine learning, semiconductor nanomanufacturing, and techniques from medical instrumentation, including advanced saline flushing, this project aims to create a micro-imaging system with 250-micron aperture and 100-micron rigid tip thickness, which will capture full-color images in a 100-degree field of view with cellular resolution. Along with academic researchers from basic science and engineering disciplines, this project includes partners associated with startups commercializing meta-optics and endoscopes as well as minimally invasive, interventional surgeons specializing in cardiovascular diseases.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.