Image-based Algorithm Research Articles

An Image-Based Algorithm for Precise and Accurate High Throughput Assessment of Drug Activity against the Human Parasite Trypanosoma cruzi

We present a customized high content (image-based) and high throughput screening algorithm for the quantification of Trypanosoma cruzi infection in host cells. Based solely on DNA staining and single-channel images, the algorithm precisely segments and identifies the nuclei and cytoplasm of mammalian host cells as well as the intracellular parasites infecting the cells. The algorithm outputs statistical parameters including the total number of cells, number of infected cells and the total number of parasites per image, the average number of parasites per infected cell, and the infection ratio (defined as the number of infected cells divided by the total number of cells). Accurate and precise estimation of these parameters allow for both quantification of compound activity against parasites, as well as the compound cytotoxicity, thus eliminating the need for an additional toxicity-assay, hereby reducing screening costs significantly. We validate the performance of the algorithm using two known drugs against T.cruzi: Benznidazole and Nifurtimox. Also, we have checked the performance of the cell detection with manual inspection of the images. Finally, from the titration of the two compounds, we confirm that the algorithm provides the expected half maximal effective concentration (EC50) of the anti-T. cruzi activity.

Temporal Coherence in Image-Based Visual Hull Rendering

Image-based visual hull rendering is a method for generating depth maps of a desired viewpoint from a set of silhouette images captured by calibrated cameras. It does not compute a view-independent data representation, such as a voxel grid or a mesh, which makes it particularly efficient for dynamic scenes. When users are captured, the scene is usually dynamic, but does not change rapidly because people move smoothly within a subsecond time frame. Exploiting this temporal coherence to avoid redundant calculations is challenging because of the lack of an explicit data representation. This paper analyzes the image-based visual hull algorithm to find intermediate information that stays valid over time and is, therefore, worth to make explicit. We then derive methods that exploit this information to improve the rendering performance. Our methods reduce the execution time by up to 25 percent. When the user's motions are very slow, reductions of up to 50 percent are achieved.

Impact of subject head motion on quantitative brain 15O PET and its correction by image-based registration algorithm

Subject head motion during sequential (15)O positron emission tomography (PET) scans can result in artifacts in cerebral blood flow (CBF) and oxygen metabolism maps. However, to our knowledge, there are no systematic studies examining this issue. Herein, we investigated the effect of head motion on quantification of CBF and oxygen metabolism, and proposed an image-based motion correction method dedicated to (15)O PET study, correcting for transmission-emission mismatch and inter-scan mismatch of emission scans. We analyzed (15)O PET data for patients with major arterial steno-occlusive disease (n = 130) to determine the occurrence frequency of head motion during (15)O PET examination. Image-based motion correction without and with realignment between transmission and emission scans, termed simple and 2-step method, respectively, was applied to the cases that showed severe inter-scan motion. Severe inter-scan motion (>3 mm translation or >5° rotation) was observed in 27 of 520 adjacent scan pairs (5.2 %). In these cases, unrealistic values of oxygen extraction fraction (OEF) or cerebrovascular reactivity (CVR) were observed without motion correction. Motion correction eliminated these artifacts. The volume-of-interest (VOI) analysis demonstrated that the motion correction changed the OEF on the middle cerebral artery territory by 17.3 % at maximum. The inter-scan motion also affected CBV, CMRO2 and CBF, which were improved by the motion correction. A difference of VOI values between the simple and 2-step method was also observed. These data suggest that image-based motion correction is useful for accurate measurement of CBF and oxygen metabolism by (15)O PET.

Reproducibility in the automated quantitative assessment of HER2/neu for breast cancer

Background: With the emerging role of digital imaging in pathology and the application of automated image-based algorithms to a number of quantitative tasks, there is a need to examine factors that may affect the reproducibility of results. These factors include the imaging properties of whole slide imaging (WSI) systems and their effect on the performance of quantitative tools. This manuscript examines inter-scanner and inter-algorithm variability in the assessment of the commonly used HER2/neu tissue-based biomarker for breast cancer with emphasis on the effect of algorithm training. Materials and Methods: A total of 241 regions of interest from 64 breast cancer tissue glass slides were scanned using three different whole-slide imagers and were analyzed using two different automated image analysis algorithms, one with preset parameters and another incorporating a procedure for objective parameter optimization. Ground truth from a panel of seven pathologists was available from a previous study. Agreement analysis was used to compare the resulting HER2/neu scores. Results: The results of our study showed that inter-scanner agreement in the assessment of HER2/neu for breast cancer in selected fields of view when analyzed with any of the two algorithms examined in this study was equal or better than the inter-observer agreement previously reported on the same set of data. Results also showed that discrepancies observed between algorithm results on data from different scanners were significantly reduced when the alternative algorithm that incorporated an objective re-training procedure was used, compared to the commercial algorithm with preset parameters. Conclusion: Our study supports the use of objective procedures for algorithm training to account for differences in image properties between WSI systems.

Assessing facial wrinkles: automatic detection and quantification

As people mature, their skin gradually presents lines, wrinkles, and folds that become more pronounced with time. Skin wrinkles are perceived as important cues in communicating information about the age of the person. Nowadays, documenting the facial appearance through imaging is prevalent in skin research, therefore detection and quantitative assessment of the degree of facial wrinkling can be a useful tool for establishing an objective baseline and for assessing benefits to facial appearance due to various dermatological treatments. However, few image-based algorithms for computationally assessing facial wrinkles are present in the literature, and those that exist have limited reliability. In this work, an algorithm for automatic detection of facial wrinkles is developed, based on estimating the orientation and the frequency of elongated spatial features, captured via digital image filtering. The algorithm is tested against one set of clinically validated 11-point wrinkle scales present on the face. The algorithm is employed for assessing the presence of forehead furrows on a set of 100 clinically graded facial images. The proposed computational assessment correlates well with the corresponding clinical scores. We find that the results are in better agreement with clinical scoring when the wrinkle depth information, approximated via filter responses, is combined with the wrinkle length information as opposed to the case when the two measures are considered separately.

Survey of Emergency Management Professionals to Assess Ideal Characteristics of a Photographic-Based Family Reunification Tool

A reunification tool that captures images of children at the time of the disaster would enable parents to locate their missing children, particularly if the children are unable to communicate their identity. This study assessed the ideal features and parameters of a photographic-based reunification tool. A convenience sample of federal, state, and hospital-based emergency management professionals were surveyed to elicit their preferences regarding an image-based reunification algorithm, to assess the parents' level of difficulty in viewing images with facial trauma, and to determine the minimum percentage of successful reunifications needed to justify adoption of a reunification tool. Of 322 emergency management professionals surveyed, 129 (40%) responded. Only 18% favored a photographic-based tool that would display images in which only the categories of age, gender, and facial features (eye, hair, and skin color) would exactly match the parent's description of the child. However, 72% preferred a broader, less-rigid system in which the images displayed would match all or most features in the parents' description of the missing child, allowing parents to view more of the image database. Most (85%) preferred a tool showing unedited images of living children, allowing parents to view facial trauma. However, more respondents reported that parents would find viewing unedited images with facial trauma somewhat or very difficult emotionally compared with edited images for both living (77% vs 20%, P < .001) and deceased children (91% vs 70%, P < .001.) In a disaster involving 1000 children, a tool that reunites a minimum of 10% of families would be adopted by over 50% of the participants. Participants were willing to accept a lower percentage of reunifications in a disaster involving 1000 children compared with disasters involving 10 (P < .001) or 100 children. (P < .001). Emergency management professionals identified desirable characteristics of a photographic-based reunification tool, including an algorithm displaying unedited photographs of missing children that loosely matches the parents' description, acknowledging the parents' emotional difficulty in viewing photographs with facial trauma. Participants were also willing to accept a lower percentage of successful reunifications as the scale of the disaster size increased.

The Hardware Solution of a New Image Processing Algorithm

This paper presents a new image-based parking space detection algorithm, the algorithm is based on a joint decision on multi-feature image processing algorithms, according to the data threshold of variance, correlation and edge-point density joint adjudicate the seize of paring space, it has a high recognition rate. And then the authors transplanted this algorithm into the FPGA platform, the hardware solution are implemented. And according to the algorithm characteristics the authors improved and optimized it. And the final, a compared test between hardware and software algorithms was executed, which results show that, the hardware algorithm for the solution is not only maintain a high recognition rate, but also more efficient.

Real-time image-based rigid registration of three-dimensional ultrasound

Registration of three-dimensional ultrasound (3DUS) volumes is necessary in several applications, such as when stitching volumes to expand the field of view or when stabilizing a temporal sequence of volumes to cancel out motion of the probe or anatomy. Current systems that register 3DUS volumes either use external tracking systems (electromagnetic or optical), which add expense and impose limitations on acquisitions, or are image-based methods that operate offline and are incapable of providing immediate feedback to clinicians. This paper presents a real-time image-based algorithm for rigid registration of 3DUS volumes designed for acquisitions in which small probe displacements occur between frames. Described is a method for feature detection and descriptor formation that takes into account the characteristics of 3DUS imaging. Volumes are registered by determining a correspondence between these features. A global set of features is maintained and integrated into the registration, which limits the accumulation of registration error. The system operates in real-time (i.e. volumes are registered as fast or faster than they are acquired) by using an accelerated framework on a graphics processing unit. The algorithm's parameter selection and performance is analyzed and validated in studies which use both water tank and clinical images. The resulting registration accuracy is comparable to similar feature-based registration methods, but in contrast to these methods, can register 3DUS volumes in real-time.

Image-based algorithm for analysis of transient trapping in single-particle trajectories

Particle tracking has become an increasingly useful tool in microfluidics and biophysics, allowing measurement of microrheology, local structure, and flow. We introduce a novel, automated approach to analyze single-particle trajectories with transient elements, based on image-processing approaches and physical analysis of probe motion. In many physical and active biological systems, such as living cells, probe particles experience thermally mediated Brownian motion combined with active transport processes that can lead to transient-trajectories of local diffusion and trapping, punctuated by segments of active transport. Analyzing such a trajectory as a single unit masks the intermittent nature of the motion. Moreover, directly applying the generalized Stokes–Einstein relation in out-of-equilibrium systems is incorrect and returns inaccurate rheological parameters. We present an automated image-processing-based method to identify and segment transient trap-escape trajectories, allowing quantitative analysis of each segment. We define and discuss effects of controlling parameters, such as particle size and camera frame rate. Our algorithm provides a general and automated method to segment and analyze transient elements in trajectories of single particles, which can be applied to many different experiments. Our image-based approach allows identification of trapping segments, unbiased by specific step sizes within those traps or the mechanism driving those steps. As an example, we successfully apply this method to experiments of laser tweezers trapped particles and show that trajectory segmentation allows us to calculate both trap and fluid parameters. We accurately identify a round trap, calculate the trap stiffness at 3.1 pN/μm, and find that significant local heating reduces fluid viscosity.

Image-based motion planning for two mobile robots in an environment with obstacles by using cellular neural networks

Abstract The paper presents an image-based algorithm for motion-planning of two mobile robots moving to the same target in an environment with obstacles. Due to parallel computing, the Cellular Neural Networks (CNN) techniques ensure the images processing in real-time and represent an advantageous solution for autonomous mobile robots guidance. The path planning algorithm can be improved increasing the speed of image processing, using advanced type of the CNN implementation and it can be extended for three or more robots.

Abstract 3185: Use of whole slide digital imaging for determination of caPCNA and Ki-67 labeling index in normal breast reductions and DCIS: Comparison of the two biomarkers using the Aperio image-based positive pixel algorithm

Abstract While mitotic rates as well as proliferation indices have been reported as significantly different among normal breast and different grades of breast cancer, a reproducible cut-off value has not emerged. The goal of this study was to determine whether more reproducible algorithms using whole slide digital images (WSDI) can define better cut-off values for the normal breast vs. DCIS. Design: Whole tissue sections from 10 normal breast reductions and 10 cases of DCIS were evaluated by immunohistochemistry for caPCNA and Ki67 (Mib-1). Having identified an isoform of PCNA that was uniquely expressed by cancer cells (caPCNA) and distinct from the isoform expressed by non-malignant and normal cells (nmPCNA), antibody selectively binding to caPCNA was developed and used for these studies. Stained slides were scanned using the Aperio (Vista, California) automated whole slide scanning system (Scanscope CS) and were viewed using ImageScopeTM. Single fields of view from each WSDI measuring ∼420,000 µ2 and representing the highest density of caPCNA and Ki-67 positive nuclei were selected for analysis. Different areas were analyzed in a subset of slides to determine the variability in each case, and the total nuclear labeling index was generated using the Aperio positive pixel algorithm. Results: We analyzed the labeling indices in 10 normal breast reductions and 10 DCIS cases. Both caPCNA and Ki-67 were nuclear specific with minimal cytoplasmic staining. The mean labeling indices for the Aperio automated positive pixel counts were caPCNA (0.15) and Ki-67 (0.12) for normal breast and caPCNA (0.52) and Ki-67 (0.29) for DCIS. The correlation coefficients between automated counts were 0.653 for normal breast and 0.616 for DCIS, with an overall correlation coefficient of 0.708 for the entire series. Conclusions: Our analysis of both biomarkers yielded a similarly increasing labeling index for DCIS cases when compared to their expression in normal breast expression. The positive pixel algorithm method has a sampling bias due to subjective selection of the area being counted. In summary, the present results show similar findings for caPCNA and Ki-67 in normal breast reductions with no evidence of breast disease and in DCIS clinical cases. Comparing caPCNA with Ki-67, and unlike traditional PCNA, caPCNA can potentially better differentiate between normal breast and DCIS, and may be a useful biomarker for early stage cancer diagnosis. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3185. doi:10.1158/1538-7445.AM2011-3185

Development of a cell-based high throughput assay system and IN-house image analysis software for screening of active compounds against dengue virus

Dengue virus (DV) is a mosquito-borne pathogen capable of infecting multiple target organs in the human host. Infection with any one of the four dengue serotypes can cause an acute febrile illness known as Dengue Fever (DF), while subsequent infection with a heterologous serotype or a highly virulent strain can result in the more severe and sometimes lethal Dengue Hemorrhagic Fever (DHF) and Dengue Shock Syndrome (DSS). According to the World Health Organization, approximately 2.5 billion people live in dengue endemic countries, with an estimated 50 million cases occurring annually. The lack of an effective dengue vaccine has prompted the need to discover compounds that can inhibit DV infection. In this study, a cell-based, high-throughput assay system that utilizes an image-based analysis algorithm for evaluating active compounds that target DV infection is described. Using representative strains of DV1, DV2, DV3 serotypes, a human-derived hepatoma cell line (Huh7.5) and Aedes albopictus mosquito-derived cell line (clone C6/36) were inoculated with a high titer of each virus in a 384-well plate culture system in the presence of varying concentration of known active compounds. After 1-2 rounds of viral replication (48~72 h), infection was arrested by para-formaldehyde fixation. DV-infected cells were visualized by probing with D1-4G2-4-15 mAb, a flavivirus group-specific monoclonal antibody that targets the E protein, and a mouse IgG-specific AlexaFluor488™ secondary antibody. Images of the DV-infected cell culture are captured with an automated confocal microscope (Evotec Opera™), and analyzed using IM v3.0, a custom-based image analysis software developed by IP-Korea’s Image Mining Group. A 10-point dose-response curve was generated for each active compound and reproduced in several experiments. This newly developed HTA (High Throughput Assay) system for Dengue can be a useful tool to screen large compound libraries for active drugs that have inhibitory effects to DV infection.

Genetic algorithms for feature selection of image analysis-based quality monitoring model: An application to an iron mine

Measuring the quality parameters of materials at mines is difficult and a costly job. In this paper, an image analysis-based method is proposed efficiently and cost effectively that determines the quality parameters of material. The image features are extracted from the samples collected from a mine and modeled using neural networks against the actual grade values of the samples generated by chemical analysis. The dimensions of the image features are reduced by applying the genetic algorithm. The results showed that only 39 features out of 189 features are sufficient to model the quality parameter. The model was tested with the testing data set and the result revealed that the estimated grade values are in good agreement with the real grade values ( R 2=0.77). The developed method was then applied to a case study mine of iron ore. The case study results show that proposed image-based algorithm can be a good alternative for estimating quality parameters of materials at a mine site. The effectiveness of the proposed method was verified by applying it on a limestone deposit and the results revealed that the method performed equally well for the limestone deposit.

Comprehensive Model and Image-Based Recognition of Hand Gestures for Interaction in 3D Environments

Interest in gesture-based interaction has been growing considerably, but most systems still limit the recognition of hand gestures to a small set of signs. We present a model for hand gestures that allows the definition of thousands of distinct signals based on the combination of a much smaller number of gesture components. This model comprehends several different kinds of gestures, both static and dynamic and using either hand or both. The choice of types of gestures and individual components is based not only on a review of the relevant literature but also on preliminary user studies, specifically for interaction in virtual and augmented environments and in entertainment and education applications. Gesture recognition based on this model is implemented as a finite state machine that incorporates the results of algorithms for the classification of each component, but is itself independent of those algorithms. The paper also describes an unencumbered gesture recognition system built using this model and recognition strategy, a single low-cost camera and relatively simple image-based algorithms to classify hand poses, movements and location and for segmentation. Tests show the model allowed the definition of the desired gestures for three target applications, a commercial computer game and two educational 3D environments. Our system was able to recognize these user gestures, and several others, in real-time. We could also perceive a need for the recognition of incomplete gestures and for a more robust segmentation strategy.

Development and Implementation of Image-based Algorithms for Measurement of Deformations in Material Testing

This paper presents the development and implementation of three image-based methods used to detect and measure the displacements of a vast number of points in the case of laboratory testing on construction materials. Starting from the needs of structural engineers, three ad hoc tools for crack measurement in fibre-reinforced specimens and 2D or 3D deformation analysis through digital images were implemented and tested. These tools make use of advanced image processing algorithms and can integrate or even substitute some traditional sensors employed today in most laboratories. In addition, the automation provided by the implemented software, the limited cost of the instruments and the possibility to operate with an indefinite number of points offer new and more extensive analysis in the field of material testing. Several comparisons with other traditional sensors widely adopted inside most laboratories were carried out in order to demonstrate the accuracy of the implemented software. Implementation details, simulations and real applications are reported and discussed in this paper.

Spatial Prediction of Weed Intensities From Exact Count Data and Image-Based Estimates

Summary Collecting weed exact counts in an agricultural field is easy but extremely time consuming. Image analysis algorithms for object extraction applied to pictures of agricultural fields may be used to estimate the weed content with a high resolution (about 1 m2), and pictures that are acquired at a large number of sites can be used to obtain maps of weed content over a whole field at a reasonably low cost. However, these image-based estimates are not perfect and acquiring exact weed counts also is highly useful both for assessing the accuracy of the image-based algorithms and for improving the estimates by use of the combined data. We propose and compare various models for image index and exact weed count and we use them to assess how such data should be combined to obtain reliable maps. The method is applied to a real data set from a 30-ha field. We show that using image estimates in addition to exact counts allows us to improve the accuracy of maps significantly. We also show that the relative performances of the methods depend on the size of the data set and on the specific methodology (full Bayes versus plug-in) that is implemented.

Crack measurement: Development, testing and applications of an automatic image-based algorithm

The paper presents an Image-based Method for Crack Analysis (IMCA) which is capable of processing a sequence of digital imagery to perform a twofold task: (i) the extraction of crack borders and the evaluation of its width across the longitudinal profile; (ii) the measurement of crack deformations (width, sliding and rotation). Here both problems are solved in 2-D, but an extension to 3-D is also addressed. The equipment needed to apply the method is made up of a digital camera (or a still video-camera in case a high frequency in data acquisition is necessary), an orientation frame which establishes the object reference system, a pair of signalized supports to be placed in a permanent way on both sides of the crack to compute deformations; however, permanent targets are mandatory only for case (ii). The measurement process is carried out in a fully automatic way, a fact also that makes this technique highly operational for unskilled people in engineering surveying or photogrammetry. The accuracy of the proposed method, evaluated in experimental tests adopting different consumer digital cameras, is about ± 5–20 μm, like the accuracy of most deformometers, but with the advantage of automation and of augmented achievable information; moreover, the image sequence can be archived and off-line measurements could be performed at any time.

Aerosol optical thickness (AOT) retrieval over land using satellite image-based algorithm

The aim of this study is to present and apply the proposed algorithm on archived time series Landsat TM over an urban area in the vicinity of Heathrow Airport (UK) acquired in 1984-1986 and to two up-to-dated Landsat TM images in the vicinity of Paphos Airport in Cyprus acquired in July-August 2008. The monitoring of aerosol concen- trations becomes a high environmental priority particularly in urban areas. The proposed algorithm has been developed to allow the quantification of the aerosol optical thickness (AOT) over land. The algorithm compares multitemporal satellite data sets and evaluates radiometric alterations due to the optical atmospheric effects of aerosols. Novel features of this algorithm which is based on the application of radiative transfer calculations are the inclusion of applying iteration procedures for selecting the suitable object for determining the aerosol optical thickness and the automatic division into working grid cells. AOT retrieval from Landsat TM band 1 images of Cyprus has been cross- validated satisfactorily by comparing AOT values found from handheld MICROTOPS II sun photometer.

Simulating Human Visual Perception in Nighttime Illumination

This paper presents an image-based algorithm for simulating the visual adaptation of the human visual system to various illuminations, especially in dark nighttime conditions. The human visual system exhibits different characteristics depending on the illumination intensity, with photopic vision in bright conditions, scotopic vision in dark conditions, and mesopic vision between these two. A computational model is designed to simulate multiple features of mesopic vision and scotopic vision, including the chromaticity change, luminance change, and visual acuity loss. The system uses a source image under bright illumination as input. Then assuming that the viewer has already adapted to the new conditions, the color spectrum of the input image is reconstructed to replace the source with modifications of the chromaticity and the luminance of the relighted scene. A bilateral filter is used to simulate the visual acuity loss. The model parameters have clear physical meanings and can be obtained from experimental data to achieve realistic results. The algorithm can be used not only for visual perception simulation, but also as a day-for-night tool to produce realistic nighttime images from daytime images.

Practice of Web Data Mining Methods Application

Practice of Web Data Mining Methods ApplicationRecent growth of information on the Internet imposes high demands on the effectiveness of processing algorithms. This paper discusses some algorithms from the field of Web Data Mining which have proved effective in many existing applications. The paper is divided into two logical parts; the first part provides a theoretical description of the algorithms, but the second one contains examples of their successful use to solve real problems. Search algorithms of vague duplicates of documents are currently actively used by all the leading search engines in the world. The paper describes the following algorithms: shingles, signature methods and image-based algorithms. Such methods of classification as a method of fuzzy clustering to-medium (Fuzzy cmeans/ FCM clustering) and clustering by ant colony (Standard Ant Clustering Algorithm SACA) are considered. In conclusion, the experience of the successful application of fuzzy clustering in conjunction with the software toolkit DataEngine to improve the efficiency of the bank "BCI Bank" is described as well as the sharing of the ant colony clustering method in conjunction with linear genetic programming to meet the increasing efficiency of predicting the load on the servers of high load Internet portal Monash Institut.