A framework for biomedical figure segmentation towards image-based document retrieval

The figures included in many of the biomedical publications play an important role in understanding the biological experiments and facts described within. Recent studies have shown that it is possible to integrate the information that is extracted from figures in classical document classification and retrieval tasks in order to improve their accuracy. One important observation about the figures included in biomedical publications is that they are often composed of multiple subfigures or panels, each describing different methodologies or results. The use of these multimodal figures is a common practice in bioscience, as experimental results are graphically validated via multiple methodologies or procedures. Thus, for a better use of multimodal figures in document classification or retrieval tasks, as well as for providing the evidence source for derived assertions, it is important to automatically segment multimodal figures into subfigures and panels. This is a challenging task, however, as different panels can contain similar objects (i.e., barcharts and linecharts) with multiple layouts. Also, certain types of biomedical figures are text-heavy (e.g., DNA sequences and protein sequences images) and they differ from traditional images. As a result, classical image segmentation techniques based on low-level image features, such as edges or color, are not directly applicable to robustly partition multimodal figures into single modal panels.In this paper, we describe a robust solution for automatically identifying and segmenting unimodal panels from a multimodal figure. Our framework starts by robustly harvesting figure-caption pairs from biomedical articles. We base our approach on the observation that the document layout can be used to identify encoded figures and figure boundaries within PDF files. Taking into consideration the document layout allows us to correctly extract figures from the PDF document and associate their corresponding caption. We combine pixel-level representations of the extracted images with information gathered from their corresponding captions to estimate the number of panels in the figure. Thus, our approach simultaneously identifies the number of panels and the layout of figures.In order to evaluate the approach described here, we applied our system on documents containing protein-protein interactions (PPIs) and compared the results against a gold standard that was annotated by biologists. Experimental results showed that our automatic figure segmentation approach surpasses pure caption-based and image-based approaches, achieving a 96.64% accuracy. To allow for efficient retrieval of information, as well as to provide the basis for integration into document classification and retrieval systems among other, we further developed a web-based interface that lets users easily retrieve panels containing the terms specified in the user queries.     

Utilizing weakly controlled vocabulary for sentence segmentation in biomedical literature

Since biomedical texts contain a wide variety of domain specific terms, building a large dictionary to perform term matching is of great relevance. However, due to the existence of null boundary between adjacent terms, this matching is not a trivial problem. Moreover, it is known that generative words cannot be comprehensively included in a dictionary because their possible variations are infinite. In this study, we report our approach to dictionary building and term matching in biomedical texts. Large amount of terms with/without part-of-speech (POS) and/or category information were gathered, and a completion program generated approximately 1.36 million term variants to avoid stemming problems when matching terms. The dictionary was stored in a relational database management system (RDBMS) for quick lookup, and used by a matching program. Since the matching operation is not restricted to a substring surrounded by space characters, we can avoid the problem of null boundaries. This feature is also useful for generative words. Experimental results on GENIA corpus are promising: nearly half of the possible terms were correctly recognized as a meaningful segment, and most of the remaining half could be correctly recognized by some post-processing process, like chunking and further decomposition. It should be remarked that although we have not used term cost, connectivity cost, or syntactic information, reasonable segmentation and dictionary lookup were performed in most cases.     

An interactive agent-based framework for materialization-informed architectural design

Concepts of swarm intelligence are becoming increasingly relevant in the field of architectural design. An example is the use of agent-based modeling and simulation methods, which can help manage the complexity of building designs that feature many similar, but geometrically unique elements. Apart from leading to effective solutions and expanding the architectural design space, agent-based design methods can also be employed in integrated planning processes, in which the contributions of various disciplines take place in an integrated loop instead of being executed consecutively. We propose a computational framework for architectural design, in which agents represent building elements and/or joints between building elements. Behavior parameters, behavior weighting, and the environment can be modified in real-time while the agent system is running. Additionally, the designer can interact with individual agents directly, while slowing down or pausing agent movement if so desired. In the resulting design approach, the designer can globally adjust behavior parameters, while retaining local control over details where needed. To facilitate an integrative design process, domain-specific data and the results of external analysis can be included, either directly as input for agent behaviors, or by modifying the environment. We illustrate the potential of this computational framework using the example of the design of plate structures and show how this method can lead to quantifiable results while also attaining aesthetic goals. Furthermore, we provide an outlook toward possible further extensions of agent-based design methods in architecture.     

A framework for automatic heart sound analysis without segmentation

Background  

A new framework for heart sound analysis is proposed. One of the most difficult processes in heart sound analysis is segmentation, due to interference form murmurs.     

BioQuery: an object framework for building queries to biomedical databases

SUMMARY: BioQuery is an application that helps scientists automate database searches. Users can build and store queries to public biomedical databases, and receive periodic updates on the results of those queries when new data is available. The application is implemented on a portable object framework that can provide database-searching capability to other applications. This framework is easily extensible, allowing users to develop plug-ins that provide access to new databases. BioQuery thus provides end-users with a complete database searching interface and updating service, and gives developers a toolkit to provide database-searching capability to their applications. AVAILABILITY: Free to all users: http://www.bioquery.org.     

Toxicity of inorganic nanomaterials in biomedical imaging

Inorganic nanoparticles have shown promising potentials as novel biomedical imaging agents with high sensitivity, high spatial and temporal resolution. To translate the laboratory innovations into clinical applications, their potential toxicities are highly concerned and have to be evaluated comprehensively both in vitro and in vivo before their clinical applications. In this review, we first summarized the in vivo and in vitro toxicities of the representative inorganic nanoparticles used in biomedical imagings. Then we further discuss the origin of nanotoxicity of inorganic nanomaterials, including ROS generation and oxidative stress, chemical instability, chemical composition, the surface modification, dissolution of nanoparticles to release excess free ions of metals, metal redox state, and left-over chemicals from synthesis, etc. We intend to provide the readers a better understanding of the toxicology aspects of inorganic nanomaterials and knowledge for achieving optimized designs of safer inorganic nanomaterials for clinical applications.     

An automatic segmentation and classification framework for anti-nuclear antibody images

Autoimmune disease is a disorder of immune system due to the over-reaction of lymphocytes against one's own body tissues. Anti-Nuclear Antibody (ANA) is an autoantibody produced by the immune system directed against the self body tissues or cells, which plays an important role in the diagnosis of autoimmune diseases. Indirect ImmunoFluorescence (IIF) method with HEp-2 cells provides the major screening method to detect ANA for the diagnosis of autoimmune diseases. Fluorescence patterns at present are usually examined laboriously by experienced physicians through manually inspecting the slides with the help of a microscope, which usually suffers from inter-observer variability that limits its reproducibility. Previous researches only provided simple segmentation methods and criterions for cell segmentation and recognition, but a fully automatic framework for the segmentation and recognition of HEp-2 cells had never been reported before. This study proposes a method based on the watershed algorithm to automatically detect the HEp-2 cells with different patterns. The experimental results show that the segmentation performance of the proposed method is satisfactory when evaluated with percent volume overlap (PVO: 89%). The classification performance using a SVM classifier designed based on the features calculated from the segmented cells achieves an average accuracy of 96.90%, which outperforms other methods presented in previous studies. The proposed method can be used to develop a computer-aided system to assist the physicians in the diagnosis of auto-immune diseases.     

Biocompatible astaxanthin as novel contrast agent for biomedical imaging

Photoacoustic imaging (PAI) is a hybrid imaging modality with high resolution and sensitivity that can be beneficial for cancer staging. Due to insufficient endogenous photoacoustic (PA) contrast, the development of exogenous agents is critical in targeting cancerous tumors. The current study demonstrates the feasibility of marine‐oriented material, astaxanthin, as a biocompatible PA contrast agent. Both silicon tubing phantoms and ex vivo bladder tissues are tested at various concentrations (up to 5 mg/ml) of astaxanthin to quantitatively explore variations in PA responses. A Q‐switched Nd : YAG laser (λ = 532 nm) in conjunction with a 5 MHz ultrasound transducer is employed to generate and acquire PA signals from the samples. The phantom results presented that the PA signal amplitudes increase linearly with the astaxanthin concentrations (threshold detection = 0.31 mg/ml). The tissue injected with astaxanthin yields up to 16‐fold higher PA signals, compared with that with saline. Due to distribution of the injected astaxanthin, PAI can image the margin of astaxanthin boles as well as quantify their volume in 3D reconstruction. Further investigations on selective tumor targeting are required to validate astaxanthin as a potential biocompatible contrast agent for PAI‐assisted bladder cancer detection.

     

Nanoparticles for biomedical imaging: fundamentals of clinical translation

Because of their large size compared to small molecules and their multifunctionality, nanoparticles (NPs) hold promise as biomedical imaging, diagnostic, and theragnostic agents. However, the key to their success hinges on a detailed understanding of their behavior after administration into the body. NP biodistribution, target binding, and clearance are complex functions of their physicochemical properties in serum, which include hydrodynamic diameter, solubility, stability, shape and flexibility, surface charge, composition, and formulation. Moreover, many materials used to construct NPs have real or potential toxicity or may interfere with other medical tests. In this review, we discuss the design considerations that mediate NP behavior in the body and the fundamental principles that govern clinical translation. By analyzing those nanomaterials that have already received regulatory approval, most of which are actually therapeutic agents, we attempt to predict which types of NPs hold potential as diagnostic agents for biomedical imaging. Finally, using quantum dots as an example, we provide a framework for deciding whether an NP-based agent is the best choice for a particular clinical application.     

Very small photoluminescent gold nanoparticles for multimodality biomedical imaging

An original synthesis method based on X-ray irradiation produced gold nanoparticles (AuNPs) with two important properties for biomedical research: intense visible photoluminescence and very high accumulation in cancer cells. The nanoparticles, coated with MUA (11-mercaptoundecanoid acid), are very small (1.4 nm diameter); the above two properties are not present for even slightly larger sizes. The small MUA-AuNPs are non-cytotoxic (except for very high concentrations) and do not interfere with cancer cell proliferation. Multimodality imaging using visible light fluorescence and X-ray microscopy is demonstrated by tracing the nanoparticle-loaded tumor cells.     

Photon counting imaging: applications in biomedical research

Photon counting imaging, a technique capable of imaging at the single photon level, is finding applications in biological research and is providing unprecedented views of ultra-low light level phenomena. In combination with the optical microscope, this technique has provided a means of directly visualizing gene expression in single cells, imaging metabolites in tumor tissue and visualizing the chemiluminescence associated with oxidative metabolism in phagocytic cells. At the macroscopic level, it has greatly extended the sensitivity of detection in protein blots and has been applied as an image luminometer to assay microtiter plates. The technique holds great promise for use with fluorescence- and luminescence-based methods in many fields of research.     

Selection of quantum dot wavelengths for biomedical assays and imaging

Fluorescent semiconductor nanocrystals (quantum dots [QDs]) are hypothesized to be excellent contrast agents for biomedical assays and imaging. A unique property of QDs is that their absorbance increases with increasing separation between excitation and emission wavelengths. Much of the enthusiasm for using QDs in vivo stems from this property, since photon yield should be proportional to the integral of the broadband absorption. In this study, we demonstrate that tissue scatter and absorbance can sometimes offset increasing QD absorption at bluer wavelengths, and counteract this potential advantage. By using a previously validated mathematical model, we explored the effects of tissue absorbance, tissue scatter, wavelength dependence of the scatter, water-to-hemoglobin ratio, and tissue thickness on QD performance. We conclude that when embedded in biological fluids and tissues, QD excitation wavelengths will often be quite constrained, and that excitation and emission wavelengths should be selected carefully based on the particular application. Based on our results, we produced near-infrared QDs optimized for imaging surface vasculature with white light excitation and a silicon CCD camera, and used them to image the coronary vasculature in vivo. Taken together, our data should prove useful in designing fluorescent QD contrast agents optimized for specific biomedical applications.     

Exposure render: an interactive photo-realistic volume rendering framework

The field of volume visualization has undergone rapid development during the past years, both due to advances in suitable computing hardware and due to the increasing availability of large volume datasets. Recent work has focused on increasing the visual realism in Direct Volume Rendering (DVR) by integrating a number of visually plausible but often effect-specific rendering techniques, for instance modeling of light occlusion and depth of field. Besides yielding more attractive renderings, especially the more realistic lighting has a positive effect on perceptual tasks. Although these new rendering techniques yield impressive results, they exhibit limitations in terms of their exibility and their performance. Monte Carlo ray tracing (MCRT), coupled with physically based light transport, is the de-facto standard for synthesizing highly realistic images in the graphics domain, although usually not from volumetric data. Due to the stochastic sampling of MCRT algorithms, numerous effects can be achieved in a relatively straight-forward fashion. For this reason, we have developed a practical framework that applies MCRT techniques also to direct volume rendering (DVR). With this work, we demonstrate that a host of realistic effects, including physically based lighting, can be simulated in a generic and flexible fashion, leading to interactive DVR with improved realism. In the hope that this improved approach to DVR will see more use in practice, we have made available our framework under a permissive open source license.     

A generalizable NLP framework for fast development of pattern-based biomedical relation extraction systems

Background

Text mining is increasingly used in the biomedical domain because of its ability to automatically gather information from large amount of scientific articles. One important task in biomedical text mining is relation extraction, which aims to identify designated relations among biological entities reported in literature. A relation extraction system achieving high performance is expensive to develop because of the substantial time and effort required for its design and implementation. Here, we report a novel framework to facilitate the development of a pattern-based biomedical relation extraction system. It has several unique design features: (1) leveraging syntactic variations possible in a language and automatically generating extraction patterns in a systematic manner, (2) applying sentence simplification to improve the coverage of extraction patterns, and (3) identifying referential relations between a syntactic argument of a predicate and the actual target expected in the relation extraction task.

Results

A relation extraction system derived using the proposed framework achieved overall F-scores of 72.66% for the Simple events and 55.57% for the Binding events on the BioNLP-ST 2011 GE test set, comparing favorably with the top performing systems that participated in the BioNLP-ST 2011 GE task. We obtained similar results on the BioNLP-ST 2013 GE test set (80.07% and 60.58%, respectively). We conducted additional experiments on the training and development sets to provide a more detailed analysis of the system and its individual modules. This analysis indicates that without increasing the number of patterns, simplification and referential relation linking play a key role in the effective extraction of biomedical relations.

Conclusions

In this paper, we present a novel framework for fast development of relation extraction systems. The framework requires only a list of triggers as input, and does not need information from an annotated corpus. Thus, we reduce the involvement of domain experts, who would otherwise have to provide manual annotations and help with the design of hand crafted patterns. We demonstrate how our framework is used to develop a system which achieves state-of-the-art performance on a public benchmark corpus.     

Spline-based semiparametric projected generalized estimating equation method for panel count data

We propose to analyze panel count data using a spline-based semiparametric projected generalized estimating equation (GEE) method with the proportional mean model E(N(t)|Z) = Λ(0)(t) e(β(0)(T)Z). The natural logarithm of the baseline mean function, logΛ(0)(t), is approximated by a monotone cubic B-spline function. The estimates of regression parameters and spline coefficients are obtained by projecting the GEE estimates into the feasible domain using a weighted isotonic regression (IR). The proposed method avoids assuming any parametric structure of the baseline mean function or any stochastic model for the underlying counting process. Selection of the working covariance matrix that accounts for overdispersion improves the estimation efficiency and leads to less biased variance estimations. Simulation studies are conducted using different working covariance matrices in the GEE to investigate finite sample performance of the proposed method, to compare the estimation efficiency, and to explore the performance of different variance estimates in presence of overdispersion. Finally, the proposed method is applied to a real data set from a bladder tumor clinical trial.     

Applications of ATR-FTIR spectroscopic imaging to biomedical samples

FTIR spectroscopic imaging in ATR (Attenuated Total Reflection) mode is a powerful tool for studying biomedical samples. This paper summarises recent advances in the applications of ATR-FTIR imaging to dissolution of pharmaceutical formulations and drug release. The use of two different ATR accessories to obtain chemical images of formulations in contact with water as a function of time is demonstrated. The innovative use of the diamond ATR accessory allowed in situ imaging of tablet compaction and dissolution. ATR-FTIR imaging was also applied to obtain images of the surface of skin and the spatial distribution of protein and lipid rich domains was obtained. Chemical images of cross-section of rabbit aorta were obtained using a diamond ATR accessory and the possibility of in situ imaging of arterial samples in contact with aqueous solution was demonstrated for the first time. This experiment opens an opportunity to image arterial samples in contact with solutions containing drug molecules. This approach may help in understanding the mechanisms of treatment of atherosclerosis.     

Bayesian structural time series for biomedical sensor data: A flexible modeling framework for evaluating interventions

The development of mobile-health technology has the potential to revolutionize personalized medicine. Biomedical sensors (e.g., wearables) can assist with determining treatment plans for individuals, provide quantitative information to healthcare providers, and give objective measurements of health, leading to the goal of precise phenotypic correlates for genotypes. Even though treatments and interventions are becoming more specific and datasets more abundant, measuring the causal impact of health interventions requires careful considerations of complex covariate structures, as well as knowledge of the temporal and spatial properties of the data. Thus, interpreting biomedical sensor data needs to make use of specialized statistical models. Here, we show how the Bayesian structural time series framework, widely used in economics, can be applied to these data. This framework corrects for covariates to provide accurate assessments of the significance of interventions. Furthermore, it allows for a time-dependent confidence interval of impact, which is useful for considering individualized assessments of intervention efficacy. We provide a customized biomedical adaptor tool, MhealthCI, around a specific implementation of the Bayesian structural time series framework that uniformly processes, prepares, and registers diverse biomedical data. We apply the software implementation of MhealthCI to a structured set of examples in biomedicine to showcase the ability of the framework to evaluate interventions with varying levels of data richness and covariate complexity and also compare the performance to other models. Specifically, we show how the framework is able to evaluate an exercise intervention’s effect on stabilizing blood glucose in a diabetes dataset. We also provide a future-anticipating illustration from a behavioral dataset showcasing how the framework integrates complex spatial covariates. Overall, we show the robustness of the Bayesian structural time series framework when applied to biomedical sensor data, highlighting its increasing value for current and future datasets.     

Applications of ATR-FTIR spectroscopic imaging to biomedical samples

FTIR spectroscopic imaging in ATR (Attenuated Total Reflection) mode is a powerful tool for studying biomedical samples. This paper summarises recent advances in the applications of ATR-FTIR imaging to dissolution of pharmaceutical formulations and drug release. The use of two different ATR accessories to obtain chemical images of formulations in contact with water as a function of time is demonstrated. The innovative use of the diamond ATR accessory allowed in situ imaging of tablet compaction and dissolution. ATR-FTIR imaging was also applied to obtain images of the surface of skin and the spatial distribution of protein and lipid rich domains was obtained. Chemical images of cross-section of rabbit aorta were obtained using a diamond ATR accessory and the possibility of in situ imaging of arterial samples in contact with aqueous solution was demonstrated for the first time. This experiment opens an opportunity to image arterial samples in contact with solutions containing drug molecules. This approach may help in understanding the mechanisms of treatment of atherosclerosis.