This PDF file contains the front matter associated with SPIE Proceedings Volume 10868 including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

Thermoregulation is a mammalian physiological function fulfilled in large part by autonomic control of blood flow. We demonstrate the variation in hematocrit (Hct) and intravascular volume (VV) in the peripheral circulation when the external means of maintaining the initial thermal disequilibrium is removed using a PV[O]H device capable of noninvasively measuring both Hct and VV with unprecedented sensitivity, accuracy and precision on a 3 second timescale. Calibrated using an FDA approved device now in standard use for monitoring Hct during dialysis, the PV[O]H detection limit for measuring Hct variation is ±0.03 where 45% is normal. Observing the return to thermal equilibrium at 2 separate anatomic locations, we observe the return to normal homeostasis in a matter of a few minutes. Heat induced vasodilation results in an antecedent increase in plasma volume in greater proportion than for red blood cells into the dilated capillaries. At equilibrium homeostasis i.e. when there is no externally maintained thermal gradient we observe periodic fluctuations in the peripheral Hct and VV on a roughly 15 second to 1.5 minute timescale.

Lipofuscin is a fluorescent lipid/protein complex in the retinal pigment epithelium (RPE) that accumulates with age and contributes to the pathogenesis of retinal dystrophies. Therefore, quantification of lipofuscin, through fundus autofluorescence (FAF) imaging, is important in diagnosis and monitoring the progression of the diseases. However, the measured AF signal is affected by the excitation light intensity, the detector sensitivity, and the optical properties of the ocular media anterior to the RPE. We previously developed a simultaneous visible-light OCT and FAF imaging technology which is able to eliminate the pre-RPE attenuations. Further, we implemented two reference targets with known fluorescence efficiency and reflectivity in the retinal intermediate imaging plane to eliminate the effects of system fluctuations. However, we used an ND filter to reduce the fluorescence signal of the commercially available reference target to FAF level which brought the necessity of a separate reflection target for OCT. In this study, we introduce a customized reference standard target to serve as a common target for both AF and OCT. This reference target is composed of PMMA and synthesized A2E, the major fluorophore of lipofuscin. Homogenous mixtures of PMMA resist and A2E were prepared with an optimum concentration of A2E to avoid the necessity of the ND filter. A2E-PMMA solution was spin-coated on a silicon wafer. The fluorescent coating with 400nm thickness serves as the common FAF and OCT reference target. Using A2E in reference fabrication provides a similar excitation and emission spectrums to RPE lipofuscin, and thus a better reference for lipofuscin/A2E quantification.

The current study investigates the beneficial combination of optical coherence tomography (OCT) and photoacoustic microscopy (PAM) as a safe method for observing retinal and choroidal vasculature. A recent addition to the field has been the integration of gold nanoparticles (AuNPs) to provide enhanced contrast in OCT and PAM images. The improved analysis of capillaries is the result of the strong optical scattering and optical absorption of gold nanoparticles due to surface plasmon resonance. Femtosecond laser ablation created the ultra-pure colloidal gold nanoparticles, which were then capped with polyethylene glycol (PEG). The AuNPs were administered to thirteen New Zealand rabbits to determine the advantages of this technology, while also investigating the safety and biocompatibility. The study determines that the synthesized PEG-AuNPs (20.0 ± 1.5 nm) were beneficial in enhancing contrast in PAM and OCT images without demonstrating cytotoxic effects to bovine retinal endothelial cells. In living rabbits, the administered PEG-AuNPs resulted in an 82% increased signal for PAM and a 45% increased signal for OCT in the retinal and choroidal vessels. A histology and biodistribution report determined that the AuNPs had mostly accumulated in the liver and spleen. TUNEL staining and histology established that no cell injury or death in the lung, liver, kidney, spleen, heart, or eyes had occurred up to 1 week after receiving a dose of AuNP. The nanoparticle technology, therefore, provides an effective and safe method to enhance contrast in ocular imaging, resulting in improved visualization of retinal microvasculature.

Osteoarthritis (OA) is the most common form of arthritis, where the protective cartilage on the ends of bones wears down over time, causing pain, tenderness, stiffness, loss of flexibility and bone spurs. Degenerative alterations start before cartilage loss happens, which include surface swelling, cartilage fibrillation, and calcification. Detecting the early degenerative alterations can assist the diagnosis of early-stage OA. In this study, two imaging modalities are applied on human hip-joint specimens in ex vivo imaging, including polarization-sensitive optical coherence tomography (PS-OCT) and multiphoton microscopy (MPM). OCT detects the layered tissue structure of cartilage and bone using backscattered light and PS-OCT is a functional extension of OCT. PS-OCT measures tissue birefringence which is sensitive to the orderly organization of collagen in cartilage. MPM can visualize collagen fibers with sub-cellular resolution. Complementary information about cartilage on the cellular and tissue level can be obtained by the multimodal imaging. Using the multimodal system, the variation of the thickness of the cartilage structural zones, abnormal birefringence caused by collagen alterations and fibrillation, and uneven structure resulted from calcification are imaged and quantified. The imaging results show distinctive features of degenerative alterations in the OA specimen, such as uneven tissue surface, fibrillation, and reduced birefringence. It is shown that PS-OCT has great potential in detecting early stage OA.

Bone is a complex matrix of organic and inorganic components whose composition and crystallinity depend on many factors such as age, gender, genetic make-up and health of the individual. The variation in composition can be useful as an indicator of disease: for example, osteoporitic bones present abnormally low mineral:protein and low mineral:lipid ratios as well as lower crystallinity. Raman microscopy offers a rapid method for evaluating the mineral, protein and fat composition of bone, and of selected areas within the bone, down to 2 µm. In this study of human cortical bone samples from young ( < 50 yrs) and older (> 50 yrs) males and females, we show that young males and females can be differentiated by the protein:lipid Raman band intensity ratios. The Raman band intensity ratio for mineral:protein is shown to be able to distinguish bones of different age groups: (47-53, 58-62 and 70+ yrs). The variation of microstructures within the cortical bone, visible as light/dark spots under optical microscopy, could be distinguished according to phosphate:protein and carbonate:protein rations. This ability of this technique to identify variation within bone microstructures and to classify bone samples according to age and gender from their Raman spectra provides a new tool for studies of bone disease.

While the use of nanomaterials in medical diagnostics has received increasing interest, in vivo detection of nanoparticles using optical techniques is still a challenge. Among other techniques, surface-enhanced Raman scattering (SERS)-labeled nanoparticles offer many potential applications in the field of disease diagnostics and biomedical monitoring, due to the advantages offered by SERS. We have previously developed a unique plasmonics-active nanoplatform, gold nanostars (GNS) for in vitro and in vivo multiplexed detection and diagnostics. To date conventional optical setups are typically limited in obtaining SERS signals at the sample surface, due to the strong attenuation caused by the highly scattering and absorbing tissue. Herein, we utilize spatially offset Raman spectroscopy (SORS) to overcome this depth limitation and obtain specific spectrochemical signatures of SERS-labeled nanoparticles, such as gold nanostars, beneath thick material and bone. In particular, we developed an optical setup for inverse spatially offset Raman spectroscopy to improve the robustness of the method developed. The efficacy of this method, referred to as inverse Surface-Enhanced Spatially Offset Raman Spectroscopy (SESORS) is demonstrated through the detection of layer-specific and subsurface SERS signals beneath different layers and substrates: (1) 4-mm tissue phantom, (2) 4-mm paraffin film, and (3) 5 mm bone of a macaque skull. Additionally, we show the possibility of recovering the pure SERS signal that belongs to a specific layer within a two-layer system using scaled subtraction. We will discuss the use of inverse SESORS in applications relevant to biomedical research.

Optical imaging for estimating physiological parameters, such as tissue oxygenation or blood volume fraction has been an active field of research for many years. In this context, machine learning -based approaches are gaining increasing attention in the literature. Following up on this trend, this talk will present recent progress in multispectral optical and photoacoustic image analysis using deep learning (DL). From a methodological point of view, it will focus on two challenges: (1) How to train a DL algorithm in the absence of reliable reference training data and (2) how to quantify and compensate the different types of uncertainties associated with the inference of physiological parameters. The research presented is being conducted in the scope of the European Research Council (ERC) starting grant COMBIOSCOPY.

Photoacoustic imaging utilizes light and sound to make images by transmitting laser pulses that illuminate regions of interest, which subsequently absorb the light, causing thermal expansion and the generation of sound waves that are detected with conventional ultrasound transducers. The Photoacoustic and Ultrasonic Systems Engineering (PULSE) Lab is developing novel methods that use photoacoustic imaging to guide surgeries with the ultimate goal of eliminating surgical complications caused by injury to important structures like major blood vessels and nerves that are otherwise hidden from a surgeons immediate view. This paper summarizes our recent work to learn from the physics of sound propagation in tissue and develop acoustic beamforming algorithms that improve image quality, using state-of-the-art deep learning methods. These deep learning methods hold promise for robotic tracking tasks, visualization and visual servoing of surgical tool tips, and assessment of relative distances between the surgical tool and nearby critical structures (e.g., major blood vessels and nerves that if injured will cause severe complications, paralysis, or patient death). Impacted surgeries and procedures include neurosurgery, spinal fusion surgery, hysterectomies, and biopsies.

Conference Presentation for "Quantitative frequency-domain pulse oximetry with deep neural network (DNN) real-time processing"

The literature articulates the importance of advancing novel solutions which enable clinicians to intraoperatively resolve pathological tissue from healthy tissue in situ in order to guide the accuracy and efficiency of surgical tumor resection. A method which non-invasively provides real-time delineation of cancer margins has great potential to improve clinical outcomes by accelerating surgical procedural times, ensuring complete tumor resection, and by enabling more conservative resection approaches which preserves healthy tissue. Autofluorescence lifetime imaging is a powerful technique which holds great promise in addressing this clinically unmet need. Using a custom built, fiber-optic based, multi-spectral time-resolved fluorescence spectroscopy (ms-TRFS) instrument (excitation 355 nm) applied to cancer within head & neck anatomy, our preliminary results from 13 human patients indicate that tumor vs. healthy tissue regions (confirmed via histology) can be distinguished on the basis of lifetime and intensity ratio for both in vivo (pre-resection) and ex vivo (post-resection) applications. Each of the three major ms-TRFS spectral bands demonstrate highly conserved lifetime and intensity ratio trends within specific tissue types (palate, palatine tonsil, lingual tonsil, & base of tongue) for cancerous regions when juxtaposed to neighboring healthy peripheral tissue. Current results demonstrate distinct lifetime and intensity ratio results when comparing across tissue types. Collectively, our initial data suggests that time-resolved autofluorescence could serve as a valuable tool for providing real-time intraoperative diagnosis and surgical guidance during robot-assisted cancer removal in otolaryngologic applications.

Photoacoustic tomography (PAT) has excellent sensitivity for hemoglobin and lipids, which make up much of human breast tissue. Our group has focused on intraoperative PAT applied to tissues obtained during breast-conserving surgery (BCS). In BCS, the tumor is excised with a margin of healthy tissue to ensure tumor removal. Margin detection can be difficult and re-excision surgeries are required in 10 to 25% of cases. Our first-generation intraoperative PAT system was capable of 3D imaging specimens up to 11 cm in diameter and several centimeters thick. The system used a semi-circular ring of low frequency transducers, resulting in a 2.5 mm spatial resolution. The current objective is to improve spatial resolution using higher frequency transducers. An array was constructed with 41 circular transducers positioned on two concentric circular rungs with a single point of focus. An optical window at the center allowed illumination. The array was tested with imaging phantoms consisting of written words on a clear plastic bag, 108 µm polyester monofilament arranged as parallel lines with spacing varying from 1 mm to 8 mm, and finally with porcine tissues. The array was positioned above and perpendicular to the imaging area and raster scanned. Signal averaging was implemented, and images were reconstructed with universal back projection. Image analysis demonstrated a 400 μm spatial resolution, but with low penetration depth and low sensitivity. Results suggest the transducers could improve spatial resolution of the first-generation intraoperative PAT system by 6-fold.

Sentinel lymph node biopsy is important in the early stage breast tumor resection surgery. Its results will determine if the axillary lymph node dissection (ALND) will be conducted afterwards. For locating sentinel lymph nodes, indocyanine green (ICG) has been widely used with a near infrared (NIR) camera to image its fluorescence. However, surgeons need to watch a screen beside the operating table to see the fluorescence, with their hands operating on the surgical site. We developed a navigation system that projects the invisible fluorescence back to the surgical site visibly in real-time. The system introduces a co-axial optics design to guarantee the projection accuracy. Phantom experiments are conducted to assess the projection resolution and accuracy of the system. Animal experiments with three mice show a good system performance and its preclinical feasibility. Furthermore, the system is tested in a clinical trial of ninety breast cancer patients in three hospitals in China. ICG and methylene blue (MB) is subcutaneously injected separately into the areola at 3-4 points to get both fluorescent and visible contrast, for further comparison. The navigation process is compared with a commercialized NIR imaging system. The results show a 100% detection rate of sentinel lymph nodes and a good consistency with the methylene blue and the commercialized imaging device. The experiments demonstrate good clinical feasibility of the co-axial projection system.

Accurate identification, precise dissection, and careful preservation of critical structures, such as nerves and blood vessels, are key to successful surgical outcomes. Unintended and/or unrecognized injuries to critical structures result in debilitating short- and long-term morbidity, avoidable mortality, and considerable socioeconomic and healthcare burdens. ChemImage has developed a Molecular Chemical Imaging endoscope (MCI-E) to be deployed as an intraoperative imaging device for real-time detection of key anatomical structures. MCI-E does not require the use of contrast agents, and employs visible-near infrared (vis-NIR) reflectance hyperspectral imaging. We tested the in vivo performance of MCI-E by collecting high quality vis-NIR signatures from several anatomical structures, including ureters, arteries, and veins in live pigs under general anesthesia. In this paper, we will present successful MCI-E detection of lymph node, ureter, vessels, nerve, bowel, and thyroid in background tissues under relevant in vivo conditions. If successful, integration of MCI-E into surgical procedures will enable real-time automated detection of anatomical structures during surgeries. The benefits of this capability include the opportunity of reduced surgery time, decreased patient risk, fewer repeat surgeries, and enhancement of surgeon training.

As much as 50% of thyroid procedures result in post-surgical hypoparathyroidism and consequent hypocalcemia. This can be due to accidental removal of the parathyroid glands or damage to their blood supply that renders them non-viable. The parathyroids are the body’s main organs for regulating calcium, so loss of their function will require lifelong medication to maintain normal calcium levels. Work has been done separately to address both causes of parathyroid function loss. Autofluorescence spectroscopy/imaging has been shown to be highly accurate in distinguishing parathyroid glands from other tissues in the neck, helping avoid accidental removal of parathyroid glands. Laser speckle contrast imaging (LSCI) is capable of accurately identifying parathyroid glands that have suffered vascular compromise, providing guidance on whether to transplant a parathyroid. Here, we present an instrument that combines both techniques to enable parathyroid identification and viability assessment. Additionally, we developed algorithms to automate the extraction of parathyroid viability information from speckle contrast images using information from fluorescence images. This makes the device more autonomous and speeds up the process of providing information to the surgeon. Testing on ex vivo parathyroid and thyroid specimens revealed that the algorithm performs best when the ratio of parathyroid to thyroid fluorescence is at least 1.5. The device will also be tested on patients undergoing parathyroidectomy at Vanderbilt University Medical Center. Autofluorescence data will be validated by histology to confirm parathyroid tissue, and LSCI data will be validated by ligating the blood supply to the diseased parathyroid gland in preparation for removal.

Real-time intraoperative guidance during neurosurgeries are often limited to endoscopy or microscopy, which are suboptimal at locating underlying blood vessels and nerves. Damaging these critical structures can have severe surgical complications. To overcome this challenge, we are developing a fast-tuning, multispectral photoacoustic approach to guiding neurological procedures. An ex vivo porcine sciatic nerve and caprine carotid artery perfused with whole human blood were suspended in a water bath. A spectroscopic analysis with wavelengths 690 nm to 1260 nm was performed on each specimen with a constant optical energy of 1.5 mJ/pulse and 11 mJ/pulse for a 1 mm diameter optical fiber and a 5 mmm diameter fiber bundle, respectively. The contrast and signal-to-noise ratio of each target was calculated from photoacoustic images, with wavelength-dependent contrast values and signal-to-noise ratios that ranged from 0.41 to 21.8 dB and 10.12 to 25.6 dB, respectively. In particular, the blood vessel contrast (18.2 dB) was greater than the nerve contrast (0.61 dB) when excited with 750 nm light. However, the nerve contrast (10.7 dB) was greater than the blood vessel contrast (6.6 dB) when excited with 1230 nm light. These results indicate that simultaneous visualization of major vessels and nerves requires an imaging system that exploits the unique optical absorption peaks of both hemoglobin and lipids by fast-tuning between 750 nm and 1230 nm excitation wavelengths.

Abstract Cardiac surgical patients are administered large heparin doses to prevent thrombosis during surgery. Activated clotting time (ACT), traditionally used to assess anticoagulation correlates poorly with heparin concentration and lacks information on key coagulation metrics such as fibrin polymerization (α-angle) and clot strength (MA). Here we assess the accuracy of our novel bedside optical sensor, iCoagLab, in measuring several coagulation parameters including ACT, α-angle and MA and evaluate its capability to monitor anticoagulation during cardiac surgery. iCoagLab measures anticoagulation by assessing changes in blood viscosity from intensity fluctuations of laser speckle patterns measured from a 25µL blood drop. In this study, blood samples from 18 volunteers spiked with increased concentrations of heparin (0.2-5USP/mL) and from 30 patients undergoing cardiac surgery were assessed using iCoagLab. Coagulation metrics, including, ACT, α-angle and MA, were derived and compared with corresponding results obtained from thromboelastography (TEG), ISTAT-kaolin-ACT and Hepcon-HMS-Plus-instruments. In volunteer samples, heparin dose significantly prolonged the ACT values measured by iCoagLab which correlated closely with TEG (r=0.95, p<0.0001) and ISTAT-ACT (r=0.87, p<0.0001). Both the iCoagLab and TEG showed a decrease in MA at high heparin concentration (p<0.01). Similarly, in cardiac surgical patients, iCoagLab-ACT correlated strongly with Hepcon (r=0.76 p<0.0001) and TEG-ACT (r=0.89, p<0.0001). The MA and α-angle were also significantly modulated throughout surgery (p<0.05-0.0001) similar to TEG. These studies showed that iCoagLab rapidly and accurately measured anticoagulation and global hemostasis using just a 25µL blood drop, likely opening the powerful opportunity for multifunctional coagulation monitoring at the bedside during surgery.

Precise and efficient guidance is of paramount importance for minimally invasive vascular access procedures. Ultrasound (US) imaging is commonly used in clinics for this purpose, but the visualization of medical needles and tissue targets are often challenging. Photoacoustic (PA) imaging holds potential in guiding vascular access procedures, but clinical translation of this technology has often been hindered by bulky and expensive excitation sources. In this work, potential of a portable LED-based PA and US imaging system in guiding minimally invasive vascular access procedures is demonstrated using phantom studies and in vivo measurements on human volunteers. In the first experiment, a 14-gauge medical needle was inserted into chicken breast tissue at multiple angles and US/PA images were acquired at a frame rate of 30 Hz, to study the effect of needle insertion angles on US/PA contrast. To obtain a preliminary indication about the potential of PA/US system in imaging superficial vasculature in vivo, brachial artery of a healthy volunteer was imaged in free-hand probe guidance. With the capability of providing real-time visualization of clinical metal needles and tissue targets at clinically relevant imaging depth and spatial resolution, the LED-based PA/US system used in this study holds strong potential in guiding minimally invasive peripheral vascular access procedures.

Development of (biophotonic) devices within a university hospital and the clinical implementation has been workable up till recently. An innovation starts with an idea or solution for a clinical problem during interactions between physicians and biomedical engineers. After a well documented design and development of a prototype by the R&D workshop within an university hospital, in vitro/ex-vivo feasibility tests are performed. The results are used to make a risk analysis/safety report in preparation for clinical feasibility study. A comprehensive dossier is submitted to the ethical committee for approval. Since most biophotonic devices are non invasive and/or non-contact during clinical application, it is less complicated to obtain ethical approval. Officially, devices without CE mark are not allowed in the hospital except with ethical approval for a clinical study. It will be harder to obtain all the approvals for a multi-center trial. Recently, European guidelines for CE marking of medical devices (MDD) have been replaced by the Medical Devices Regulations (MDR) which has a major impact on existing and new medical devices. Prototypes are only allowed for clinical studies if they are produced by an ISO 13485 certified R&D department. The clinical effectiveness has to be proven and CE approved medical devices need recertification within 3 years. The capacity of ethics committees and clinical studies in hospitals is limited and expensive. Furthermore, capacity of notified bodies is potentially decreasing by 50% due to the Brexit. In contrast to the less strict directions of the FDA, the introduction of the MDR in Europe makes it very challenging to obtain CE marking and the introduction of new medical devices could dramatically slow down in the next years.

The process of translating lab-built innovations into viable tools for clinical applications is complex and costly. Clearing the regulatory processes is the pivotal step that eventually enables these devices to be implemented for the intended clinical applications. Unfortunately this task could be challenging and time-consuming for unprepared academics aiming to translate their inventions/discoveries from bench to bedside. Therefore there is a vital need to educate researchers on adopting the best approach when dealing with regulatory submissions to ensure smoother translation of their respective technologies. To understand the bench to bedside pathway more clearly, we will utilize the example of the first-ever discovery of near infrared autofluorescence in parathyroid glands at Vanderbilt University. Subsequently a lab-built system was designed for label-free intraoperative parathyroid identification during thyroid and parathyroid surgeries, which was tested across 162 patients with high accuracy. Subsequently Vanderbilt University partnered with AiBiomed (Santa Barbara, California) to develop a clinical prototype called ‘PTeye’ that was user-friendly for surgeons and ready-to-use in operation rooms. The ‘PTeye’ was then evaluated across 81 patients in a single-blinded, multi-centric study that yielded 96% accuracy. Relying on this data, Vanderbilt University and AiBiomed initiated the ‘de novo’ application process with the Food and Drug Administration (FDA) for regulatory clearance of the ‘PTeye’. The ‘de novo’ approach was selected since the instrument design and intended use of ‘PTeye’ did not resemble that of any pre-existing medical devices. Following a successful review, the FDA eventually granted permission to market ‘PTeye’ as an adjunct intraoperative tool for label-free parathyroid identification.

You have a novel optical technology and have demonstrated feasibility in solving a real clinical need. Better yet, you have further matured the technology and shown early success within an academic clinical study. Now it’s time to take the leap towards commercialization to disseminate and bring your disruptive product to the masses. Successful commercialization of a Medical Device extends beyond good product development and in fact is underpinned by establishing the right Regulatory strategy. In this presentation, we will provide an overview for the rapid development of an Optical Coherence Tomography Imaging System and discuss regulatory strategy, challenges and best practices to help future entrepreneurs navigate this critical yet delicate pathway.

Barrett’s oesophagus is an acquired condition that predisposes patients to the development of oesophageal adenocarcinoma through intermediate stages of dysplasia. Early detection of dysplasia allows curative endoscopic therapy, but current standard of care surveillance achieves only around 40% sensitivity for dysplasia. Multispectral imaging (MSI) allows simultaneous collection of morphological (spatial) and biochemical (spectral) information from tissue, which can help to more effectively delineate disease. This motivated the design and construction of a compact, clinically translatable multispectral endoscope (MuSE) that can be introduced through the accessory channel of a standard gastroscope to collect multispectral images in vivo. MuSE is based around a spectrally resolved detector array (SRDA) with 9 spectral filters (8 narrow bands; average FWHM 30nm, center wavelengths 553, 587, 629, 665, 714, 749, 791, 829nm; 1 broadband; 500–850nm). The SRDA was coupled to a clinically approved 10,000-fibre endoscope (PolyScope) for imaging. Illumination was provided by sequentially by a broadband (400–750nm) and narrowband (400–480nm) source for reflectance and autofluorescence imaging respectively. Subjects due to undergo clinically indicated endoscopy with a previous diagnosis of dysplasia or early adenocarcinoma were enrolled for experimental imaging using MuSE in a pilot clinical study. Patients with clearly visible lesions were selected to allow co-registration of the image cubes with pathology of biopsies. Here, we present the results from these first-in-human tests of MuSE, including evaluation of the image quality and classification potential of the multispectral image cubes.

In this study, FTIR spectroscopy was used to evaluate the overall biochemical status of necrotic tissue areas of cutaneous squamous cell carcinoma chemically-induced on mice. FTIR hyperspectral image collected from specimen showed high correlation with the photomicrograph obtained by light microscopy, in which we were able to identify clusters associated to keratin, necrosis and regions with no tissue. Alterations in the protein content were documented in the necrotic tissue areas, indicating changes on protein conformation.

Sentinel lymph node biopsy is a primary mean of staging cancer; however, the time-intensive nature of standard pathology limits the volume of the node that can be assessed. As a result, micrometastases can be missed, which have been shown to affect treatment decisions and therefore clinical outcomes. Optical imaging offers a potential solution for improved sensitivity and larger tissue evaluation, but an understanding of optical properties is necessary because of the high scattering nature of biological tissue. Here, time-domain optical imaging and measures of transmittance are used to characterize the optical properties of porcine lymph nodes at 685 nm and 780 nm. Results demonstrated values comparable to that of other soft biological tissue (685 nm: μ_a = 0.09 ± 0.01cm^-1 , μ_s’ = 2.60 ± 0.42 cm^-1 , g = 0.95; 780 nm: μ_a = 0.24 ± 0.10cm^-1 , μ_s’ = 3.35 ± 0.14 cm^-1 , g = 0.92). Based on these coefficients, optical properties of TiO₂ were investigated so that a protocol to fabricate a lymph node tissue-mimicking phantom could be defined.

Retinal optical coherence tomography (OCT) is increasingly used for quantifying neuroaxonal damage in diseases of the central nervous system such as multiple sclerosis. High-quality OCT images are essential for accurate intraretinal segmentation and for correct quantification of retinal thickness changes. The quality of OCT images depends largely on the operator and patient compliance. Quality evaluation is time-consuming, and current OCT image quality criteria depend on the experience of the grader and are therefore subjective. The automatic graderindependent real-time feedback system for quality evaluation of retinal OCT images, AQuA, was developed to standardize quality evaluation and data accuracy. It classifies by signal quality, anatomical completeness and segmentation plausibility and has been validated by experienced graders. However, it is currently limited to OCT scans taken with one device from a single vendor. The aim of this work is to improve the capability of the AQuA quality classifier to generalize to new data, by developing a convolutional neural network (CNN), AQuANet. Moreover, this CNN may serve as a basic quality classifier, that can be adapted to specific problems by transfer learning. AQuANet is trained on A-Scan batches with quality labels automatically obtained with AQuA. Thus, a large set of training data of about 13000 A-Scan batches could be used, leading to an accuracy of 99.53%.

Accurate tumour margin detection is a crucial step in tumour resection surgeries as progression-free survival is linked to rates of complete resection. Despite this, post-surgical positive margin rates remain high for a host of cancers. While spectroscopic techniques have shown promise as highly accurate diagnostic systems, they are inherently limited by their point-based application. Current spectroscopic diagnostic implementations fail to adequately capture spatial diagnostic information, resulting in these systems operating as one-dimensional tools suboptimal for tumour margin delineation. Here we demonstrate a computer vision-based technique that captures spatial information, enabling the transformation of spectroscopic systems from one-dimensional tools to clinically-relevant two-dimensional diagnostic platforms. We show that through visual tracking of a spectroscopic probe’s location relative to the tissue, we can display spatially co-registered spectroscopic diagnoses over clinical tumour imaging data to enhance tumour margin visualisation and aid tumour resection. Our visual, marker-based tracking approach enables real-time spectroscopic diagnostics and is designed for rapid application to different spectroscopic probe modalities and geometries with robust performance under different lighting conditions and with patient movement during procedures. We demonstrate the utility of this spatial diagnostic platform using a Raman spectroscopy probe for ex vivo margin delineation, with ongoing in vivo investigations for subcutaneous xenograft tumour models in nude mice. The associated software developed for this system permits clinical-user interaction for diagnostic threshold adjustment and tumour boundary delineation, enabling clinical diagnostic control for complex tumour geometries. Our system captures essential spatial diagnostic information, transforming point-based spectroscopic systems into effective platforms for tumour delineation.

In traditional histopathology images, and emerging ex vivo microscopy techniques that generate large (gigapixel) image data, new strategies are needed to efficiently search the images to identify salient areas for human recognition or computer-aided diagnosis, to aid in rapid and efficient review workflows. One strategy is to learn from pathologists to develop model observers that can predict the most significant areas of an image for human review or application of computationally-expensive diagnostic algorithms. To get further understanding of the pathologists’ perception and cognition, we developed a custom web-based multiresolution viewer based on OpenSeadragon, which can record view coordinate, zoom level, coordinate dwell time, and coordinate scan path of users viewing images, under the assumption that the view coordinate represents the real-time visual area. We conducted experiments on two types of data with multiple reviewers: 1) traditional histopathology images of radical prostatectomy specimens, and 2) whole surface images of prostate margins collected on intact organs using structured illumination microscopy. Overall error rate, viewing pattern, saliency maps, dwell time and zoom level on fixation clusters were analyzed on normal tissue and cancer loci/positive margins. The result of the pilot experiments demonstrates that saliency maps correspond well with known areas of tumor in histopathology images and residual cancer in tumor margin images. These data will be used to predict saliency maps on new images based on low-level image features and to test accuracy vs. expert reviewers. This tool has promise for automated image dimensionality reduction and diagnosis of histopathology images.

A miniaturized non-contact laser ranging sensor for endoscopic inspection and surgery tools is demonstrated. The sensor can measure the absolute distance between the endoscopic tool and the tissue in the 0-20 mm range with an axial resolution of 5 µm based on frequency-modulated continuous wave interferometry. As a tunable laser source, a commercially available 850 nm VCSEL is wavelength-modulated via self-heating, achieving a tuning range of 2.5 nm at 2.9 kHz repetition rate. Once coupled into a single-mode fiber, the laser light is projected onto the tissue using a loosely-focusing, Φ350 μm GRIN lens located at the distal end of the probe. The spectrum of the light collected by the same GRIN lens on the return path, which encodes the tissue distance in a well-defined modulation frequency, is detected by a silicon photodetector. Due to its low-cost, small diameter, flexibility and simplicity this sensor can be integrated monolithically into an endoscope or employed as a stand-alone sensor operating through the working channel.

Many percutaneous needle-based procedures such as foetal interventions, tumor biopsies, nerve blocks, and central venous catheterizations are guided by ultrasound (US) imaging to identify the procedural target and to visualize the needle. A key challenge associated with ultrasound-guided needle insertions is accurate and efficient identification of the needle tip, as thin needles can readily stray from the imaging plane and can have poor visibility at large insertion angles. Ultrasonic tracking is a method for localising the needle tip relative to the imaging plane in real-time, using an ultrasonic transmitter or receiver integrated into the needle that is in communication with an external ultrasound imaging probe. This study had two foci. The first was to increase the sensitivity with which ultrasound reception was performed, using a custom fiber optic hydrophone with a high-finesse Fabry-Pérot cavity based ultrasound sensor. This sensor, which comprised of a polymer layer sandwiched between dielectric mirrors, was interrogated continuously during insertions into tissue. The second focus of the study was to develop a custom needle stylet into which the fiber optic hydrophone was integrated, which was fully compatible with clinical practice and which could be adapted to different needles. We tested the sensitivity of the sensorized stylet across a wide range of needle angulations, depths and insertion angles in different biological tissues. We demonstrated, for the first time, needle tip localization in ex-vivo tissues at depths beyond 6 cm and insertion angles steeper than 80°. We conclude that ultrasonic tracking with high-finesse Fabry-Pérot fiber optic hydrophone is very promising for use in clinical practice.

Treatment of aneurysms and other neurological conditions would benefit from improved guidance of wires in arteries and the resulting reduction in time required to insert catheters and other imaging or treatment devices. By making a small opening in a blood vessel in the groin/upper thigh, arm, or neck, a catheter can be inserted with a guide wire to maneuver the catheter to the desired site. The ability to navigate arteries more accurately and thus insert a catheter more quickly would reduce the probability of aneurysm rupture and thus has the potential to reduce the rate of fatality or neurological deficit. This technology can also be used to treat complex aneurysms such as wide neck bifurcation aneurysms by angulating the neck of aneurysm temporarily. Nitinol is a shape memory alloy of nickel and titanium, with a long record of biocompatibility, particularly when an oxide and/or another passivating layer is applied. Nitinol coils have been used widely as stents because the alloy can be programmed to expand at human body temperatures and solidly fix to blood vessel walls. In this work, Nitinol wire is used with a programmed heat activation above body temperature, and thus guidance can be externally controlled using resistive heating. We present results of current-controlled steering of Nitinol wire, including the programming, control, and material response to varying current levels and pulse durations. We also demonstrate the viability of a light emitting diode attached to the guidewire with additional potential for image guidance, catheter navigation, and other treatment techniques.

Current surgical microscope systems have excellent optical properties but still involve some limitations. A future fully digital surgical microscope may overcome some major limitations of typical optomechanical systems, like ergonomic restrictions or limited number of observers. Furthermore, it can leverage and provide the full potential of digital reality. To achieve this, the frontend, the reconstruction of the digital twin of the surgical scenery, as well as the backend, the 3-D visualization interface for the surgeon, need to work in real-time. To investigate the visualization chain, we developed a virtual reality environment allowing pretesting this new form of 3-D data presentation. In this study, we wanted to answer the following question: How must the visualization pipeline look like to achieve a real-time update of the 3-D digital reality scenery. With our current approach, we were able to obtain visualizations with a frame rate of 120 frames per second and a 3-D data update rate of approximately 90 datasets per second. In a further step, a first prototype of a real-time mixed-reality head mounted visualization system could be manufactured based on the knowledge gained during the virtual reality pretesting.

Preoperative planning systems (PPS) currently are very important for evidence-based medicine promotion in surgery worldwide. Design and development of such systems is the mainstream of health care provision in Russia and let to apply preventive, diagnostic and remedial technologies based on the evidenced effectiveness and safety. PPS allow to reduce planning time, increase its accuracy and to minimize medical malpractice. Modern PPS perform a geometric planning helping surgeon to make the correct decision regarding surgery management. Special PPS designed for vertebral-pelvic complex surgery allow evaluate bone and joint deformation degree, planning simultaneous implants replacement, osteotomy. The evaluation is based on 2D and 3D medical images (if available) studying. The analysis of the most demanded modern PPS (MediCad, TraumaCad, OrthoView and Surgimap) properties shows that all of these products allow to measure a massive number of orthopedic parameters: scoliosis, lordosis, sagittal balance, the angular deviation of the sacrum, the cervico-diaphyseal angle, the Hilgenreiner angle etc. However not all of indicated systems allow to automatically search such parameters as the Reimer index (percentage) and epiphyseal index. Also not all the systems have the detection of calibration devices function and Auto-Hip function. So the aim of this study is to identify and prove the sufficient number of parameters to be evaluated by the surgeon to perform correct geometric preoperative planning in vertebral and pelvic complex surgery.

The objective of this study was to the created surface of quantitative uptake value with radioactive tracer PET/CT in normal Thai brain. The surface was generated from the matrix of quantitative uptake value by MATLAB software. Data of PET/CT image was modified to High dynamic range imaging file format by MRI convert and merge together with ana75_2.mat, in this step surfacedata.mat was obtained. The surface data was taken to create a surface of quantitative uptake value to observing the distribution of radiopharmaceuticals in the region of interest.

In this paper, we present a 3D surface imaging technique to re-engineer traditional endoscopy into a quantitatively capable instrument. In our design concept, we demonstrate that, by utilizing two fiber bundle channels and structured light based 3D reconstruction principle, we can obtain the 3D information on a GI tract surface in all most real time fashion. The two fiber bundles are used for pattern light projection and image capture. The implied significance of our design and experiment includes: (a) It is possible to convert the traditional 2D video based endoscope into a 3D endoscope through minimum modifications; and (b) The proposed 3D endoscope allows clinicians to obtain the actual size information on any target of interest during the procedures.

Capsule endoscopy (CE) uses a miniature on-board camera in a pill for imaging gastrointestinal (GI) tract. It has provided a non-invasive and non-ionization way for gastroenterologists to diagnose GI tract diseases. However, CE has major drawbacks such as ineffective forward-looking field of view (FOV), abundant data, and lengthy viewing and interpreting time, significantly lowering the chance of finding a GI disease through the video screening process. We present a concept of utilizing full spherical field of view imaging for easy visualization. Built on camera pose tracking and 3D algorithms under spherical viewing field of view (FOV), through immersive display or VR, the technology is shown to allow clinicians to visualize interested pathological structures at finger tips on a VR device. Initial test results on phantoms show that our design is feasible.

Venipuncture is a medical practice which is ordinarily performed in hospital. However, it is a difficult procedure as the occurrence of many failure of the puncture and sometimes medical accidents such as nerve damage and blood vessel damage are reported. This is caused by the difficulty of visually identification of the blood vessel. Although the depth information of the blood vessel is also important, the existing system in clinical practice can visualize vessels only by two dimensional images. In our previous study, to estimate the three dimensional position of the blood vessel, we have proposed the system based on refocusing using light field imaging. This method can obtain cross-sectional information of blood vessel emphasized using near infrared light at each depth. However, under the influence of the strong scattering characteristics of living tissue, vascular fluoroscopic images obtained using infrared light are blurred. This is because the light scattered from other paths overlaps with the light traveling straight through the influence of blood vessel absorption. In order to suppress this blur, a system for removing scattered light is constructed. In order to eliminate the influence of scattering, a method of angular filtering using a lens array is adopted. Since this method uses lens arrays, it has high affinity with refocusing technology. First, we organize the expression related our proposed system. Then, evaluate the basic principles of the proposed system by using blood vessel simulated object.

In this study, we clinically evaluated a pulse oximeter-based device for automated capillary refill time (CRT) estimation in dogs and cats. CRT can reveal conditions like shock or anemia in dogs and cats. However, visual CRT estimation has low repeatability, and the available optical systems for automated estimation are not suitable for pets. We evaluated a custom-made portable CRT measuring device on various measurement sites of 12 dogs and 11 cats with parallel visual CRT estimation on the gum by treating veterinarian. The capillary refill was also recorded by a video camera for reference. The visual and video procedures were moderately correlated with the coefficient of 0.61; visual CRT values were on average for 0.18 s longer than the reference. On average, ~32% of measurements with the proposed device were successful. The rest failed due to excessive pigmentation, motion artifacts, and other pressure-induced effects. The measurement sites of the metacarpal pad, digit, and tail were moderately correlated with the reference values with coefficients of 0.53, 0.58, and 0.42, respectively.