Multispectral Imaging Camera Research Articles

Application of Multispectral Drone for Quantitative Assessment of the Fertilizer Requirement for the Up-country Tea Plantation in Sri Lanka


 
 
 The amount of chlorophyll content and structure would differ between diseased and healthy plants, affecting the NDVI values, which strongly correlated with fertilizer applications. The primary goal of this research is to use a multispectral drone to quantify the fertilizer requirements for the sample tea plantations in Nuwara Eliya, Sri Lanka. In contrast, a drone with multispectral imaging camera sensors was used to gather images covering blue, green, red, red-edge, and near-infrared bands. The collected images were analyzed, and geographic outputs were generated. The processed geospatial data were used to determine the NDVI values for each tea field, and the general health of the tea was estimated. Finally, NDVI readings and geospatial products from each tea field related to soil analytical data. Based on the NDVI values, 25 selected soil samples were collected using sampling plots. The concentrations of pH, electrical conductivity (EC), nitrogen (N), phosphorus (P) and potassium (K) were then determined using standard soil testing procedures. Finally, soil analysis results were combined with the geospatial data. In addition, the amount of Dolomite for each tea field was calculated based on the pH values, and the amount of nitrogen, phosphorus, and potassium was calculated as yield and area and based on the number of tea bushes. The tea plantations show an NDVI range of 0.76-0.81 for low healthy tea, 0.81-0.86 for medium healthy tea and 0.86-0.97 for healthy tea. The healthiness percentages of one of the tea plantation show 18% healthy tea, 54% medium healthy tea, and 22% low healthy tea. Conversely, the healthiness percentages of the other tea plantation show 44% healthy tea, 39% medium healthy tea and 17% low healthy tea based on the NDVI value. The average value of pH, EC, N, P, and K are 4.84, 41.58, 0.29, 0.11, 0.04 in tea plantations, respectively, and in other planation, average values of pH, EC, N%, P%, and K% are 4.83, 40.96, 0.36, 0.25, 0.04 respectively. The amounts of Dolomite for each tea field are 2685.32 Kg/ha and 977.34 Kg/ha. In one tea field, an average annual NPK fertilizer need is 685 kg per hectare, 94 kg per hectare, and 269 kg per hectare, respectively. Additionally, calculations based on tea bushes indicate that the average annual NPK fertilizer requirements are 685 kg, 89 kg, and 254 kg, respectively. Other tea filed average annual NPK fertilizer need is 88 kg per hectare, 24 kg per hectare, and 68 kg per hectare, respectively. Additionally, calculations based on tea bushes indicate that the average annual NPK fertilizer requirements are 90 kg, 25 kg, and 70 kg, respectively. However, these values have several limitations, such as climate, soil, plant, and cultural practices. This study mainly focuses on and creates Insights into Plant Health by assessing precise plant-level data using a multispectral imaging system. It had not been correlated with other factors except soil conditions.
 Keywords: Unmanned aerial vehicles, Multispectral, NDVI, Tea, Fertilizer
 
 

A-197 - Novel, non-invasive method for classification of melanocytic lesions using a direct illumination multispectral imaging camera

A-197 - Novel, non-invasive method for classification of melanocytic lesions using a direct illumination multispectral imaging camera

INNV-29. MULTISPECTRAL ANALYSIS FOR INTRA-OPERATIVE CHARACTERIZATION OF BRAIN TUMOURS, MARGINS OF RESECTION, AND ELOQUENT AREAS ACTIVATION – PRELIMINARY RESULTS

Abstract BACKGROUND The identification of intrinsic brain tumours’ boundaries and their anatomical relationship with eloquent connectomes is one of the main challenges in neurosurgery. Several new intra-operative spectroscopic technologies have been recently investigated for this purpose. Multispectral imaging (MSI) analysis is based on analysis of reflected light, divided in specific wavelengths ranges across the electromagnetic spectrum. By doing so, it can identify wavelengths invisible to the naked eye, and make them visible by applying specific filters. METHODS We have used MSI to acquire intraoperative images during brain tumours surgeries. The MSI camera has been adapted to be connected to the intra-operative microscope and acquire a stack of images of different spectra to be analysed on a separate time. The focus of the images acquisition includes tumour structure, its margins, and eloquent connectomes during awake surgery or neurostimulation to analyse the possibility of different perfusion signal during activation. Image segmentation using K-mean clustering has been used to detect structural information and identify tumour’s area and tumour’s margins, while oxygen maps have been created using linear least squares regression to detect areas of metabolic activity. RESULTS Twenty-three patients have been enrolled in the current study at the time of this submission. We have analysed the intra-operative acquisition using intra-operative microscope picture, pre- and post-operative MRI scan, and neurophysiological findings as ground truth. Our preliminary results show that MSI is a potential additional technology to visually show tumour core, margins, and different degree of perfusions on the tissues analysed during tumour surgery. Although on a very limited subgroup of patients, MSI seems to be reliable in identify eloquent areas activation during awake surgery. These results are promising and warrant more data acquisition to be confirmed.

CD13: Determining the spectral response of positive patch test results with the view of developing a multispectral imaging camera to improve the accuracy of virtual patch testing

CD13: Determining the spectral response of positive patch test results with the view of developing a multispectral imaging camera to improve the accuracy of virtual patch testing

Spectral technology and multispectral imaging for estimating the photosynthetic pigments and SPAD of the Chinese cabbage based on machine learning

• The estimation models constructed by spectra and multispectral imaging combined with various machine learning algorithms were compared. • A one-dimensional convolutional neural network model based on spectral data was established. • A two-dimensional convolutional neural network model based on multispectral images was established. • Spectra and multispectral images were combined with deep learning to achieve the estimation of multiple indicators, such as photosynthetic pigments and SPAD. The contents of photosynthetic pigment, which directly affect the growth of crops, could be evaluated with spectral and multispectral imaging technologies in an accurate and rapid way. For commodity germplasm resources on the market, the estimation of photosynthetic pigments and soil and plant analyzer development (SPAD) value was accomplished using the two techniques combined with machine learning. The spectrometer used in this study employed 781 bands from 320 nm to 1100 nm, and a multispectral imaging camera was used to acquire images in visible and near-infrared. Convolutional neural network (CNN), multiple linear regression (MLR) and generalized linear model (GLM) were used to establish the machine learning models, which established by preprocessed spectral data or 4-channel multispectral images. For estimating photosynthetic pigments (chlorophyll a , chlorophyll b , total chlorophyll and carotenoids), the GLM model established by spectral data was the optimal among all the models. For the SPAD optimal estimation model, the GLM model established by the spectral data and CNN model established by the multispectral images were fair. The R 2 and RMSE of the CNN model validation set in estimating SPAD were 0.87 and 2.31, respectively. The R 2 and RMSE of the GLM model validation set in estimating SPAD were 0.88 and 2.39, respectively. By combining two techniques with different machine learning methods, a comprehensive analysis of photosynthetic pigments and SPAD was accomplished in this paper.

A comparison of ground-based hyperspectral imaging and red-edge multispectral imaging for façade material classification

This paper compares the feasibility of classifying building façade materials in situ via a low-cost multispectral imaging camera with 5 bands compared to a high-cost, short-wave infrared hyperspectral unit with 240 bands. The sensors were used to classify five common façade materials: brick, mortar, stone, painted window frames, and rendering. The data were subjected to a Partial Least Squares Discriminant Analysis by selecting 95 training pixels and 65 validation pixels for each material. Across all materials, the hyperspectral imaging demonstrated a moderate advantage over multispectral imaging (with a correct classification rate of 99% versus 80%) for the selected data and stronger qualitative matching to the entire façade. While the hyperspectral data were consistently superior, the multispectral data still produced valuable results, thereby demonstrating potential as a fast, easy, and much cheaper technology compared to hyperspectral units for in situ façade materials classification.

Pluto’s Haze Abundance and Size Distribution from Limb Scatter Observations by MVIC

Abstract The New Horizons spacecraft observed Pluto and Charon at solar-phase angles between 16° and 169°. In this work, we use the Multispectral Visible Imaging Camera (MVIC) observations to construct multiwavelength phase curves of Pluto’s atmosphere, using the limb scatter technique. Observational artifacts and biases were removed using Charon as a representative airless body. The size and distribution of the haze particles were constrained using a Titan fractal aggregate phase function. We find that monodispersed and log-normal populations cannot simultaneously describe the observed steep forward scattering, indicative of wavelength-scale particles, and the non-negligible backscattering indicative of particles much smaller than the wavelength. Instead, we find it necessary to use bimodal or power-law distributions, especially below ∼200 km, to properly describe the MVIC observations. Above 200 km, where the atmosphere is isotropically scattering, a monodisperse, log-normal, or a bimodal/power law approximating a monodispersed population is able to fit the phase curves well. As compared to the results of previously published articles, we find that Pluto’s atmosphere must contain haze particle number densities an order of magnitude greater for small (∼10 nm) and large (∼1 μm) radii, and relatively fewer intermediate sizes (∼100 nm). These conclusions support a lower aggregate aerosol growth rate than that found by Gao et al., indicating a higher charge-to-radius ratio, upwards of 60e − μm−1. In order to generate large particles with a lower growth rate, the atmosphere must also have a lower sedimentation velocity (<∼0.01 m s−1 at 200 km), which is possible with a fractal dimension of less than 2.

Pluto's Sputnik Planitia: Composition of geological units from infrared spectroscopy

We have compared spectroscopic data of Sputnik Planitia on Pluto, as acquired by New Horizons' Linear Etalon Imaging Spectral Array (LEISA) instrument, to the geomorphology as mapped by White et al. (2017) using visible and panchromatic imaging acquired by the LOng-Range Reconnaissance Imager (LORRI) and the Multi-spectral Visible Imaging Camera (MVIC). We have focused on 13 of the geologic units identified by White et al. (2017), which include the plains and mountain units contained within the Sputnik basin. We divided the map of Sputnik Planitia into 15 provinces, each containing one or more geologic units, and we use LEISA to calculate the average spectra of the units inside the 15 provinces. Hapke-based modeling was then applied to the average spectra of the units to infer their surface composition, and to determine if the composition resulting from the modeling of LEISA spectra reflects the geomorphologic analyses of LORRI data, and if areas classified as being the same geologically, but which are geographically separated, share a similar composition. We investigated the spatial distribution of the most abundant ices on Pluto's surface - CH4, N2, CO, H2O, and a non-ice component presumed to be a macromolecular carbon-rich material, termed a tholin, that imparts a positive spectral slope in the visible spectral region and a negative spectral slope longward of ~1.1 μm. Because the exact nature of the non-ice component is still debated and because the negative spectral slope of the available tholins in the near infrared does not perfectly match the Pluto data, for spectral modeling purposes we reference it generically as the negative spectral slope endmember (NSS endmember). We created maps of variations in the integrated band depth (from LEISA data) and areal mass fraction (from the modeling) of the components. The analysis of correlations between the occurrences of the endmembers in the geologic units led to the observation of an anomalous suppression of the strong CH4 absorption bands in units with compositions that are dominated by H2O ice and the NSS endmember. Exploring the mutual variation of the CH4 and N2 integrated band depths with the abundance of crystalline H2O and NSS endmember revealed that the NSS endmember is primarily responsible for the suppression of CH4 absorptions in mountainous units located along the western edge of Sputnik Planitia. Our spectroscopic analyses have provided additional insight into the geological processes that have shaped Sputnik Planitia. A general increase in volatile abundance from the north to the south of Sputnik Planitia is observed. Such an increase first observed and interpreted by Protopapa et al., 2017 and later confirmed by climate modeling (Bertrand et al., 2018) is expressed geomorphologically in the form of preferential deposition of N2 ice in the upland and mountainous regions bordering the plains of southern Sputnik Planitia. Relatively high amounts of pure CH4 are seen at the southern Tenzing Montes, which are a natural site for CH4 deposition owing to their great elevation and the lower insolation they are presently receiving. The NSS endmember correlates the existence of tholins within certain units, mostly those coating the low-latitude mountain ranges that are co-latitudinal with the tholin-covered Cthulhu Macula. The spectral analysis has also revealed compositional differences between the handful of occurrences of northern non-cellular plains and the surrounding cellular plains, all of which are located within the portion of Sputnik Planitia that is presently experiencing net sublimation of volatiles, and which do not therefore exhibit a surface layer of bright, freshly-deposited N2 ice. The compositional differences between the cellular and non-cellular plains here hint at the effectiveness of convection in entraining and trapping tholins within the body of the cellular plains, while preventing the spread of such tholins to abutting non-cellular plains.

A single sensor based multispectral imaging camera using a narrow spectral band color mosaic integrated on the monochrome CMOS image sensor

A multispectral image camera captures image data within specific wavelength ranges in narrow wavelength bands across the electromagnetic spectrum. Images from a multispectral camera can extract a additional information that the human eye or a normal camera fails to capture and thus may have important applications in precision agriculture, forestry, medicine, and object identification. Conventional multispectral cameras are made up of multiple image sensors each fitted with a narrow passband wavelength filter and optics, which makes them heavy, bulky, power hungry, and very expensive. The multiple optics also create an image co-registration problem. Here, we demonstrate a single sensor based three band multispectral camera using a narrow spectral band red–green–blue color mosaic in a Bayer pattern integrated on a monochrome CMOS sensor. The narrow band color mosaic is made of a hybrid combination of plasmonic color filters and a heterostructured dielectric multilayer. The demonstrated camera technology has reduced cost, weight, size, and power by almost n times (where n is the number of bands) compared to a conventional multispectral camera.

Phase Curves of Nix and Hydra from the New Horizons Imaging Cameras

Abstract NASA’s New Horizons spacecraft’s voyage through the Pluto system centered on 2015 July 14 provided images of Pluto’s small satellites Nix and Hydra at viewing angles unattainable from Earth. Here, we present solar phase curves of the two largest of Pluto’s small moons, Nix and Hydra, observed by the New Horizons LOng Range Reconnaissance Imager and Multi-spectral Visible Imaging Camera, which reveal the scattering properties of their icy surfaces in visible light. Construction of these solar phase curves enables comparisons between the photometric properties of Pluto’s small moons and those of other icy satellites in the outer solar system. Nix and Hydra have higher visible albedos than those of other resonant Kuiper Belt objects and irregular satellites of the giant planets, but not as high as small satellites of Saturn interior to Titan. Both Nix and Hydra appear to scatter visible light preferentially in the forward direction, unlike most icy satellites in the outer solar system, which are typically backscattering.

Visible/near‐infrared spectral diversity from in situ observations of the Bagnold Dune Field sands in Gale Crater, Mars

AbstractAs part of the Bagnold Dune campaign conducted by Mars Science Laboratory rover Curiosity, visible/near‐infrared reflectance spectra of dune sands were acquired using Mast Camera (Mastcam) multispectral imaging (445–1013 nm) and Chemistry and Camera (ChemCam) passive point spectroscopy (400–840 nm). By comparing spectra from pristine and rover‐disturbed ripple crests and troughs within the dune field, and through analysis of sieved grain size fractions, constraints on mineral segregation from grain sorting could be determined. In general, the dune areas exhibited low relative reflectance, a weak ~530 nm absorption band, an absorption band near 620 nm, and a spectral downturn after ~685 nm consistent with olivine‐bearing sands. The finest grain size fractions occurred within ripple troughs and in the subsurface and typically exhibited the strongest ~530 nm bands, highest relative reflectances, and weakest red/near‐infrared ratios, consistent with a combination of crystalline and amorphous ferric materials. Coarser‐grained samples were the darkest and bluest and exhibited weaker ~530 nm bands, lower relative reflectances, and stronger downturns in the near‐infrared, consistent with greater proportions of mafic minerals such as olivine and pyroxene. These grains were typically segregated along ripple crests and among the upper surfaces of grain flows in disturbed sands. Sieved dune sands exhibited progressive decreases in reflectance with increasing grain size, as observed in laboratory spectra of olivine size separates. The continuum of spectral features observed between the coarse‐ and fine‐grained dune sands suggests that mafic grains, ferric materials, and air fall dust mix in variable proportions depending on aeolian activity and grain sorting.

Focus on microscopy, imaging, and nanotechnology

Focus on microscopy, imaging, and nanotechnology

Haze in Pluto's atmosphere

Haze in Pluto's atmosphere was detected in images by both the Long Range Reconnaissance Imager (LORRI) and the Multispectral Visible Imaging Camera (MVIC) on New Horizons. LORRI observed haze up to altitudes of at least 200km above Pluto's surface at solar phase angles from ∼20° to ∼169°. The haze is structured with about ∼20 layers, and the extinction due to haze is greater in the northern hemisphere than at equatorial or southern latitudes. However, more haze layers are discerned at equatorial latitudes. A search for temporal variations found no evidence for motions of haze layers (temporal changes in layer altitudes) on time scales of 2 to 5 hours, but did find evidence of changes in haze scale height above 100km altitude. An ultraviolet extinction attributable to the atmospheric haze was also detected by the ALICE ultraviolet spectrograph on New Horizons. The haze particles are strongly forward-scattering in the visible, and a microphysical model of haze is presented which reproduces the visible phase function just above the surface with 0.5µm spherical particles, but also invokes fractal aggregate particles to fit the visible phase function at 45km altitude and account for UV extinction. A model of haze layer generation by orographic excitation of gravity waves is presented. This model accounts for the observed layer thickness and distribution with altitude. Haze particles settle out of the atmosphere and onto Pluto's surface, at a rate sufficient to alter surface optical properties on seasonal time scales. Pluto's regional scale albedo contrasts may be preserved in the face of the haze deposition by atmospheric collapse.

Inflight radiometric calibration of New Horizons’ Multispectral Visible Imaging Camera (MVIC)

We discuss two semi-independent calibration techniques used to determine the inflight radiometric calibration for the New Horizons’ Multi-spectral Visible Imaging Camera (MVIC). The first calibration technique compares the measured number of counts (DN) observed from a number of well calibrated stars to those predicted using the component-level calibration. The ratio of these values provides a multiplicative factor that allows a conversation between the preflight calibration to the more accurate inflight one, for each detector. The second calibration technique is a channel-wise relative radiometric calibration for MVIC's blue, near-infrared and methane color channels using Hubble and New Horizons observations of Charon and scaling from the red channel stellar calibration. Both calibration techniques produce very similar results (better than 7% agreement), providing strong validation for the techniques used. Since the stellar calibration described here can be performed without a color target in the field of view and covers all of MVIC's detectors, this calibration was used to provide the radiometric keyword values delivered by the New Horizons project to the Planetary Data System (PDS). These keyword values allow each observation to be converted from counts to physical units; a description of how these keyword values were generated is included. Finally, mitigation techniques adopted for the gain drift observed in the near-infrared detector and one of the panchromatic framing cameras are also discussed.

Charon's light curves, as observed by New Horizons’ Ralph color camera (MVIC) on approach to the Pluto system

Light curves produced from color observations taken during New Horizons’ approach to the Pluto-system by its Multi-spectral Visible Imaging Camera (MVIC, part of the Ralph instrument) are analyzed. Fifty seven observations were analyzed, they were obtained between 9th April and 3rd July 2015, at a phase angle of 14.5° to 15.1°, sub-observer latitude of 51.2 °N to 51.5 °N, and a sub-solar latitude of 41.2°N. MVIC has four color channels; all are discussed for completeness but only two were found to produce reliable light curves: Blue (400–550nm) and Red (540–700nm). The other two channels, Near Infrared (780–975nm) and Methane-Band (860–910nm), were found to be potentially erroneous and too noisy respectively. The Blue and Red light curves show that Charon's surface is neutral in color, but slightly brighter on its Pluto-facing hemisphere. This is consistent with previous studies made with the Johnson B and V bands, which are at shorter wavelengths than that of the MVIC Blue and Red channel respectively.

USE OF A MULTISPECTRAL UAV PHOTOGRAMMETRY FOR DETECTION AND TRACKING OF FOREST DISTURBANCE DYNAMICS

This study presents a new methodological approach for assessment of spatial and qualitative aspects of forest disturbance based on the use of multispectral imaging camera with the UAV photogrammetry. We have used the miniaturized multispectral sensor Tetracam Micro Multiple Camera Array (μ-MCA) Snap 6 with the multirotor imaging platform to get multispectral imagery with high spatial resolution. The study area is located in the Sumava Mountains, Central Europe, heavily affected by windstorms, followed by extensive and repeated bark beetle (Ips typographus [L.]) outbreaks in the past 20 years. After two decades, there is apparent continuous spread of forest disturbance as well as rapid regeneration of forest vegetation, related with changes in species and their diversity. For testing of suggested methodology, we have launched imaging campaign in experimental site under various stages of forest disturbance and regeneration. The imagery of high spatial and spectral resolution enabled to analyse the inner structure and dynamics of the processes. The most informative bands for tree stress detection caused by bark beetle infestation are band 2 (650nm) and band 3 (700nm), followed by band 4 (800 nm) from the, red-edge and NIR part of the spectrum. We have identified only three indices, which seems to be able to correctly detect different forest disturbance categories in the complex conditions of mixture of categories. These are Normalized Difference Vegetation Index (NDVI), Simple 800/650 Ratio Pigment specific simple ratio B1 and Red-edge Index.

USE OF A MULTISPECTRAL UAV PHOTOGRAMMETRY FOR DETECTION AND TRACKING OF FOREST DISTURBANCE DYNAMICS

Abstract. This study presents a new methodological approach for assessment of spatial and qualitative aspects of forest disturbance based on the use of multispectral imaging camera with the UAV photogrammetry. We have used the miniaturized multispectral sensor Tetracam Micro Multiple Camera Array (μ-MCA) Snap 6 with the multirotor imaging platform to get multispectral imagery with high spatial resolution. The study area is located in the Sumava Mountains, Central Europe, heavily affected by windstorms, followed by extensive and repeated bark beetle (Ips typographus [L.]) outbreaks in the past 20 years. After two decades, there is apparent continuous spread of forest disturbance as well as rapid regeneration of forest vegetation, related with changes in species and their diversity. For testing of suggested methodology, we have launched imaging campaign in experimental site under various stages of forest disturbance and regeneration. The imagery of high spatial and spectral resolution enabled to analyse the inner structure and dynamics of the processes. The most informative bands for tree stress detection caused by bark beetle infestation are band 2 (650nm) and band 3 (700nm), followed by band 4 (800 nm) from the, red-edge and NIR part of the spectrum. We have identified only three indices, which seems to be able to correctly detect different forest disturbance categories in the complex conditions of mixture of categories. These are Normalized Difference Vegetation Index (NDVI), Simple 800/650 Ratio Pigment specific simple ratio B1 and Red-edge Index.

Assessment of advanced glycated end product accumulation in skin using auto fluorescence multispectral imaging

Several studies have shown that advanced glycation end products (AGE) play a role in both the microvascular and macrovascular complications of diabetes and are closely linked to inflammation and atherosclerosis. AGEs accumulate in skin and can be detected using their auto fluorescence (AF).A significant correlation exists between AGE AF and the levels of AGEs as obtained from skin biopsies. A commercial device, the AGE Reader, has become available to assess skin AF for clinical purposes but, while displaying promising results, it is limited to single-point measurements performed in contact to skin tissue. Furthermore, in vivo imaging of AGE accumulation is virtually unexplored.We proposed a non-invasive, contact-less novel technique for quantifying fluorescent AGE deposits in skin tissue using a multispectral imaging camera setup (MSI) during ultraviolet (UV) exposure. Imaging involved applying a region-of-interest mask, avoiding specular reflections and a simple calibration. Results of a study conducted on 16 subjects with skin types ranging from fair to deeply pigmented skin, showed that AGE measured with MSI in forearm skin was significantly correlated with the AGE reference method (AGE Reader on forearm skin, R=0.68, p=0.005). AGE measured in facial skin was borderline significantly related to AGE Reader on forearm skin (R=0.47, p=0.078). These results support the use of the technique in devices for non-touch measurement of AGE content in either facial or forearm skin tissue over time.

Diagnostic accuracy of diffuse reflectance imaging for early detection of pre-malignant and malignant changes in the oral cavity: a feasibility study

BackgroundDiffusely reflected light is influenced by cytologic and morphologic changes that take place during tissue transformation, such as, nuclear changes, extracellular matrix structure and composition as well as blood flow. Albeit with varying degree of sensitivity and specificity, the properties of diffusely reflected light in discriminating a variety of oral lesions have been demonstrated by our group in multiple studies using point monitoring systems. However, the point monitoring system could not identify the region with the most malignant potential in a single sitting.MethodsIn order to scan the entire lesion, we developed a multi-spectral imaging camera system that records diffuse reflectance (DR) images of the oral lesion at 545 and 575 nm with white light illumination. The diagnostic accuracy of the system for 2-dimensional DR imaging of pre-malignant and malignant changes in the oral cavity was evaluated through a clinical study in 55 patients and 23 healthy volunteers. The DR imaging data were compared with gold standard tissue biopsy and histopathology results.ResultsIn total 106- normal/clinically healthy sites, 20- pre-malignant and 29- malignant (SCC) sites were compared. While the median pixel value of the R545/R575 image ratio for normal/clinically healthy tissue was 0.87 (IQR = 0.82-0.94), they were 1.35 (IQR = 1.13-1.67) and 2.44 (IQR = 1.78-3.80) for pre-malignant and malignant lesions, respectively. Area under the ROC curve to differentiate malignant from normal/clinically healthy [AUC = 0.99 (95% CI: 0.99-1.00)], pre-malignant from normal/clinically healthy [AUC = 0.94 (95% CI: 0.86-1.00)], malignant from pre-malignant [AUC = 0.84 (95% CI: 0.73-0.95)] and pre-malignant and malignant from normal/clinically healthy [AUC = 0.97 (95% CI: 0.94-1.00)] lesions were desirable.ConclusionWe find DR imaging to be very effective as a screening tool in locating the potentially malignant areas of oral lesions with relatively good diagnostic accuracy while comparing it to the gold standard histopathology.

Development of automated quantitative multiparameter immunocytochemical profiling of circulating tumor cells (CTCs) in breast cancer (BC).

e21124 Background: The currently accepted definition of a CTC is a nucleated cell that is positive by qualitative fluorescence for EpCAM, expresses cytokeratin, and is negative for the pan-hematopoietic marker CD45. Such approach has limitations such as operator bias, difficulty distinguishing true signal from autofluoresence or filter bleed-through, and most importantly, it cannot differentiate expression levels that are low, intermediate or high. This is a problem because CTCs are in fact heterogeneous with respect to phenotypic expression of various markers. Methods: Peripheral blood samples were subjected to our previously described, negative depletion technology. Cytospins of enriched samples were stained with 3 to 5 antibodies conjugated to various Alexa Fluor dyes plus a nuclear dye and then analyzed with both confocal microscopy and a traditional epifluorescence microscope equipped with a Nuance (CRi) multispectral imaging camera and associated software. This imaging system allows not only fluorescent signals to be quantified between specific excitation and emission filters, but also the spectral structure between these boundaries through the use of a liquid crystal that takes images in as little as 10nm increments. Such imaging allows spectra from specific dyes to identify from the contributions of other dyes, filter leakage, and auto fluorescence. Results: Using this technology, and appropriate controls, we are developing continuous variable quantification of the expression of: cytokeratins, vimentin, CD45, Her2, CD44, EpCAM and EGFR in negatively enriched, peripheral blood samples from metastatic breast cancer patients. To date, over 10 metastatic patient samples have been quantified and the study is ongoing. Initial quantification indicates that significant variability in expression levels exists in multiple markers including Her2 neu. Conclusions: Using multispectral technology, we have shown that CTCs are heterogeneous, with different CTCs having varying expression levels of EpCAM, HER2 neu, vimentin and EGFR. Further multiparameter quantitative characterization with up to 6 color analysis on the same cell is in progress.