Open source bioimage informatics for cell biology

Synaptonemal & CO analyzer: A tool for synaptonemal complex and crossover analysis in immunofluorescence images

During the formation of ova and sperm, homologous chromosomes get physically attached through the synaptonemal complex and exchange DNA at crossover sites by a process known as meiotic recombination. Chromosomes that do not recombine or have anomalous crossover distributions often separate poorly during the subsequent cell division and end up in abnormal numbers in ova or sperm, which can lead to miscarriage or developmental defects. Crossover numbers and distribution along the synaptonemal complex can be visualized by immunofluorescent microscopy. However, manual analysis of large numbers of cells is very time-consuming and a major bottleneck for recombination studies. Some image analysis tools have been created to overcome this situation, but they are not readily available, do not provide synaptonemal complex data, or do not tackle common experimental difficulties, such as overlapping chromosomes. To overcome these limitations, we have created and validated an open-source ImageJ macro routine that facilitates and speeds up the crossover and synaptonemal complex analyses in mouse chromosome spreads, as well as in other vertebrate species. It is free, easy to use and fulfills the recommendations for enhancing rigor and reproducibility in biomedical studies.

Evaluation of ImageJ software in ultrasonic image analysis: Follicular and luteal morphological characteristics of cattle

This study was conducted to compare the effectiveness of two methodologies in evaluating B- and Doppler-mode ultrasonic images: analysis using ultrasonic software and utilizing a computer with ImageJ software. To determine if ImageJ software utilization is an efficacious alternative to ultrasonic software device- analysis, there were comparisons of values when using the two methods for morphological and vascular characteristics of follicular dynamics and luteal function in 18 crossbred cattle. From day 8 of an ovarian dynamics synchronization treatment regimen period until the time of ovulation (Day 10), B-mode and power-flow ultrasonic cineloops were obtained every 12 h to assess follicular diameter, wall area, and wall blood perfusion area. On Day 14 after ovulation, US cineloops of ovaries were obtained in B mode and power flow to evaluate various morphological and vascular characteristics of the corpus luteum (CL), including luteal diameter, luteal area, and CL blood perfusion area. Cineloops were evaluated and analyzed using ultrasonic software, and in a computer with ImageJ software. To evaluate consistency in results between the two methods, there was evaluation utilizing paired t-test, Pearson correlation coefficient, Bland-Altman plot, and Linear Regression Test procedures to calculate proportion of bias between values for measurements of variables evaluated. Results indicated none of the values for variables before and after ovulation differed (P > 0.05). It, therefore, was concluded that utilization of ImageJ software is an efficacious biomedical technique to analyze ultrasonic images of morphological and vascular characteristics before and after ovulation in cattle.

THE USAGE OF RADIOFREQUENCY ABLATION FOR TREATMENT OF KELOIDS AND HYPERTROPHIC SCARS

OBJECTIVE

To evaluate the effectiveness of radiofrequency ablation (RFA) for scars tissue volume reduction, and influence on the clinical manifestations of keloid and hypertrophic scars.

MATERIALS AND METHODS

Seventeen patients (9 males, 8 females), 19-62 years old, with some scars were enrolled into the prospective randomized controlled study: 10 with keloids, 7 with hypertrophic scars. Previous failed attempts to correct the scars were undertaken in 8 patients with keloids and in 3 patients with hypertrophic scars. The efficacy of four scar volume reduction methods was compared (12 scars in each group, 5 sessions at 10-day inter- vals): the first group - RFA, the second - intralesional 5-fluorouracil (5-FU) injections, the third - RFA followed by 5-FU injections and the fourth - RFA, injections of verapamil and then 5-FU.

RESULTS

The scars volume reduction was faster after RFA (by 65.3 %) than after chemotherapy. Local verapamil application after RFA followed by 5-FU injections reduced scar volume after the fifth session by 78.3 %. Relieving of the associated symptoms and scars hyperaemia was faster after RFA than after cytostatic drug administration. Their simultaneous application, especially in combination with verapamil, intensified the effect. There were no infectious complications and haemorrhages. Ulcers 3-5 mm in diameter in the place of puncture on one scar in the first group, on two scars in the third and fourth groups were registered as the side effects. Seventeen scars in seven patients, who received RFA as a monotherapy or in combination with chemotherapy, were assessed 6 months after treatment. The average scars volume decreased from (3260.5 ± 2057.36) mm3 at the moment after the fifth session to (2110.6 ± 1296.16) mm3, p = 0.0033.

CONCLUSIONS

Scars volume reduction was faster after five sessions of RFA than following local 5-FU injections. Combination of RFA with scars infiltration with verapamil and 5-FU strengthened the effect. In the same way, the above-mentioned methods influenced clinical symptomatology associated with the scars. Scars hyperaemia was also reduced by RFA. Radiofrequency ablation of excess scar tissue is advisable to use as an initial method for the treat- ment of keloids and hypertrophic scars, since it is easy to carry, is not expensive and effectively reduces the scar vol- ume, demonstrating the patient the opportunity to achieve a positive result.

Implementation of novel statistical procedures and other advanced approaches to improve analysis of CASA data

Computer-aided sperm analysis (CASA) produces a wealth of data that is frequently ignored. The use of multiparametric statistical methods can help explore these datasets, unveiling the subpopulation structure of sperm samples. In this review we analyse the significance of the internal heterogeneity of sperm samples and its relevance. We also provide a brief description of the statistical tools used for extracting sperm subpopulations from the datasets, namely unsupervised clustering (with non-hierarchical, hierarchical and two-step methods) and the most advanced supervised methods, based on machine learning. The former method has allowed exploration of subpopulation patterns in many species, whereas the latter offering further possibilities, especially considering functional studies and the practical use of subpopulation analysis. We also consider novel approaches, such as the use of geometric morphometrics or imaging flow cytometry. Finally, although the data provided by CASA systems provides valuable information on sperm samples by applying clustering analyses, there are several caveats. Protocols for capturing and analysing motility or morphometry should be standardised and adapted to each experiment, and the algorithms should be open in order to allow comparison of results between laboratories. Moreover, we must be aware of new technology that could change the paradigm for studying sperm motility and morphology.

Evaluation of the scars' vascularization using computer processing of the digital images

BACKGROUND

The application of new techniques of the scars' correction requires the objective evaluation of their vascularization.

OBJECTIVE

To evaluate the effectiveness of digital program ImageJ in assessing neovascularization of pathologic scars.

MATERIAL AND METHODS

In this cross-sectional study, a total of 25 patients with pathologic scars were enrolled. Vessel selection in the first set of digital images of their scars was performed by computer processing started from thresholding with subsequent manual correction. In the second set of the same pictures, Vessel Analysis plugin was used. Comparison of both approaches was performed by three independent investigators. The time required for images processing was measured.

RESULTS

The average time that image processing and calculation have taken in the first group (753.3 ± 88.02 seconds) was statistically longer (P < 0.0001) than in the second one (358.1 ± 105.91 seconds). Independent investigators scored the precision of vessel selection in the first group as 80.4 ± 9.82, in the second group as 72.6 ± 10.53 (P < 0.0001). Kolmogorov-Smirnov test demonstrated higher precision of vessel selection by method that involves manual correction (P < 0.001). The results of Vascular Density measurements were obviously overestimated in the second group. More expedient looks calculation of the Vascular Length Density: ratio of skeletonized vasculature area to total area. Skeletonization avoids overestimation of Vascular Density, but the density of the vessel mesh can be determined.

CONCLUSIONS

Computer processing of the scars' digital photographs using ImageJ software gives cheap, technically easy and not cumbersome way of superficial scars' vascularization objectifying. Vessel selection with subsequent manual correction has advantage of higher precision.

Strategies and Solutions to Maintain and Retain Data from High Content Imaging, Analysis, and Screening Assays

Data analysis and management in high content screening (HCS) has progressed significantly in the past 10 years. The analysis of the large volume of data generated in HCS experiments represents a significant challenge and is currently a bottleneck in many screening projects. In most screening laboratories, HCS has become a standard technology applied routinely to various applications from target identification to hit identification to lead optimization. An HCS data management and analysis infrastructure shared by several research groups can allow efficient use of existing IT resources and ensures company-wide standards for data quality and result generation. This chapter outlines typical HCS workflows and presents IT infrastructure requirements for multi-well plate-based HCS.

ImageJ2: ImageJ for the next generation of scientific image data

BACKGROUND

ImageJ is an image analysis program extensively used in the biological sciences and beyond. Due to its ease of use, recordable macro language, and extensible plug-in architecture, ImageJ enjoys contributions from non-programmers, amateur programmers, and professional developers alike. Enabling such a diversity of contributors has resulted in a large community that spans the biological and physical sciences. However, a rapidly growing user base, diverging plugin suites, and technical limitations have revealed a clear need for a concerted software engineering effort to support emerging imaging paradigms, to ensure the software's ability to handle the requirements of modern science.

RESULTS

We rewrote the entire ImageJ codebase, engineering a redesigned plugin mechanism intended to facilitate extensibility at every level, with the goal of creating a more powerful tool that continues to serve the existing community while addressing a wider range of scientific requirements. This next-generation ImageJ, called "ImageJ2" in places where the distinction matters, provides a host of new functionality. It separates concerns, fully decoupling the data model from the user interface. It emphasizes integration with external applications to maximize interoperability. Its robust new plugin framework allows everything from image formats, to scripting languages, to visualization to be extended by the community. The redesigned data model supports arbitrarily large, N-dimensional datasets, which are increasingly common in modern image acquisition. Despite the scope of these changes, backwards compatibility is maintained such that this new functionality can be seamlessly integrated with the classic ImageJ interface, allowing users and developers to migrate to these new methods at their own pace.

CONCLUSIONS

Scientific imaging benefits from open-source programs that advance new method development and deployment to a diverse audience. ImageJ has continuously evolved with this idea in mind; however, new and emerging scientific requirements have posed corresponding challenges for ImageJ's development. The described improvements provide a framework engineered for flexibility, intended to support these requirements as well as accommodate future needs. Future efforts will focus on implementing new algorithms in this framework and expanding collaborations with other popular scientific software suites.

ImageJ2: ImageJ for the next generation of scientific image data

BACKGROUND

ImageJ is an image analysis program extensively used in the biological sciences and beyond. Due to its ease of use, recordable macro language, and extensible plug-in architecture, ImageJ enjoys contributions from non-programmers, amateur programmers, and professional developers alike. Enabling such a diversity of contributors has resulted in a large community that spans the biological and physical sciences. However, a rapidly growing user base, diverging plugin suites, and technical limitations have revealed a clear need for a concerted software engineering effort to support emerging imaging paradigms, to ensure the software's ability to handle the requirements of modern science.

RESULTS

We rewrote the entire ImageJ codebase, engineering a redesigned plugin mechanism intended to facilitate extensibility at every level, with the goal of creating a more powerful tool that continues to serve the existing community while addressing a wider range of scientific requirements. This next-generation ImageJ, called "ImageJ2" in places where the distinction matters, provides a host of new functionality. It separates concerns, fully decoupling the data model from the user interface. It emphasizes integration with external applications to maximize interoperability. Its robust new plugin framework allows everything from image formats, to scripting languages, to visualization to be extended by the community. The redesigned data model supports arbitrarily large, N-dimensional datasets, which are increasingly common in modern image acquisition. Despite the scope of these changes, backwards compatibility is maintained such that this new functionality can be seamlessly integrated with the classic ImageJ interface, allowing users and developers to migrate to these new methods at their own pace.

CONCLUSIONS

Scientific imaging benefits from open-source programs that advance new method development and deployment to a diverse audience. ImageJ has continuously evolved with this idea in mind; however, new and emerging scientific requirements have posed corresponding challenges for ImageJ's development. The described improvements provide a framework engineered for flexibility, intended to support these requirements as well as accommodate future needs. Future efforts will focus on implementing new algorithms in this framework and expanding collaborations with other popular scientific software suites.

Workflows for microscopy image analysis and cellular phenotyping

In large scale biological experiments, like high-throughput or high-content cellular screening, the amount and the complexity of images to be analyzed are steadily increasing. To handle and process these images, well defined image processing and analysis steps need to be performed by applying dedicated workflows. Multiple software tools have emerged with the aim to facilitate creation of such workflows by integrating existing methods, tools, and routines, and by adapting them to different applications and questions, as well as making them reusable and interchangeable. In this review, we describe workflow systems for the integration of microscopy image analysis techniques with focus on KNIME and Galaxy.

From morphology to biochemical state - intravital multiphoton fluorescence lifetime imaging of inflamed human skin

The application of multiphoton microscopy in the field of biomedical research and advanced diagnostics promises unique insights into the pathophysiology of inflammatory skin diseases. In the present study, we combined multiphoton-based intravital tomography (MPT) and fluorescence lifetime imaging (MPT-FLIM) within the scope of a clinical trial of atopic dermatitis with the aim of providing personalised data on the aetiopathology of inflammation in a non-invasive manner at patients' bedsides. These 'optical biopsies' generated via MPT were morphologically analysed and aligned with classical skin histology. Because of its subcellular resolution, MPT provided evidence of a redistribution of mitochondria in keratinocytes, indicating an altered cellular metabolism. Two independent morphometric algorithms reliably showed an even distribution in healthy skin and a perinuclear accumulation in inflamed skin. Moreover, using MPT-FLIM, detection of the onset and progression of inflammatory processes could be achieved. In conclusion, the change in the distribution of mitochondria upon inflammation and the verification of an altered cellular metabolism facilitate a better understanding of inflammatory skin diseases and may permit early diagnosis and therapy.

It's all about the pentiums: The use, manipulation, and storage of digital microscopy imaging data for the biological sciences

Digital microscopy has revolutionized quantitative imaging, with binary-encoded computer files serving to capture and store imaging data sets for analysis. With the ever-present use of computers to generate and store imaging data, it becomes increasingly important to understand how these files are created, and how they can be both used and mis-used. This is a particularly important task for the biologist who may have limited background in computer science. Here we discuss some of the basic aspects of digital data storage and use, including file types, storage media, and the choice between commercial and open-source software. Often, open-source software is written by a user or group of users, and then distributed to the scientific community at large. These can be important tools, but there are some hidden costs to this freeware that we will discuss. We will also compare open-source software to commercial imaging software, which is often written for use by non-computer scientists.

Machine learning in cell biology - teaching computers to recognize phenotypes

Recent advances in microscope automation provide new opportunities for high-throughput cell biology, such as image-based screening. High-complex image analysis tasks often make the implementation of static and predefined processing rules a cumbersome effort. Machine-learning methods, instead, seek to use intrinsic data structure, as well as the expert annotations of biologists to infer models that can be used to solve versatile data analysis tasks. Here, we explain how machine-learning methods work and what needs to be considered for their successful application in cell biology. We outline how microscopy images can be converted into a data representation suitable for machine learning, and then introduce various state-of-the-art machine-learning algorithms, highlighting recent applications in image-based screening. Our Commentary aims to provide the biologist with a guide to the application of machine learning to microscopy assays and we therefore include extensive discussion on how to optimize experimental workflow as well as the data analysis pipeline.

Interpretation of medical imaging data with a mobile application: a mobile digital imaging processing environment

Digital Imaging Processing (DIP) requires data extraction and output from a visualization tool to be consistent. Data handling and transmission between the server and a user is a systematic process in service interpretation. The use of integrated medical services for management and viewing of imaging data in combination with a mobile visualization tool can be greatly facilitated by data analysis and interpretation. This paper presents an integrated mobile application and DIP service, called M-DIP. The objective of the system is to (1) automate the direct data tiling, conversion, pre-tiling of brain images from Medical Imaging NetCDF (MINC), Neuroimaging Informatics Technology Initiative (NIFTI) to RAW formats; (2) speed up querying of imaging measurement; and (3) display high-level of images with three dimensions in real world coordinates. In addition, M-DIP provides the ability to work on a mobile or tablet device without any software installation using web-based protocols. M-DIP implements three levels of architecture with a relational middle-layer database, a stand-alone DIP server, and a mobile application logic middle level realizing user interpretation for direct querying and communication. This imaging software has the ability to display biological imaging data at multiple zoom levels and to increase its quality to meet users’ expectations. Interpretation of bioimaging data is facilitated by an interface analogous to online mapping services using real world coordinate browsing. This allows mobile devices to display multiple datasets simultaneously from a remote site. M-DIP can be used as a measurement repository that can be accessed by any network environment, such as a portable mobile or tablet device. In addition, this system and combination with mobile applications are establishing a virtualization tool in the neuroinformatics field to speed interpretation services.

Light microscopy applications in systems biology: opportunities and challenges

Biological systems present multiple scales of complexity, ranging from molecules to entire populations. Light microscopy is one of the least invasive techniques used to access information from various biological scales in living cells. The combination of molecular biology and imaging provides a bottom-up tool for direct insight into how molecular processes work on a cellular scale. However, imaging can also be used as a top-down approach to study the behavior of a system without detailed prior knowledge about its underlying molecular mechanisms. In this review, we highlight the recent developments on microscopy-based systems analyses and discuss the complementary opportunities and different challenges with high-content screening and high-throughput imaging. Furthermore, we provide a comprehensive overview of the available platforms that can be used for image analysis, which enable community-driven efforts in the development of image-based systems biology.

Approaches to automatic parameter fitting in a microscopy image segmentation pipeline: An exploratory parameter space analysis

Introduction:

Research and diagnosis in medicine and biology often require the assessment of a large amount of microscopy image data. Although on the one hand, digital pathology and new bioimaging technologies find their way into clinical practice and pharmaceutical research, some general methodological issues in automated image analysis are still open.

Methods:

In this study, we address the problem of fitting the parameters in a microscopy image segmentation pipeline. We propose to fit the parameters of the pipeline's modules with optimization algorithms, such as, genetic algorithms or coordinate descents, and show how visual exploration of the parameter space can help to identify sub-optimal parameter settings that need to be avoided.

Results:

This is of significant help in the design of our automatic parameter fitting framework, which enables us to tune the pipeline for large sets of micrographs.

Conclusion:

The underlying parameter spaces pose a challenge for manual as well as automated parameter optimization, as the parameter spaces can show several local performance maxima. Hence, optimization strategies that are not able to jump out of local performance maxima, like the hill climbing algorithm, often result in a local maximum.

Fiji: an open-source platform for biological-image analysis

Fiji is a distribution of the popular open-source software ImageJ focused on biological-image analysis. Fiji uses modern software engineering practices to combine powerful software libraries with a broad range of scripting languages to enable rapid prototyping of image-processing algorithms. Fiji facilitates the transformation of new algorithms into ImageJ plugins that can be shared with end users through an integrated update system. We propose Fiji as a platform for productive collaboration between computer science and biology research communities.

Fiji: an open-source platform for biological-image analysis

Fiji is a distribution of the popular open-source software ImageJ focused on biological-image analysis. Fiji uses modern software engineering practices to combine powerful software libraries with a broad range of scripting languages to enable rapid prototyping of image-processing algorithms. Fiji facilitates the transformation of new algorithms into ImageJ plugins that can be shared with end users through an integrated update system. We propose Fiji as a platform for productive collaboration between computer science and biology research communities.

Fiji: an open-source platform for biological-image analysis

Fiji is a distribution of the popular open-source software ImageJ focused on biological-image analysis. Fiji uses modern software engineering practices to combine powerful software libraries with a broad range of scripting languages to enable rapid prototyping of image-processing algorithms. Fiji facilitates the transformation of new algorithms into ImageJ plugins that can be shared with end users through an integrated update system. We propose Fiji as a platform for productive collaboration between computer science and biology research communities.

Open source tools for fluorescent imaging

As microscopy becomes increasingly automated and imaging expands in the spatial and time dimensions, quantitative analysis tools for fluorescent imaging are becoming critical to remove both bottlenecks in throughput as well as fully extract and exploit the information contained in the imaging. In recent years there has been a flurry of activity in the development of bio-image analysis tools and methods with the result that there are now many high-quality, well-documented, and well-supported open source bio-image analysis projects with large user bases that cover essentially every aspect from image capture to publication. These open source solutions are now providing a viable alternative to commercial solutions. More importantly, they are forming an interoperable and interconnected network of tools that allow data and analysis methods to be shared between many of the major projects. Just as researchers build on, transmit, and verify knowledge through publication, open source analysis methods and software are creating a foundation that can be built upon, transmitted, and verified. Here we describe many of the major projects, their capabilities, and features. We also give an overview of the current state of open source software for fluorescent microscopy analysis and the many reasons to use and develop open source methods.

OMERO: flexible, model-driven data management for experimental biology

Data-intensive research depends on tools that manage multidimensional, heterogeneous datasets. We built OME Remote Objects (OMERO), a software platform that enables access to and use of a wide range of biological data. OMERO uses a server-based middleware application to provide a unified interface for images, matrices and tables. OMERO's design and flexibility have enabled its use for light-microscopy, high-content-screening, electron-microscopy and even non-image-genotype data. OMERO is open-source software, available at http://openmicroscopy.org/.

RAMTaB: robust alignment of multi-tag bioimages

Background

In recent years, new microscopic imaging techniques have evolved to allow us to visualize several different proteins (or other biomolecules) in a visual field. Analysis of protein co-localization becomes viable because molecules can interact only when they are located close to each other. We present a novel approach to align images in a multi-tag fluorescence image stack. The proposed approach is applicable to multi-tag bioimaging systems which (a) acquire fluorescence images by sequential staining and (b) simultaneously capture a phase contrast image corresponding to each of the fluorescence images. To the best of our knowledge, there is no existing method in the literature, which addresses simultaneous registration of multi-tag bioimages and selection of the reference image in order to maximize the overall overlap between the images.

Methodology/Principal Findings

We employ a block-based method for registration, which yields a confidence measure to indicate the accuracy of our registration results. We derive a shift metric in order to select the Reference Image with Maximal Overlap (RIMO), in turn minimizing the total amount of non-overlapping signal for a given number of tags. Experimental results show that the Robust Alignment of Multi-Tag Bioimages (RAMTaB) framework is robust to variations in contrast and illumination, yields sub-pixel accuracy, and successfully selects the reference image resulting in maximum overlap. The registration results are also shown to significantly improve any follow-up protein co-localization studies.

Conclusions

For the discovery of protein complexes and of functional protein networks within a cell, alignment of the tag images in a multi-tag fluorescence image stack is a key pre-processing step. The proposed framework is shown to produce accurate alignment results on both real and synthetic data. Our future work will use the aligned multi-channel fluorescence image data for normal and diseased tissue specimens to analyze molecular co-expression patterns and functional protein networks.

Data ontology and an information system realization for web-based management of image measurements

Image acquisition, processing, and quantification of objects (morphometry) require the integration of data inputs and outputs originating from heterogeneous sources. Management of the data exchange along this workflow in a systematic manner poses several challenges, notably the description of the heterogeneous meta-data and the interoperability between the software used. The use of integrated software solutions for morphometry and management of imaging data in combination with ontologies can reduce meta-data loss and greatly facilitate subsequent data analysis. This paper presents an integrated information system, called LabIS. The system has the objectives to automate (i) the process of storage, annotation, and querying of image measurements and (ii) to provide means for data sharing with third party applications consuming measurement data using open standard communication protocols. LabIS implements 3-tier architecture with a relational database back-end and an application logic middle tier realizing web-based user interface for reporting and annotation and a web-service communication layer. The image processing and morphometry functionality is backed by interoperability with ImageJ, a public domain image processing software, via integrated clients. Instrumental for the latter feat was the construction of a data ontology representing the common measurement data model. LabIS supports user profiling and can store arbitrary types of measurements, regions of interest, calibrations, and ImageJ settings. Interpretation of the stored measurements is facilitated by atlas mapping and ontology-based markup. The system can be used as an experimental workflow management tool allowing for description and reporting of the performed experiments. LabIS can be also used as a measurements repository that can be transparently accessed by computational environments, such as Matlab. Finally, the system can be used as a data sharing tool.

TLM-Converter: reorganization of long time-lapse microscopy datasets for downstream image analysis

Automated microscopy enables in vivo studies in developmental biology over long periods of time. Time-lapse recordings in three or more dimensions to study the dynamics of developmental processes can produce huge data sets that extend into the terabyte range. However, depending on the available computational resources and software design, downstream processing of very large image data sets can become highly inefficient, if not impossible. To address the lack of available open source and commercial software tools to efficiently reorganize time-lapse data on a desktop computer with limited system resources, we developed TLM-Converter. The software either fragments oversized files or concatenates multiple files representing single time frames and saves the output files in open standard formats. Our application is undemanding on system resources as it does not require the whole data set to be loaded into the system memory. We tested our tool on time-lapse data sets of live Drosophila specimens recorded by laser scanning confocal microscopy. Image data reorganization dramatically enhances the productivity of time-lapse data processing and allows the use of downstream image analysis software that is unable to handle large data sets of ≥2 GB. In addition, saving the outputs in open standard image file formats enables data sharing between independently developed software tools.

Visual pattern mining in histology image collections using bag of features

OBJECTIVE

The paper addresses the problem of finding visual patterns in histology image collections. In particular, it proposes a method for correlating basic visual patterns with high-level concepts combining an appropriate image collection representation with state-of-the-art machine learning techniques.

METHODOLOGY

The proposed method starts by representing the visual content of the collection using a bag-of-features strategy. Then, two main visual mining tasks are performed: finding associations between visual-patterns and high-level concepts, and performing automatic image annotation. Associations are found using minimum-redundancy-maximum-relevance feature selection and co-clustering analysis. Annotation is done by applying a support-vector-machine classifier. Additionally, the proposed method includes an interpretation mechanism that associates concept annotations with corresponding image regions. The method was evaluated in two data sets: one comprising histology images from the different four fundamental tissues, and the other composed of histopathology images used for cancer diagnosis. Different visual-word representations and codebook sizes were tested. The performance in both concept association and image annotation tasks was qualitatively and quantitatively evaluated.

RESULTS

The results show that the method is able to find highly discriminative visual features and to associate them to high-level concepts. In the annotation task the method showed a competitive performance: an increase of 21% in f-measure with respect to the baseline in the histopathology data set, and an increase of 47% in the histology data set.

CONCLUSIONS

The experimental evidence suggests that the bag-of-features representation is a good alternative to represent visual content in histology images. The proposed method exploits this representation to perform visual pattern mining from a wider perspective where the focus is the image collection as a whole, rather than individual images.

'Sciencenet'--towards a global search and share engine for all scientific knowledge

SUMMARY

Modern biological experiments create vast amounts of data which are geographically distributed. These datasets consist of petabytes of raw data and billions of documents. Yet to the best of our knowledge, a search engine technology that searches and cross-links all different data types in life sciences does not exist. We have developed a prototype distributed scientific search engine technology, 'Sciencenet', which facilitates rapid searching over this large data space. By 'bringing the search engine to the data', we do not require server farms. This platform also allows users to contribute to the search index and publish their large-scale data to support e-Science. Furthermore, a community-driven method guarantees that only scientific content is crawled and presented. Our peer-to-peer approach is sufficiently scalable for the science web without performance or capacity tradeoff.

AVAILABILITY AND IMPLEMENTATION

The free to use search portal web page and the downloadable client are accessible at: http://sciencenet.kit.edu. The web portal for index administration is implemented in ASP.NET, the 'AskMe' experiment publisher is written in Python 2.7, and the backend 'YaCy' search engine is based on Java 1.6.

A hybrid blob-slice model for accurate and efficient detection of fluorescence labeled nuclei in 3D

BACKGROUND

To exploit the flood of data from advances in high throughput imaging of optically sectioned nuclei, image analysis methods need to correctly detect thousands of nuclei, ideally in real time. Variability in nuclear appearance and undersampled volumetric data make this a challenge.

RESULTS

We present a novel 3D nuclear identification method, which subdivides the problem, first segmenting nuclear slices within each 2D image plane, then using a shape model to assemble these slices into 3D nuclei. This hybrid 2D/3D approach allows accurate accounting for nuclear shape but exploits the clear 2D nuclear boundaries that are present in sectional slices to avoid the computational burden of fitting a complex shape model to volume data. When tested over C. elegans, Drosophila, zebrafish and mouse data, our method yielded 0 to 3.7% error, up to six times more accurate as well as being 30 times faster than published performances. We demonstrate our method's potential by reconstructing the morphogenesis of the C. elegans pharynx. This is an important and much studied developmental process that could not previously be followed at this single cell level of detail.

CONCLUSIONS

Because our approach is specialized for the characteristics of optically sectioned nuclear images, it can achieve superior accuracy in significantly less time than other approaches. Both of these characteristics are necessary for practical analysis of overwhelmingly large data sets where processing must be scalable to hundreds of thousands of cells and where the time cost of manual error correction makes it impossible to use data with high error rates. Our approach is fast, accurate, available as open source software and its learned shape model is easy to retrain. As our pharynx development example shows, these characteristics make single cell analysis relatively easy and will enable novel experimental methods utilizing complex data sets.

MIDG-Emerging grid technologies for multi-site preclinical molecular imaging research communities

PURPOSE

Molecular imaging is the visualization and identification of specific molecules in anatomy for insight into metabolic pathways, tissue consistency, and tracing of solute transport mechanisms. This paper presents the Molecular Imaging Data Grid (MIDG) which utilizes emerging grid technologies in preclinical molecular imaging to facilitate data sharing and discovery between preclinical molecular imaging facilities and their collaborating investigator institutions to expedite translational sciences research. Grid-enabled archiving, management, and distribution of animal-model imaging datasets help preclinical investigators to monitor, access and share their imaging data remotely, and promote preclinical imaging facilities to share published imaging datasets as resources for new investigators.

METHODS

The system architecture of the Molecular Imaging Data Grid is described in a four layer diagram. A data model for preclinical molecular imaging datasets is also presented based on imaging modalities currently used in a molecular imaging center. The MIDG system components and connectivity are presented. And finally, the workflow steps for grid-based archiving, management, and retrieval of preclincial molecular imaging data are described.

RESULTS

Initial performance tests of the Molecular Imaging Data Grid system have been conducted at the USC IPILab using dedicated VMware servers. System connectivity, evaluated datasets, and preliminary results are presented. The results show the system's feasibility, limitations, direction of future research.

CONCLUSIONS

Translational and interdisciplinary research in medicine is increasingly interested in cellular and molecular biology activity at the preclinical levels, utilizing molecular imaging methods on animal models. The task of integrated archiving, management, and distribution of these preclinical molecular imaging datasets at preclinical molecular imaging facilities is challenging due to disparate imaging systems and multiple off-site investigators. A Molecular Imaging Data Grid design, implementation, and initial evaluation is presented to demonstrate the secure and novel data grid solution for sharing preclinical molecular imaging data across the wide-area-network (WAN).

Descriptive no more: the dawn of high-throughput microscopy

The next revolution in microscopy is upon us: it is High-Throughput Imaging (HTI). In HTI large numbers of images from many samples are acquired and analyzed. This has become possible due to a confluence of dramatic progress in microscope engineering, enabling efficient image collection, and the availability of high computing power for data analysis. Combining HTI with genome-wide RNA interference (RNAi)-based gene knockdown technology offers a powerful approach for unbiased discovery of cellular mechanisms.