This PDF file contains the front matter associated with SPIE Proceedings Volume 9711, including the Title Page, Copyright information, Table of Contents, and Conference Committee listing.

The ability to accurately assess burn wound severity in a timely manner is a critical component of wound management as it dictates the course of treatment. While full thickness and superficial burns can be easily diagnosed through visual inspection, burns that fall in between these categories are difficult to classify. Additionally, the ability to better quantify different stages of wound healing from a burn of any severity would be important for evaluating the efficacy of different treatment options. Here we present a longitudinal (28 day) study that employs spatial frequency domain imaging (SFDI) and laser speckle imaging (LSI) as non-invasive technologies to characterize in-vivo burn wounds and healing in a murine model. Burn wounds were created using an established technique of a brass comb heated to a given temperature and applied for a set amount of time. They were imaged immediately after the initial injury and then at 2, 4, 7, 14, 21, and 28 days following the injury. Biopsies were taken on the day of the injury in order to verify the extent of the burn damage as well as at different time points after the injury in order to visualize different stages of inflammation and healing. The results of this study suggest that the reduced scattering coefficient measured using SFDI and blood flow as measured using LSI have the potential to provide useful metrics for quantifying the severity of burn injuries as well as track the different stages associated with wound healing progression.

The combination of optical tweezers and Raman micro-spectroscopy is frequently referred as Raman tweezers. A single focused laser beam is utilized here both as a source of Raman scattering and a source forming an optical trap. Raman tweezers have been recently used in variety of applications in cell biology as a useful tool for non-contact and non-destructive determination of living cells properties. Here we use Raman tweezers to follow response of cells on the length of their cultivation in mineral oil. Analyses of obtained Raman spectra are based on 2D correlation analysis and allow us to determine the chemical background of the cell response in a gentle way.

A profound remodeling of the extracellular matrix (ECM) occurs in human ovarian cancer but it unknown how this affects tumor growth, where this understanding could lead to better diagnostics and therapeutic approaches. We investigate the role of these ECM alterations by using multiphoton excited (MPE) polymerization to fabricate biomimetic models to investigate operative cell-matrix interactions in invasion/metastasis. First, we create nano/microstructured gradients mimicking the basal lamina to study adhesion/migration dynamics of ovarian cancer cells of differing metastatic potential. We find a strong haptotactic response that depends on both contact guidance and ECM binding cues. While we found enhanced migration for more invasive cells, the specifics of alignment and directed migration also depend on cell polarity. We further use MPE fabrication to create collagen scaffolds with complex, 3D submicron morphology. The stromal scaffold designs are derived directly from “blueprints” based on SHG images of normal, high risk, and malignant ovarian tissues. The models are seeded with different cancer cell lines and this allows decoupling of the roles of cell characteristics (metastatic potential) and ECM structure and composition (normal vs cancer) on adhesion/migration dynamics. We found the malignant stroma structure promotes enhanced migration and proliferation and also cytoskeletal alignment. Creating synthetic models based on fibers patterns further allows decoupling the topographic roles of the fibers themselves vs their alignment within the tissue. These models cannot be synthesized by other conventional fabrication methods and we suggest the MPE image-based fabrication method will enable a variety of studies in cancer biology.

The health of the tracheal mucosa and submucosa is a vital yet poorly understood component of critical care medicine, and a minimally-invasive method is needed to monitor tracheal health in patients. Of particular interest are the ciliated cells of the tracheal epithelium that move mucus away from the lungs and prevent respiratory infection. Optical metabolic imaging (OMI) allows cellular-level measurement of metabolism, and is a compelling method for assessing tracheal health because ciliary motor proteins require ATP to function. In this pilot study, we apply multiphoton imaging of the fluorescence intensities and lifetimes of metabolic co-enzymes NAD(P)H and FAD to the mucosa and submucosa of ex vivo mouse trachea. We demonstrate the feasibility and potential diagnostic utility of these measurements for assessing tracheal health and pathophysiology at the single-cell level.

Drug delivery to tumors is well known to be chaotic and limited, partly from dysfunctional vasculature, but also because of microscopic regional variations in composition. Modeling the of transport of nanoparticle therapeutics, therefore must include not only a description of vascular permeability, but also of the movement of the drug as suspended in tumor interstitial fluid (TIF) once it leaves the blood vessel. Understanding of this area is limited because we currently lack the tools and analytical methods to characterize it. We have previously shown that directional anisotropy of drug delivery can be detected using Directional Fourier Spatial Frequency (DFSF) Analysis. Here we extend this approach to generate flow line maps of nanoparticle transport in TIF relative to tumor ultrastructure, and show that features of tumor spatial heterogeneity can be identified that are directly related to local flow isometries. The identification of these regions of limited flow may be used as a metric for determining response to therapy, or for the optimization of adjuvant therapies such as radiation pre-treatment, or enzymatic degradation.

Protein therapeutics are a rapidly growing portion of the pharmaceuticals market and have many significant advantages over traditional small molecule drugs. As this market expands, however, critical regulatory and quality control issues remain, most notably the problem of protein aggregation. Individual target proteins often aggregate into larger masses which trigger an immune response in the body, which can reduce the efficacy of the drug for its intended purpose, or cause serious anaphylactic side-effects. Although detecting and minimizing aggregate formation is critical to ensure an effective product, aggregation dynamics are often highly complicated and there is little hope of reliable prediction and prevention from first principles. This problem is compounded for aggregates in the subvisible range of 100 nm to 10 micrometers where traditional techniques for detecting aggregates have significant limitations. Here, we present an integrated optofluidic platform for detecting nanoscale protein aggregates and characterizing interactions between these aggregates and a reference surface. By delivering light to a solution of proteins with an optical waveguide, scattered light from individual protein aggregates can be detected and analyzed to determine the force profile between each particle and the waveguide surface. Unlike existing methods which only determine size or charge, our label-free screening technique can directly measure the surface interaction forces between single aggregates and the glass substrate. This direct measurement capability may allow for better empirical predictions of the stability of protein aggregates during drug manufacturing and storage.

Osteogenesis imperfecta (OI) is a currently uncurable disease where a mutation in collagen type I yields brittle bones. One potential therapy is transplantation of mesenchymal stem cells (MSCs), but controlling and enhancing transplanted cell survival has proven challenging. Therefore, we use a 2- photon imaging system to study individual transplanted cells in the living bone marrow. We ablated cells deep in the bone marrow and observed minimal collateral damage to surrounding tissue. Future work will evaluate the local impact of transplanted MSCs on bone deposition in vivo.

Understanding the mechanisms behind the proliferation of Mesenchymal Stem cells (MSCs) can offer a greater insight into the behaviour of these cells throughout their life cycles. Traditional methods of determining the rate of MSC differentiation rely on population based studies over an extended time period. However, such methods can be inadequate as they are unable to track cells as they interact; for example, in autologous cell therapies for osteoarthritis, the development of biological assays that could predict in vivo functional activity and biological action are particularly challenging. Here further research is required to determine non-histochemical biomarkers which provide correlations between cell survival and predictive functional outcome. This paper proposes using a (previously developed) advanced texture-based analysis algorithm to facilitate in vitro cells tracking using time-lapsed microscopy. The technique was adopted to monitor stem cells in the context of unlabelled, phase contrast imaging, with the goal of examining the cell to cell interactions in both monoculture and co-culture systems. The results obtained are analysed using established exploratory procedures developed for time series data and compared with the typical fluorescent-based approach of cell labelling. A review of the progress and the lessons learned are also presented.

Optical coherence tomography (OCT) is one of powerful 3D tissue imaging tools with no fluorescence staining. We have reported that Phase-Diversity Homodyne OCT developed in Hitachi could be useful for non-invasive regeneration tissue evaluation test. The OCT enables cell imaging because of high resolution (axial resolution; ~2.6 μm, lateral resolution; ~1 μm, in the air), whereas conventional OCT was not used for cell imaging because of low resolution (10~20 μm). Furthermore, the OCT has advantage over other 3D imaging devices in cost because the light source and the objective were originally used as an optical pickup of compact disc. In this report, we aimed to assess effectiveness and safety of Phase-Diversity Homodyne OCT cell imaging. Effectiveness of OCT was evaluated by imaging a living cell sheet of human oral mucosal epithelial cells. OCT images were compared with reflection confocal microscopy (RCM) images, because confocal optical system is the highest resolution (<1 μm) 3D in vivo imaging technique. Similar nuclei images were confirmed with OCT and RCM, which suggested the OCT has enough resolution to image nuclei inside a cell sheet. Degree of differentiation could be estimated using OCT images, which becomes possible because the size of cells depends on distribution of differentiation. Effect of the OCT light irradiation on cells was studied using NIH/3T3 cells. Light irradiation, the exposure amount of which is equivalent to OCT, had no impact on cell shape, cell viability, and proliferation rate. It suggested that the light irradiation has no cell damage under the condition.

The characterization of the ion channels in corneal endothelial cells and the elucidation of their involvement in corneal pathologies would lead to the identification of new molecular target for pharmacological treatments and to the clarification of corneal physiology. The corneal endothelium is an amitotic cell monolayer with a major role in preserving corneal transparency and in regulating the water and solute flux across the posterior surface of the cornea. Although endothelial cells are non-excitable, they express a range of ion channels, such as voltage-dependent Na⁺ channels and K⁺ channels, L-type Ca² channels and many others. Interestingly, purinergic receptors have been linked to a variety of conditions within the eye but their presence in the endothelium and their role in its pathophysiology is still uncertain. In this study, we were able to extract endothelial cells from single human corneas, thus obtaining primary cultures that represent the peculiarity of each donor. Corneas were from tissues not suitable for transplant in patients. We characterized the endothelial cells by confocal microscopy, both within the intact cornea and in the primary endothelial cells cultures. We also studied the functional role of the purinergic system (adenosine, ATP and their receptors) by means of electrophysiological recordings. The experiments were performed by patch clamp recordings and confocal time-lapse microscopy and our results indicate that the application of purinergic compounds modulates the amplitude of outward currents in the isolated endothelial cells. These findings may lead to the proposal of new therapies for endothelium-related corneal diseases.

Biofilms can be defined as a community of microorganisms attached to a surface, living embedded in a self- produced matrix of hydrated extracellular polymeric substances (EPS) which comprises most of the biofilm mass. We have recently used an extensive pool of microscopy techniques (confocal fluorescence, electron and scanning probe microscopies) at the micro and nanoscales in order to create a detailed temporal observation of Xylella fastidiosa biofilm formation, using both wild type strain and Green Fluorescent Protein (GFP)-modified cells of this citrus phytopathogen. We have identified three different EPS compositions, as well as their spatial and temporal distribution from single cell to mature biofilm formation stages. In the initial adhesion stage, soluble-EPS (S-EPS) accumulates at cell polar regions and forms a surface layer which facilitates irreversible cell attachment and cell cluster formation. These small clusters are subsequently connected by filamentous cells; further S-EPS surface coverage facilitates cell attachment and form filaments, leading to a floating framework of mature biofilms. The important role of EPS in X.fastidiosa biology was further investigated by imunolabelling experiments to detect the distribution of XadA1 adhesin, which is expressed in early stages of biofilm formation and released in outer membrane vesicles. This protein is located mainly in S-EPS covered areas, as well as on the filaments, indicating a molecular pathway to the enhanced cell attachment previously observed. These results suggest that S-EPS may thus represent an important target for disease control, slow plant colonization by the bacteria, keeping the plant more productive in the field.

A histogram of oriented gradient (HOG) feature is applied to the field of diseased cell detection, which can detect diseased cells in high resolution tissue images rapidly, accurately and efficiently. Firstly, motivated by symmetrical cellular forms, a new HOG symmetrical feature based on the traditional HOG feature is proposed to meet the condition of cell detection. Secondly, considering the high feature dimension of traditional HOG feature leads to plenty of memory resources and long runtime in practical applications, a classical dimension reduction method called principal component analysis (PCA) is used to reduce the dimension of high-dimensional HOG descriptor. Because of that, computational speed is increased greatly, and the accuracy of detection can be controlled in a proper range at the same time. Thirdly, support vector machine (SVM) classifier is trained with PCA-HOG symmetrical features proposed above. At last, practical tissue images is detected and analyzed by SVM classifier. In order to verify the effectiveness of this new algorithm, it is practically applied to conduct diseased cell detection which takes 200 pieces of H&E (hematoxylin & eosin) high resolution staining histopathological images collected from 20 breast cancer patients as a sample. The experiment shows that the average processing rate can be 25 frames per second and the detection accuracy can be 92.1%.

Erythrocytes, or red blood cells, transport oxygen to and carbon dioxide from the body's tissues and organs. Red blood cell mechanical properties are altered in a number of diseases such as sickle cell anaemia and malaria. Additionally, mechanically modified red blood cell ghosts are being considered as a long-term, biocompatible carrier for drug delivery and for blood analyte sensing. Brillouin spectroscopy enables viscoelastic characterization of samples at the microscale. In this report, Brillouin spectroscopy is applied to characterize the mechanical properties of red blood cells and red blood cell ghosts.

Cancer cell morphology is closely related to their phenotype and activity. These characteristics are important in drug-response prediction for personalized cancer therapeutics. We used multi-channel fluorescence microscopy images to analyze the morphology of highly cohesive cancer cells. First, we detected individual nuclei regions in single-channel images using advanced simple linear iterative clustering. The center points of the nuclei regions were used as seeds for the Voronoi diagram method to extract spatial arrangement features from cell images. Human cancer cell populations form irregularly shaped aggregates, making their detection more difficult. We overcame this problem by identifying individual cells using an image-based shape descriptor. Finally, we analyzed the correlation between cell agglutination and cell shape.

Rapid point of care diagnostic tests and tests to provide therapeutic information are now available for a range of specific conditions from the measurement of blood glucose levels for diabetes to card agglutination tests for parasitic infections. Due to a lack of specificity these test are often then backed up by more conventional lab based diagnostic methods for example a card agglutination test may be carried out for a suspected parasitic infection in the field and if positive a blood sample can then be sent to a lab for confirmation. The eventual diagnosis is often achieved by microscopic examination of the sample. In this paper we propose a computerized vision system for aiding in the diagnostic process; this system used a novel particle recognition algorithm to improve specificity and speed during the diagnostic process. We will show the detection and classification of different types of cells in a diluted blood sample using regression analysis of their size, shape and colour. The first step is to define the objects to be tracked by a Gaussian Mixture Model for background subtraction and binary opening and closing for noise suppression. After subtracting the objects of interest from the background the next challenge is to predict if a given object belongs to a certain category or not. This is a classification problem, and the output of the algorithm is a Boolean value (true/false). As such the computer program should be able to "predict" with reasonable level of confidence if a given particle belongs to the kind we are looking for or not. We show the use of a binary logistic regression analysis with three continuous predictors: size, shape and color histogram. The results suggest this variables could be very useful in a logistic regression equation as they proved to have a relatively high predictive value on their own.

The Directional Fourier Spatial Frequencies (DFSF) of a 2D image can identify similarity in spatial patterns within groups of related images. A Support Vector Machine (SVM) can then be used to classify images if the inter-image variance of the FSF in the training set is bounded. However, if variation in FSF increases with training set size, accuracy may decrease as the size of the training set increases. This calls for a method to identify a set of training images from among the originals that can form a vector basis for the entire class. Applying the Cauchy product method we extract the DFSF spectrum from radiographs of osteoporotic bone, and use it as a matched filter set to eliminate noise and image specific frequencies, and demonstrate that selection of a subset of superclassifiers from within a set of training images improves SVM accuracy. Central to this challenge is that the size of the search space can become computationally prohibitive for all but the smallest training sets. We are investigating methods to reduce the search space to identify an optimal subset of basis training images.

A deployable mobile medical application is presented that employs a smartphone camera, patient input, internet connectivity, and cloud-based image processing techniques to document and analyze physiological characteristics of hands in osteoarthritis (OA) patients. The application performs digital image processing that spatially calibrates the image, locates hand fiduciary features, and quantifies hand features to identify abnormal distal and proximal interphalangeal joints. The algorithm determines the finger centerlines and joint coordinates. From these anatomical fiduciary points, it measures the width of fingers, location and size of joints, and finger joint angulation. The diagnostically relevant features measured by the mobile application can be applied to current diagnostic protocols such as the American College of Rheumatology (ACR) criteria for OA. Based on the results from a pilot study, the mobile application was modified to include interactive user guidance built into the smartphone. This app makes improvements on the algorithm that validate the image quality and makes the algorithm less dependent on precise capture conditions. Based on clinical feedback, a web-based portal and dashboard for advanced analysis was developed and presented. Clinicians, researchers, and patients can use this to explore relationships between pain, treatment, environmental parameters, and lifestyle factors.

This project introduces a new smartphone mobile medical application that utilizes cloud-based image processing to measure and characterize acne lesions. The application uses image segmentation and image analysis to measure erythema regions and compare them to known physiological characteristics to identify lesion type and stage. It then predicts the time when the lesion will heal, the “Vanishing Point”. The system can update and refine its prediction of the vanishing point as additional images of actual lesion progression are captured on subsequent days. Data from 40 subjects are analyzed for classification accuracy.

A conventional fluorescence microscope is used to detect muscle foods (meat and poultry) spoilage. Fluorescence images are acquired based on native intrinsic tissue emission at selective excitation wavelengths. Multivariate data analysis algorithm, e.g. nonnegative matrix factorization (NMF) is used to unmix the fluorescence signal due to different native fluorohpores. Relative concentrations of these fluorophores are retrieved and used to detect the spoilage of the meat. The method can potentially be used to design a portable device which may utilize a conventional color camera such as a cell phone camera for data acquisition and analysis, and achieve rapid in-situ spoilage detection.

Nutrient and oxygen transport, as well as the removal of metabolic waste are essential processes to support and maintain viable tissue. Current bioreactor technology used to grow tissue cultures in vitro has a fundamental limit to the thickness of tissues. Based on the low diffusion limit of oxygen a maximum tissue thickness of 200 μm is possible. The efficiency of those systems is currently under investigation. During the cultivation process of the artificial tissue in bioreactors, which lasts 28 days or longer, there are no possibilities to investigate the viability of cells. This work is designed to determine the influence of a non-disruptive cell viability measuring system on cellular activity. The measuring system uses a natural cellular marker produced during normal metabolic activity.

Nicotinamide adenine dinucleotide (NADH) is a coenzyme naturally consumed and produced during cellular metabolic processes and has thoroughly been studied to determine the metabolic state of a cell. Measuring the fluorescence of NADH within the cell represents a non-disruptive marker for cell viability. Since the measurement process is optical in nature, NADH fluorescence also provides a pathway for sampling at different measurement depths within a given tissue sample. The measurement system we are using utilizes a special UV light source, to excite the NADH fluorescence state. However, the high energy potentially alters or harms the cells. To investigate the influence of the excitation signal, the cells were irradiated with a laser operating at a wavelength of 355 nm and examined for cytotoxic effects. The aim of this study was to develop a non-cytotoxic system that is applicable for large-scale operations during drug-tissue interaction testing.

Current microscopic and endoscopic technologies for cancer screening utilize white-light illumination sources. Hyper-spectral imaging has been shown to improve sensitivity while retaining specificity when compared to white-light imaging in both microscopy and in vivo imaging. However, hyperspectral imaging methods have historically suffered from slow acquisition times due to the narrow bandwidth of spectral filters. Often minutes are required to gather a full image stack. We have developed a novel approach called excitation-scanning hyperspectral imaging that provides 2–3 orders of magnitude increased signal strength. This reduces acquisition times significantly, allowing for live video acquisition. Here, we describe a preliminary prototype excitation-scanning hyperspectral imaging system that can be coupled with endoscopes or microscopes for hyperspectral imaging of tissues and cells.

Our system is comprised of three subsystems: illumination, transmission, and imaging. The illumination subsystem employs light-emitting diode arrays to illuminate at different wavelengths. The transmission subsystem utilizes a unique geometry of optics and a liquid light guide. Software controls allow us to interface with and control the subsystems and components. Digital and analog signals are used to coordinate wavelength intensity, cycling and camera triggering. Testing of the system shows it can cycle 16 wavelengths at as fast as 1 ms per cycle. Additionally, more than 18% of the light transmits through the system. Our setup should allow for hyperspectral imaging of tissue and cells in real time.

We have developed a multimode dermoscope (SkinSpect™) capable of illuminating human skin samples in-vivo with spectrally-programmable linearly-polarized light at 33 wavelengths between 468nm and 857 nm. Diffusely reflected photons are separated into collinear and cross-polarized image paths and images captured for each illumination wavelength. In vivo human skin nevi (N = 20) were evaluated with the multimode dermoscope and melanin and hemoglobin concentrations were compared with Spatially Modulated Quantitative Spectroscopy (SMoQS) measurements. Both systems show low correlation between their melanin and hemoglobin concentrations, demonstrating the ability of the SkinSpect™ to separate these molecular signatures and thus act as a biologically plausible device capable of early onset melanoma detection.

Hyperspectral imaging (HSI) can combine morphological and molecular information, yielding potential for real-time and high throughput multiplexed fluorescent contrast agent imaging. Multiplexed readout from targets, such as cell surface receptors overexpressed in cancer cells, could improve both sensitivity and specificity of tumor identification. There remains, however, a need for compact and cost effective implementations of the technology. We have implemented a low-cost wide-field multiplexed fluorescence imaging system, which combines LED excitation at 590, 655 and 740 nm with a compact commercial solid state HSI system operating in the range 600 - 1000 nm. A key challenge for using reflectance-based HSI is the separation of contrast agent fluorescence from the reflectance of the excitation light. Here, we illustrate how it is possible to address this challenge in software, using two offline reflectance removal methods, prior to least-squares spectral unmixing. We made a quantitative comparison of the methods using data acquired from dilutions of contrast agents prepared in well-plates. We then established the capability of our HSI system for non-invasive in vivo fluorescence imaging in small animals using the optimal reflectance removal method. The HSI presented here enables quantitative unmixing of at least four fluorescent contrast agents (Alexa Fluor 610, 647, 700 and 750) simultaneously in living mice. A successful unmixing of the four fluorescent contrast agents was possible both using the pure contrast agents and with mixtures. The system could in principle also be applied to imaging of ex vivo tissue or intraoperative imaging in a clinical setting. These data suggest a promising approach for developing clinical applications of HSI based on multiplexed fluorescence contrast agent imaging.

Anastomosis, the connection of two structures, is a critical procedure for reconstructive surgery with over 1 million cases/year for visceral indication alone. However, complication rates such as strictures and leakage affect up to 19% of cases for colorectal anastomoses and up to 30% for visceral transplantation anastomoses. Local ischemia plays a critical role in anastomotic complications, making blood perfusion an important indicator for tissue health and predictor for healing following anastomosis. In this work, we apply a real time multispectral imaging technique to monitor impact on tissue perfusion due to varying interrupted suture spacing and suture tensions. Multispectral tissue images at 470, 540, 560, 580, 670 and 760 nm are analyzed in conjunction with an empirical model based on diffuse reflectance process to quantify the hemoglobin oxygen saturation within the suture site. The investigated tissues for anastomoses include porcine small (jejunum and ileum) and large (transverse colon) intestines. Two experiments using interrupted suturing with suture spacing of 1, 2, and 3 mm and tension levels from 0 N to 2.5 N are conducted. Tissue perfusion at 5, 10, 20 and 30 min after suturing are recorded and compared with the initial normal state. The result indicates the contrast between healthy and ischemic tissue areas and assists the determination of suturing spacing and tension. Therefore, the assessment of tissue perfusion will permit the development and intra-surgical monitoring of an optimal suture protocol during anastomosis with less complications and improved functional outcome.

The natural fluorescence (autofluorescence) of tissues has been noted as a biomarker for cancer for several decades. Autofluorescence contrast between tumors and healthy tissues has been of significant interest in endoscopy, leading to development of autofluorescence endoscopes capable of visualizing 2-3 fluorescence emission wavelengths to achieve maximal contrast. However, tumor detection with autofluorescence endoscopes is hindered by low fluorescence signal and limited quantitative information, resulting in prolonged endoscopic procedures, prohibitive acquisition times, and reduced specificity of detection.

Our lab has designed a novel excitation-scanning hyperspectral imaging system with high fluorescence signal detection, low acquisition time, and enhanced spectral discrimination. In this study, we surveyed a comprehensive set of excised tissues to assess the feasibility of detecting tissue-specific pathologies using excitation-scanning. Fresh, untreated tissue specimens were imaged from 360 to 550 nm on an inverted fluorescence microscope equipped with a set of thin-film tunable filters (Semrock, A Unit of IDEX). Images were subdivided into training and test sets. Automated endmember extraction (ENVI 5.1, Exelis) with PCA identified endmembers within training images of autofluorescence. A spectral library was created from 9 endmembers. The library was used for identification of endmembers in test images. Our results suggest (1) spectral differentiation of multiple tissue types is possible using excitation scanning; (2) shared spectra between tissue types; and (3) the ability to identify unique morphological features in disparate tissues from shared autofluorescent components. Future work will focus on isolating specific molecular signatures present in tissue spectra, and elucidating the contribution of these signatures in pathologies.

In recent years, it has been suggested that epicardial adipose tissue (EAT) plays an important role in development of coronary artery disease (CAD) and diabetes mellitus (DM). In this article, we used two-photon fluoresce microscope to measure the fluorescence metabolic image of EAT, which obtained from the patient with/without CAD/DM. We used 740nm and 890nm infrared light to excite the auto-fluorescence of metabolic molecules NADH and FAD respectively. We collected the fluorescence signal at wavelength 450nm to 500nm and 500nm to 550nm to obtain the metabolic image. Through the image, we computed the redox ratio (NADH/FAD) by analyzing the intensity. The preliminary result showed that the redox ratio increase in the patients with CAD. It indicates EAT adipocytes of patient with CAD have decreased cellular metabolic activity. But there were no significant variation of redox ratio in the patients with DM.

We discuss the theoretical framework of a new method of performing spectroscopy: labeling a unique intensity modulation frequency onto the optical frequencies of an incident or illumination power spectrum. In a manner similar to Fourier transform spectroscopy, we rescale the optical angular frequencies of the power spectrum down to readily measured frequencies on a square-law detector enabling rapid spectral update rates on a single element detector.

This project's goal is to determine how to effectively implement a technology continuum from a low cost, remotely deployable imaging device to a more sophisticated multimode imaging system within a standard clinical practice. In this work a smartphone is used in conjunction with an optical attachment to capture cross-polarized and collinear color images of a nevus that are analyzed to quantify chromophore distribution. The nevus is also imaged by a multimode hyperspectral system, our proprietary SkinSpect™ device. Relative accuracy and biological plausibility of the two systems algorithms are compared to assess aspects of feasibility of in-home or primary care practitioner smartphone screening prior to rigorous clinical analysis via the SkinSpect.

Introduction: The International Society for Advancement of Cytometry (ISAC) has created a standard for the Minimum Information about a Flow Cytometry Experiment (MIFlowCyt 1.0). CytometryML will serve as a common metadata standard for flow and image cytometry (digital microscopy).

Methods: The MIFlowCyt data-types were created, as is the rest of CytometryML, in the XML Schema Definition Language (XSD1.1). The datatypes are primarily based on the Flow Cytometry and the Digital Imaging and Communication (DICOM) standards. A small section of the code was formatted with standard HTML formatting elements (p, h1, h2, etc.).

Results:1) The part of MIFlowCyt that describes the Experimental Overview including the specimen and substantial parts of several other major elements has been implemented as CytometryML XML schemas (www.cytometryml.org). 2) The feasibility of using MIFlowCyt to provide the combination of an overview, table of contents, and/or an index of a scientific paper or a report has been demonstrated. Previously, a sample electronic publication, EPUB, was created that could contain both MIFlowCyt metadata as well as the binary data.

Conclusions: The use of CytometryML technology together with XHTML5 and CSS permits the metadata to be directly formatted and together with the binary data to be stored in an EPUB container. This will facilitate: formatting, data- mining, presentation, data verification, and inclusion in structured research, clinical, and regulatory documents, as well as demonstrate a publication’s adherence to the MIFlowCyt standard, promote interoperability and should also result in the textual and numeric data being published using web technology without any change in composition.

The processing and analysis of surface-immobilized single molecule FRET (Förster resonance energy transfer) data follows systematic steps (e.g. single molecule localization, clearance of different sources of noise, selection of the conformational and kinetic model, etc.) that require a solid knowledge in optics, photophysics, signal processing and statistics. The present proceeding aims at standardizing and facilitating procedures for single molecule detection by guiding the reader through an optimization protocol for a particular experimental data set. Relevant features were determined from single molecule movies (SMM) imaging Cy3- and Cy5-labeled Sc.ai5γ group II intron molecules synthetically recreated, to test the performances of four different detection algorithms. Up to 120 different parameterizations per method were routinely evaluated to finally establish an optimum detection procedure. The present protocol is adaptable to any movie displaying surface-immobilized molecules, and can be easily reproduced with our home-written software MASH (multifunctional analysis software for heterogeneous data) and script routines (both available in the download section of www.chem.uzh.ch/rna).

Atherosclerosis is characterized by the growth of fibrous plaques due to the retention of cholesterol and lipids within the artery wall, which can lead to vessel occlusion and cardiac events. One way to evaluate arterial disease is to quantify the amount of lipid present in these plaques, since a higher disease burden is characterized by a higher concentration of lipid. Although therapeutic stimulation of reverse cholesterol transport to reduce cholesterol deposits in plaque has not produced significant results, this may be due to current image analysis methods which use averaging techniques to calculate the total amount of lipid in the plaque without regard to spatial distribution, thereby discarding information that may have significance in marking response to therapy. Here we use Directional Fourier Spatial Frequency (DFSF) analysis to generate a characteristic spatial frequency spectrum for atherosclerotic plaques from C57 Black 6 mice both treated and untreated with a cholesterol scavenging nanoparticle. We then use the Cauchy product of these spectra to classify the images with a support vector machine (SVM). Our results indicate that treated plaque can be distinguished from untreated plaque using this method, where no difference is seen using the spatial averaging method. This work has the potential to increase the effectiveness of current in-vivo methods of plaque detection that also use averaging methods, such as laser speckle imaging and Raman spectroscopy.

The skin contains several fluorescent molecules or fluorophores that serve as markers of structure, function and composition. UV fluorescence excitation photography is a simple and effective way to image specific intrinsic fluorophores, such as the one ascribed to tryptophan which emits at a wavelength of 345 nm upon excitation at 295 nm, and is a marker of cellular proliferation. Earlier, we built a clinical UV photography system to image cellular proliferation. In some samples, the naturally low intensity of the fluorescence can make it difficult to separate the fluorescence of cells in higher proliferation states from background fluorescence and other imaging artifacts -- like electronic noise. In this work, we describe a statistical image segmentation method to separate the fluorescence of interest. Statistical image segmentation is based on image averaging, background subtraction and pixel statistics. This method allows to better quantify the intensity and surface distributions of fluorescence, which in turn simplify the detection of borders. Using this method we delineated the borders of highly-proliferative skin conditions and diseases, in particular, allergic contact dermatitis, psoriatic lesions and basal cell carcinoma. Segmented images clearly define lesion borders. UV fluorescence excitation photography along with statistical image segmentation may serve as a quick and simple diagnostic tool for clinicians.

The literature contains several reports of Mie-like fits to angular-domain elastic scattering measurements from multiple cells or isolated mitochondria. In these studies, the sampling volume typically contains hundreds or thousands of mitochondria, allowing for the size distribution of mitochondria to be modeled as a smooth function, (e.g. Gaussian or log-normal) with a small number of free parameters. In the case of a single-cell volume containing significantly fewer mitochondria, the true size distribution will no longer be as smooth. Increasing the number of free parameters can lead to unstable fits, however, as the forward-directed angular scattering pattern from such a population illuminated with 785 nm light is a monotonically decaying radial function with few distinct features. Using simulations, we have investigated the limitations of modeling single-cell mitochondrial scattering using smooth population distributions of Mie scatterers. In different instances, the fidelity of the estimated size information can be limited by the number of organelles, the angular detection range, or the non-ideality of the data (both speckle and shot noise). We will describe the conditions under which each of these effects dominates. We will also discuss whether mean and standard deviation are the best sizes to report from such Mie modeling, or if there are other size parameters that have greater fidelity to the true, non-smooth size distributions.

Light Sheet Microscopy is a high-resolution fluorescence microscopic technique which enables to observe the mouse brain vascular network clearly with immunostaining. However, micro-vessels are stained with few fluorescence antibodies and their signals are much weaker than large vessels, which make micro-vessels unclear in LSM images. In this work, we developed a vascular image enhancement method to enhance micro-vessel details which should be useful for vessel statistics analysis. Since gradient describes the edge information of the vessel, the main idea of our method is to increase the gradient values of the enhanced image to improve the micro-vessels contrast. Our method contained two steps: 1) calculate the gradient image of LSM image, and then amplify high gradient values of the original image to enhance the vessel edge and suppress low gradient values to remove noises. Then we formulated a new L1-norm regularization optimization problem to find an image with the expected gradient while keeping the main structure information of the original image. 2) The split Bregman iteration method was used to deal with the L1-norm regularization problem and generate the final enhanced image. The main advantage of the split Bregman method is that it has both fast convergence and low memory cost. In order to verify the effectiveness of our method, we applied our method to a series of mouse brain vascular images acquired from a commercial LSM system in our lab. The experimental results showed that our method could greatly enhance micro-vessel edges which were unclear in the original images.

In this work, the microencapsulation of water-soluble drug (doxorubicin, Dox) and imaging agent (perfluorocarbon, PFC) is performed by a novel liquid driven tri-axial flow focusing (LDTFF) device. The formation of stable triple-layered cone-jet mode can be observed in the simple well-assembled LDTFF device, providing an easy approach to fabricate mono-disperse triple-layered microcapsules with high encapsulation efficiency, high throughput and low cost in just one step. The fluorescence images show that the microcapsules have a satisfactory core-shell structure. The SEM micrographs show spherical and smooth surface views of the triple-layered microcapsules after being stirred 72h to remove the organic solvent totally. The results of thermo-responsive release experiments of the produced triple-layered microcapsules show these multifunctional capsules can be well stimulated when the environment temperature is beyond 55 degree centigrade. In a word, this novel approach has a great potential in applications such as drug delivery and image-guided therapy.

The prevailing hypothesis for the existence and healing of the avascular corneal epithelium is that this layer of cells are continually produced by stem cells in the limbus and transported onto the cornea to mature into corneal epithelium. In the event that the cornea is damaged and the limbal stem cell population is severely reduced, this condition known as Limbal Stem Cell Deficiency and can lead to blindness. There are numerous treatments but most have high long term failure rates. Most treatment methods include the transplantation of limbal stem cells into damaged limbus with hope of repopulating the region and regenerating at healthy corneal epithelium. Optical Coherence Tomography (OCT) is well known for its high resolution in vivo images. A bespoke OCT has been built to investigate the trajectories of these limbal stem cells after transplantation to see whether if they do repopulate the damaged limbus or not. In the experimentation magneto-labelling was used to track the limbal stem cells. For the magneto-labelling a mixture of limbal stem cells and cornea epithelium are cultured with super paramagnetic iron (Fe3O4) nanoparticles (20-30nm in size) for 24hours, to allow for uptake. The cells are then transplanted onto the denuded cornea. The transplanted cell mixture with the encapsulated magnetic nanoparticles is actuated with an external magnetic field 0.08T leading to a phase modulation on the signal. A Phase sensitive Magneto-motive OCT is used to locate the transplanted cells. The location of the cells with embed SPIOs were located both in 2D and 3D.

Imaging cells and tissues through opaque and turbid media is challenging and presents a major barrier for monitoring maturation and remodeling of bioengineered tissues. The fiber optics based imaging system described here offers a new approach for fluorescent cell imaging. A micro imaging channel is embedded in a Polycaprolactone (PCL) electrospun scaffold designed for cell seeding, which allows us to use an optical fiber to locally deliver excitation laser close to the fluorescent cells. The emission is detected by an Electron Multiplying Charge Coupled Device (EMCCD) detector and image reconstruction of multiple excitation points is achieved with a working distance of several centimeters. The objective of this study is to assess the effects of system parameters on image reconstruction outcomes. Initial studies using fluorescent beads indicated that scaffold thickness had a small effect on image quality, whereas scaffold composition (collagen content), fluorophore spectra, and the reconstruction window size had a large effect. The results also suggest that a far-red fluorescent emission is preferential when using collagenous scaffolds with a thickness of up to 500 μm. Using these optimized parameters, we were able to image fluorescently labeled cells on a scaffold with a resolution of 15-20 μm, and have also measured muscle progenitor cell differentiation and scaffold surface coverage with endothelial cells. In the future, this imaging platform can be applied to other bioengineered tissues for non-invasive monitoring both in vitro and in vivo.

Optical projection tomography is a new optical imaging method for visualizing small biological specimens in three dimension. The most important advantage of OPT is to fill the gap between MRI and confocal microscope for the specimen having the range of 1-10 mm. Thus, it has been mainly used for whole-mount small animals and developmental study since this imaging modality was developed. The ability of OPT delivering anatomical and functional information of relatively large tissue in 3D has made it a promising platform in biomedical research. Recently, the potential of OPT spans its coverage to cellular scale. Even though there are increasing demand to obtain better understanding of cellular dynamics, only few studies to visualize cellular structure, shape, size and functional morphology over tissue has been investigated in existing OPT system due to its limited field of view. In this study, we develop a novel optical imaging system for 3D cellular imaging with OPT integrated with dynamic focusing technique. Our tomographic setup has great potential to be used for identifying cell characteristic in tissue because it can provide selective contrast on dynamic focal plane allowing for fluorescence as well as absorption. While the dominant contrast of optical imaging technique is to use the fluorescence for detecting certain target only, the newly developed OPT system will offer considerable advantages over currently available method when imaging cellar molecular dynamics by permitting contrast variation. By achieving multi-contrast, it is expected for this new imaging system to play an important role in delivering better cytological information to pathologist.

We present a proof-of-principle prototype of a fiber-based snapshot spectrometer to provide high spatial and spectral sampling for biomedical application such as cell signaling or diagnostics. An image is collected by a custom fiber bundle and then divided into spatial groups with spaces in between for dispersion. The image is later scaled down by an image taper (to scale down the image size and allow smaller optical components), dispersed with a prism and captured by a CCD camera. An interpolation algorithm is used to locate each wavelength and reconstruct the image for each spectral channel. The fiber bundle is fabricated by aligning multi-mode bare fiber ribbons as matrix, gluing together in Teflon molds, laser cutting and polishing. We present preliminary finger occlusion results obtained with the spectrometer where the oxy- and deoxy-hemoglobin spectrum could be differentiated.

We report on efforts to improve the lifetime of biological lasers through the use of ascorbic acid (also commonly known as vitamin C). Fluorescent proteins and dyes, used in biological lasers, suffer from photobleaching due to the build-up of reactive oxygen species (ROS) which causes damage leading to a decrease in emission over time. This is an issue both for laser lifetime and cell health. It has previously been shown that ascorbic acid can be effective in reducing ROS levels in a variety of applications. For our experiments human embryonic kidney cells (HEK293), containing the fluorescent dye Calcein AM, were placed between two dielectric plane mirrors to form a laser cavity. The cells were pumped using the output of a Ti:Sapphire femtosecond OPO system, frequency doubled twice in BBO crystals, giving an output of 474 nm. Initial results have shown an increase in laser lifetime when ascorbic acid is added to cells indicating a reduction in the build-up of ROS.

Pathologist examination of tissue slides provides insightful information about a patient’s disease. Traditional analysis of tissue slides is performed under a binocular microscope, which requires staining of the sample and delays the examination. We present a simple cost-effective lensfree imaging method to record 2–4μm resolution wide-field (10 mm² to 6 cm²) images of unstained tissue slides. The sample processing time is reduced as there is no need for staining. A wide field of view (10 mm²) lensfree hologram is recorded in a single shot and the image is reconstructed in 2s providing a very fast acquisition chain. The acquisition is multispectral, i.e. multiple holograms are recorded simultaneously at three different wavelengths, and a dedicated holographic reconstruction algorithm is used to retrieve both amplitude and phase. Whole tissue slides imaging is obtained by recording 130 holograms with X-Y translation stages and by computing the mosaic of a 25 x 25 mm² reconstructed image. The reconstructed phase provides a phase-contrast-like image of the unstained specimen, revealing structures of healthy and diseased tissue. Slides from various organs can be reconstructed, e.g. lung, colon, ganglion, etc. To our knowledge, our method is the first technique that enables fast wide-field lensfree imaging of such unlabeled dense samples. This technique is much cheaper and compact than a conventional phase contrast microscope and could be made portable. In sum, we present a new methodology that could quickly provide useful information when a rapid diagnosis is needed, such as tumor margin identification on frozen section biopsies during surgery.

Measuring molecular orientation properties is very appealing for scientists in molecular and cell biology, as well as biomedical research. Orientational organization at the molecular scale is indeed an important brick to cells and tissues morphology, mechanics, functions and pathologies. Recent work has shown that polarized fluorescence imaging, based on excitation polarization tuning in the sample plane, is able to probe molecular orientational order in biological samples; however this applies only to information in 2D, projected in the sample plane. To surpass this limitation, we extended this approach to excitation polarization tuning in 3D. The principle is based on the decomposition of any arbitrary 3D linear excitation in a polarization along the longitudinal z-axis, and a polarization in the transverse xy–sample plane. We designed an interferometer with one arm generating radial polarization light (thus producing longitudinal polarization under high numerical aperture focusing), the other arm controlling a linear polarization in the transverse plane. The amplitude ratio between the two arms can vary so as to get any linear polarized excitation in 3D at the focus of a high NA objective. This technique has been characterized by polarimetry imaging at the back focal plane of the focusing objective, and modeled theoretically. 3D polarized fluorescence microscopy is demonstrated on actin stress fibers in non-flat cells suspended on synthetic polymer structures forming supporting pillars, for which heterogeneous actin orientational order could be identified. This technique shows a great potential in structural investigations in 3D biological systems, such as cell spheroids and tissues.

Fluorescence molecular tomography (FMT) can visualize biological activities at cellular and molecular levels in vivo, and has been extensively used in drug delivery and tumor detection research of small animals. The ill-posedness of the FMT inverse problem makes it difficult to reconstruct and resolve multiple adjacent fluorescent targets that have different functional features but are labeled with the same fluorochrome. An algorithm based on independent component analysis (ICA) for multispectral excited FMT is proposed to resolve multiple fluorescent targets in this study. Fluorescent targets are excited by multispectral excitation, and the three-dimensional distribution of fluorescent yields under the excitation spectrum is reconstructed by an iterative Tikhonov regularization algorithm. Subsequently, multiple fluorescent targets are resolved from mixed fluorescence signals by employing ICA. Simulations were performed and the results demonstrate that multiple adjacent fluorescent targets can be resolved if the number of excitation wavelengths is not smaller than that of fluorescent targets with different concentrations. The algorithm obtains both independent components that provide spatial information of different fluorescent targets and spectral courses that reflect variation trends of fluorescent yields along with the excitation spectrum. By using this method, it is possible to visualize the metabolism status of drugs in different structure organs, and quantitatively depict the variation trends of fluorescent yields of each functional organ under the excitation spectrum. This method may provide a pattern for tumor detection, drug delivery and treatment monitoring in vivo.

The goal of this project is to estimate non-nuclear organelle size distributions in single cells by measuring angular scattering patterns and fitting them with Mie theory. Simulations have indicated that the large relative size distribution of organelles (mean:width≈2) leads to unstable Mie fits unless scattering is collected at polar angles less than 20 degrees. Our optical system has therefore been modified to collect angles down to 10 degrees. Initial validations will be performed on polystyrene bead populations whose size distributions resemble those of cell organelles. Unlike with the narrow bead distributions that are often used for calibration, we expect to see an order-of-magnitude improvement in the stability of the size estimates as the minimum angle decreases from 20 to 10 degrees. Scattering patterns will then be acquired and analyzed from single cells (EMT6 mouse cancer cells), both fixed and live, at multiple time points. Fixed cells, with no changes in organelle sizes over time, will be measured to determine the fluctuation level in estimated size distribution due to measurement imperfections alone. Subsequent measurements on live cells will determine whether there is a higher level of fluctuation that could be attributed to dynamic changes in organelle size. Studies on unperturbed cells are precursors to ones in which the effects of exogenous agents are monitored over time.

Although picrosirius red (PSR) is known to be in quantifying collagen under polarized light (PL), commonly used linearly PL can result in an underestimation of collagen, as some of the fibers may appear dark if aligned with the transmission axis of the polarizers. To address this, a sample may be imaged with circularly polarized light at the expense of higher background intensity. However, the quality and alignment of the microscope illumination optics, polarizers and waveplates can still produce imaging variability with circular polarization. A simpler technique was tested that minimized variability and background intensity with linear polarization by acquiring images at multiple angles of histology slide rotation to create a composite co-registered image, permitting the optimal semi-quantitative visualization of collagen.

Linear polarization imaging was performed on PSR stained artery sections. By rotating the slide at 60° intervals while maintaining illumination, polarization and exposure parameters, 6 images were acquired for each section. A composite image was created from the 6 co-registered images, and comprised of the maximum pixel intensity at each point.

Images from any of the 6 rotation positions consistently showed variation in PSR signal. A composite image compensates for this variability, without loss of spatial resolution. Additionally, grayscale analysis showed an increased intensity range of 15 – 50% with a linearly polarized composite image over a circularly polarized image after background correction, indicating better SNR. This proposed technique will be applied in the development of a near infrared spectroscopy algorithm to detect vulnerable atherosclerotic plaques in vivo.

Various types of collagens, e.g. type I and III, represent the main load-bearing components in biological tissues. Their composition changes during processes like wound healing and fibrosis. Collagens exhibit autofluorescence when excited by ultra-violet light, distinguishable by their unique fluorescent lifetimes across a range of emission wavelengths. Therefore, we designed a miniaturized spectral-lifetime detection system for collagens as a non-invasive probe for monitoring tissue in wound healing and scarring applications. A sine modulated LED illumination was applied to enable frequency domain (FD) fluorescence lifetime measurements under different wavelengths bands, separated via a series of longpass dichroics at 387nm, 409nm and 435nm. To achieve the minute scale of optomechanics, we employed a stereolithography based 3D printer with <50 μm resolution to create a custom designed optical mount in a hand-held form factor. We examined the characteristics of the 3D printed optical system with finite element modeling to simulate the effect of thermal (LED) and mechanical (handling) strain on the optical system. Using this device, the phase shift and demodulation of collagen types were measured, where the separate spectral bands enhanced the differentiation of their lifetimes.

We have developed and tested a system for real-time intra-operative optical identification and classification of brain tissues using time-resolved fluorescence spectroscopy (TRFS). A supervised learning algorithm using linear discriminant analysis (LDA) employing selected intrinsic fluorescence decay temporal points in 6 spectral bands was employed to maximize statistical significance difference between training groups. The linear discriminant analysis on in vivo human tissues obtained by TRFS measurements (N = 35) were validated by histopathologic analysis and neuronavigation correlation to pre-operative MRI images. These results demonstrate that TRFS can differentiate between normal cortex, white matter and glioma.

Circulating tumor cell (CTC) identification has applications in both early detection and monitoring of solid cancers. The rarity of CTCs, expected at ~1-50 CTCs per million nucleated blood cells (WBCs), requires identifying methods based on biomarkers with high sensitivity and specificity for accurate identification. Discovery of biomarkers with ever higher sensitivity and specificity to CTCs is always desirable to potentially find more CTCs in cancer patients thus increasing their clinical utility. Here, we investigate quantitative image cytometry measurements of lipids with the biomarker panel of DNA, Cytokeratin (CK), and CD45 commonly used to identify CTCs. We engineered a device for labeling suspended cell samples with fluorescent antibodies and dyes. We used it to prepare samples for 4 channel confocal laser scanning microscopy. The total data acquired at high resolution from one sample is ~ 1.3 GB. We developed software to perform the automated segmentation of these images into regions of interest (ROIs) containing individual cells. We quantified image features of total signal, spatial second moment, spatial frequency second moment, and their product for each ROI. We performed measurements on pure WBCs, cancer cell line MCF7 and mixed samples. Multivariable regressions and feature selection were used to determine combination features that are more sensitive and specific than any individual feature separately. We also demonstrate that computation of spatial characteristics provides higher sensitivity and specificity than intensity alone. Statistical models allowed quantification of the required sensitivity and specificity for detecting small levels of CTCs in a human blood sample.