This PDF file contains the front matter associated with SPIE Proceedings Volume 9314, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and Conference Committee listing.

Quantification of plant health is crucial for agriculture and can even be used to monitor environmental factors and climate related changes. Plant health is known to be directly related to the chlorophyll content of leaves, which correlates with the capacity of the leaves to transmit or absorb light. The gold-standard method for measuring the chlorophyll concentration of a leaf is based on chemical extraction, which is complex, destructive and time-consuming. As an alternative, here we present a field-portable, cost-effective, and colorimetric method to quantify the chlorophyll content of leaves using Google Glass. For this purpose, we created a custom designed handheld device which is battery-powered and 3D-printed to separately provide uniform illumination of a selected region of interest on the leaf surface using red and white light-emitting-diodes (LEDs). The design of this device minimizes the interference of ambient light conditions to our chlorophyll measurements performed through the Glass camera. We tested this platform by using fifteen randomly selected plant species from UCLA Botanical Garden and imaging fully-grown leaves of these species using Glass. An image-processing algorithm was developed to process the acquired images and obtain the chlorophyll concentration information using the red channel intensities in our region-of-interest for both the white and red LED illumination conditions. The results obtained by this algorithm are in good agreement with the SPAD indices measured for each plant, demonstrating that Google Glass, in combination with our custom-designed illumination platform, can expand its functionality to be used as a chlorophyll meter in field-settings.

Cervical cancer is a leading cause of cancer death for women all across the developing world, where much of the infrastructure required for effective cervical cancer screening is unavailable because of limited resources. One of the most common method to screen for cervical cancer is by visual inspection with acetic acid (VIA), in which the cervix is imaged with the naked eye. Given inherent challenges in analysis and documentation when characterizing cervical tissue with the naked eye, an optical solution is needed. To address this challenge, a smartphone was modified and transformed into a mobile colposcope (a device used to image the cervix from outside) by adding a custom-fit light source and optics. The mobile smartphone colposcope was designed such that it augments VIA and easily integrates within the standard of care. The mobile smartphone colposcope is controlled by an app, which, stores cervical images captured on the mobile smartphone colposcope on a portal, enabling remote doctors to evaluate images and the treatment chosen by the health worker. Images from patients undergoing cervical cancer screening by a nurse using VIA in the University Hospital of Mirebalais (HUM) GYN outpatient clinic in Haiti were captured on the mobile smartphone colposcope. These images were later analyzed by an experienced OB/GYN at HUM, who determined whether or not the patient should be treated with cryoablation; more complicated cases were also shared with a consulting doctor in the US. The opinions of the experienced OB/GYN doctors at HUM, as well as the experts from the US, were used to educate nurses and midwives performing mobile colposcopy. These results suggest that remote assessment offered by mobile colposcopy can improve training of health workers performing VIA, and ultimately affect the therapy administered to patients.

Colposcopes, the gold standard devices for imaging the cervix at high magnfication, are expensive and sparse in low resource settings. Using a lens attachment, any smartphone camera can be turned into an imaging device for tissues such as the cervix. We create a smartphone-based colposcope using a simple lens design for high magnification. This particular design is useful because it allows parameters such as F-number, depth of field, and magnification to be controlled easily. We were therefore able to determine a set of design steps which are general to mobile medical imaging devices and allow them to maintain requisite image quality while still being rugged and affordable.

Objective: Compare an inexpensive cell-phone based Mobile Colposcope, with a standard colposcope in the evaluation of women with abnormal Pap smear screening. Methodology: The study was a prospective, parallel noninferiority trial. Thirty women underwent colposcopy for the evaluation of an abnormal Pap smear. After application of acetic acid, images of the cervix were obtained with both a standard colposcope and the Mobile Colposcope. An additional set of images using both devices were obtained using the red-free (green filter) mode. Eight experienced gynecologists then evaluated 100 paired images (plain and green filter) from two different sites in random order using a web based assessment program. After reviewing each set of paired images, the expert would make an assessment of: 1) normal (no biopsy/ random biopsy), or 2) abnormal. For abnormal images, the expert then electronically marked the site(s) on the image where a biopsy was recommended. In image analysis, the cervical image was divided into 12 radial sectors and the marked sites for biopsy on the matched pairs were compared. Matched pairs that were considered normal, or those where biopsy site recommendations were within +/- 30° were considered equivalent; unmatched biopsy sites were considered non-equivalent. Results were compared using Wilcoxon Matched Pairs Signed Ranks Test. Expert assessment of Mobile Colposcope images compared with assessment by standard colposcope is currently onging. Conclusions: if the Mobile Colposcope demonstrates non-inferiority to imaging obtained with a standard colposcope and due to its low cost, it has the potential help improve cervical cancer screening in low resource settings.

Efficient management of chronic sinusitis remains a great challenge for primary care physicians. Unlike ENT specialists using Computed Tomography scans, they lack an affordable and safe method to accurately screen and monitor sinus diseases in primary care settings. Lack of evidence-based sinusitis management leads to frequent under-treatments and unnecessary over-treatments (i.e. antibiotics). Previously, we reported low-cost optical imaging designs for oral illumination and facial optical imaging setup. It exploits the sensitivity of NIR transmission intensity and their unique patterns to the sinus structures and presence of fluid/mucous-buildup within the sinus cavities. Using the improved NIR system, we have obtained NIR sinus images of 45 subjects with varying degrees of sinusitis symptoms. We made diagnoses of these patients based on two types of evidence: symptoms alone or NIR images along. These diagnostic results were then compared to the gold standard diagnosis using computed tomography through sensitivity and specificity analysis. Our results indicate that diagnosis of mere presence of sinusitis that is, distinguishing between healthy individuals vs. diseased individuals did not improve much when using NIR imaging compared to the diagnosis based on symptoms alone (69% in sensitivity, 75% specificity). However, use of NIR imaging improved the differential diagnosis between mild and severe diseases significantly as the sensitivity improved from 75% for using diagnosis based on symptoms alone up to 95% for using diagnosis based on NIR images. Reported results demonstrate great promise for using NIR imaging system for management of chronic sinusitis patients in primary care settings without resorting to CT.

A time-domain optical coherence tomography system with respect to multiple reflections in the reference arm is simulated with Zemax and the optimal detector position is discussed based on the wavefront error. Multiple reference optical coherence tomography (MR-OCT) is a novel optical imaging platform using a single miniature actuator and partial mirror to recirculate light multiple times within the interferometric system. The recirculation corresponds to multiple path delays extending effectively the axial scan-range. The simulation of MR-OCT with lenses of 5 mm diameter shows that it is suitable to build a compact system using low-cost components. Subsequently MR-OCT enables an affordable miniature OCT for low recourse settings and personal or point-of-care applications.

Over the past decade, crowd-sourcing complex image analysis tasks to a human crowd has emerged as an alternative to energy-inefficient and difficult-to-implement computational approaches. Following this trend, we have developed a mathematical framework for statistically combining human crowd-sourcing of biomedical image analysis and diagnosis through games. Using a web-based smart game (BioGames), we demonstrated this platform’s effectiveness for telediagnosis of malaria from microscopic images of individual red blood cells (RBCs). After public release in early 2012 (http://biogames.ee.ucla.edu), more than 3000 gamers (experts and non-experts) used this BioGames platform to diagnose over 2800 distinct RBC images, marking them as positive (infected) or negative (non-infected). Furthermore, we asked expert diagnosticians to tag the same set of cells with labels of positive, negative, or questionable (insufficient information for a reliable diagnosis) and statistically combined their decisions to generate a gold standard malaria image library. Our framework utilized minimally trained gamers’ diagnoses to generate a set of statistical labels with an accuracy that is within 98% of our gold standard image library, demonstrating the “wisdom of the crowd”. Using the same image library, we have recently launched a web-based malaria training and educational game allowing diagnosticians to compare their performance with their peers. After diagnosing a set of ~500 cells per game, diagnosticians can compare their quantified scores against a leaderboard and view their misdiagnosed cells. Using this platform, we aim to expand our gold standard library with new RBC images and provide a quantified digital tool for measuring and improving diagnostician training globally.

For simultaneous measurement of multi-components on-site like factories, the ultra-compact (diameter: 9[mm], length: 45[mm], weight: 200[g]) one-shot ATR (Attenuated Total Reflection) Fourier spectroscopic imager was proposed. Because the proposed one-shot Fourier spectroscopic imaging is based on spatial-phase-shift interferometer, interferograms could be obtained with simple optical configurations. We introduced the transmission-type relativeinclined phase-shifter, that was constructed with a cuboid prism and a wedge prism, onto the optical Fourier transform plane of infinity corrected optical systems. And also, small light-sources and cameras in the mid-infrared light region, whose size are several millimeter on a side, are essential components for the ultra-compact spectroscopic configuration. We selected the Graphite light source (light source area: 1.7×1.7[mm], maker: Hawkeye technologies) whose radiation factor was high. Fortunately, in these days we could apply the cost-effective 2-dimensional light receiving device for smartphone (e.g. product name: LEPTON, maker: FLIR, price: around 400USD). In the case of alcoholic drinks factory, conventionally workers measure glucose and ethanol concentrations by bringing liquid solution back to laboratories every day. The high portable spectroscopy will make it possible to measure multi-components simultaneously on manufacturing scene. But we found experimentally that absorption spectrum of glucose and water and ethanol were overlapped each other in near infrared light region. But for mid-infrared light region, we could distinguish specific absorption peaks of glucose (@10.5[μm]) and ethanol (@11.5[μm]) independently from water absorption. We obtained standard curve between absorption (@9.6[μm]) and ethanol concentration with high correlation coefficient 0.98 successfully by ATR imaging-type 2-dimensional Fourier spectroscopy (wavelength resolution: 0.057[μm]) with the graphite light source (maker: Hawkeye technologies, type: IR-75).

We propose the extremely-compact-size line-imaging Fourier spectroscopy for smartphones. We realize the near common-path interferometer with strong robustness for mechanical vibrations by installing the transmission-type relative-inclined phase-shifter. The interferogram of an imaging line is formed as 2-dimensional fringe pattern on imaging sensor, such as CCD camera. In other words, the horizontal axis on an imaging sensor is assigned to phase-shift value. And the vertical axis is corresponds to image formation coordinate. Thus, by installing a relatively-inclined thin glass into imaging optics, such as smartphone, we will realize the line-imaging Fourier spectroscopy for healthcare sensor in daily-life environments.

We proposed the imaging-type 2-dimensional Fourier spectroscopy that is a near-common-path interferometer with strong robustness against mechanical vibrations. We introduced the miniature uncooled infrared microbolometer arrays for smartphone (e.g. product name: FILR ONE price: around 400USD). And we constructed the phase-shifter with the piezo impact drive mechanism (maker: Technohands.co.Ltd., stroke: 4.5mm, resolution: 0.01μm, size: 20mm, price: around 800USD). Thus, we realized the palm-size mid-infrared spectroscopic imager [size: L56mm×W69mm×H43mm weight: 500g]. And by using wide-angle lens as objective lens, the proposed method can obtain the wide-field 2- dimensional middle-infrared (wavelength: 7.5-13.5[μm]) spectroscopic imaging of radiation lights emitted from human bodies itself

As a bio/chemical sensing technique, surface enhanced Raman spectroscopy (SERS) offers sensitivity comparable to that of fluorescence detection while providing highly specific information about the analyte. Although single molecule identification with SERS was demonstrated nearly 20 years ago, today a need exists to develop practical solutions for point‐of‐sample and point‐of‐care SERS systems. Recently, we demonstrated the fabrication of SERS substrates by inkjet printing silver and gold nanostructures onto paper and other microporous membranes. Using these devices, we have been able to achieve detection limits comparable to conventional nanofabricated plasmonic substrates. Furthermore, we leverage the fluidic properties of paper to enhance the performance of the SERS devices while also enabling unprecedented ease of use. Here we report the use of inkjet‐fabricated paper SERS substrates as a detection device for biological macromolecules in an easy‐to‐use format with a low number of steps. The targeted biomarker is specifically detected with SERS through a single step competitive displacement, which dramatically reduces the number of steps as compared to conventional assays. Moreover, we further improve the usability of the assay by incorporating a paper SERS device with a fluidic cartridge format. The wicking nature of the paper sensor eliminates manual sample application steps and is much simpler than the world‐to‐chip interface of microfluidic devices. The introduction of this paper‐based SERS assay is a significant step towards highly sensitive, low‐cost, and, importantly, easy to use multiplexed biological assays.

Giardia lamblia is a waterborne parasite that causes an intestinal infection, known as giardiasis, and it is found not only in countries with inadequate sanitation and unsafe water but also streams and lakes of developed countries. Simple, sensitive, and rapid detection of this pathogen is important for monitoring of drinking water. Here we present a cost-effective and field portable mobile-phone based fluorescence microscopy platform designed for automated detection of Giardia lamblia cysts in large volume water samples (i.e., 10 ml) to be used in low-resource field settings. This fluorescence microscope is integrated with a disposable water-sampling cassette, which is based on a flow-through porous polycarbonate membrane and provides a wide surface area for fluorescence imaging and enumeration of the captured Giardia cysts on the membrane. Water sample of interest, containing fluorescently labeled Giardia cysts, is introduced into the absorbent pads that are in contact with the membrane in the cassette by capillary action, which eliminates the need for electrically driven flow for sample processing. Our fluorescence microscope weighs ~170 grams in total and has all the components of a regular microscope, capable of detecting individual fluorescently labeled cysts under light-emitting-diode (LED) based excitation. Including all the sample preparation, labeling and imaging steps, the entire measurement takes less than one hour for a sample volume of 10 ml. This mobile phone based compact and cost-effective fluorescent imaging platform together with its machine learning based cyst counting interface is easy to use and can even work in resource limited and field settings for spatio-temporal monitoring of water quality.

Contamination of the water source and air pollution are two major problems that must be faced in the coming years. The increasing worldwide contamination of freshwater systems with thousands of industrial and natural chemical compounds is one of the key environmental problems facing humanity today. It is estimated that pathogens in water cause more than 2 million deaths annually. Additionally, traditional water quality assessment methods, such as liquid chromatography and mass spectroscopy, are expensive and time consuming from sample collection to analysis. Low cost tools are needed which can provide high sensitivity in sensing, while remaining portable and providing near real time analysis. Here, we present a low cost integrating cavity that can be used for highly sensitive environmental sensing.

Fluorescence microscopy offers a number of advantages for cell- and biomarker-based diagnostics with regards to ease of use and interpretation, sensitivity, and specificity. However, its use in low-resource settings is often hindered by the need for bulky microscopes with expensive excitation and filter setups. While many advances have been made towards utilizing smartphones as microscopes, there remains a reliance on complex attachments to facilitate fluorescence microscopy. Here, we report progress towards a filter-less fluorescent assay utilizing ultraviolet light, an unmodified smartphone, and pyrene-labeled aptamers. The pyrene monomer is excited at a wavelength of 350 nm and emits at approximately 390 nm; when two pyrene molecules are brought into close proximity, however, they form an excimer which emits at approximately 490 nm. We have engineered pyrene-conjugated DNA sequences such that the fluorophores, normally in monomeric configuration, are brought into proximity upon binding of the DNA to its target. The large Stokes shift between excitation and emission of the excimer allows us to detect such biorecognition events with an unfiltered smartphone camera, enabling the use of this assay in low-resource settings where portability and easeof- use are paramount.

The modern food processing and storage environments require the real-time monitoring and rapid microbiological testing. Optical spectroscopy with selective excitation wavelengths can be the basis of a novel, rapid, reagent less, noncontact and non-destructive technique for monitoring the food spoilage. The native fluorescence spectra of muscle foods stored at 2-4°C (in refrigerator) and 20-24°C (in room temperature) were measured as a function of time with a selective excitation wavelength of 340nm. The contributions of the principal molecular components to the native fluorescence spectra of meat were measured spectra of each fluorophore: collagen, reduced nicotinamide adenine dinucleotide (NADH), and flavin. The responsible components were extracted using a method namely Multivariate Curve Resolution with Alternating Least-Squares (MCR-ALS). The native fluorescence combined with MCR-ALS can be used directly on the surface of meat to produce biochemically interpretable “fingerprints”, which reflects the microbial spoilage of foods involved with the metabolic processes. The results show that with time elapse, the emission from NADH in meat stored at 24°C increases much faster than that at 4°C. This is because multiplying of microorganisms and catabolism are accompanied by the generation of NADH. This study presents changes of relative content of NADH may be used as criterion for detection of spoilage degree of meat using native fluorescence spectroscopy.