Event trigger identification for biomedical events extraction using domain knowledge

Motivation: In molecular biology, molecular events describe observable alterations of biomolecules, such as binding of proteins or RNA production. These events might be responsible for drug reactions or development of certain diseases. As such, biomedical event extraction, the process of automatically detecting description of molecular interactions in research articles, attracted substantial research interest recently. Event trigger identification, detecting the words describing the event types, is a crucial and prerequisite step in the pipeline process of biomedical event extraction. Taking the event types as classes, event trigger identification can be viewed as a classification task. For each word in a sentence, a trained classifier predicts whether the word corresponds to an event type and which event type based on the context features. Therefore, a well-designed feature set with a good level of discrimination and generalization is crucial for the performance of event trigger identification. Results: In this article, we propose a novel framework for event trigger identification. In particular, we learn biomedical domain knowledge from a large text corpus built from Medline and embed it into word features using neural language modeling. The embedded features are then combined with the syntactic and semantic context features using the multiple kernel learning method. The combined feature set is used for training the event trigger classifier. Experimental results on the golden standard corpus show that >2.5% improvement on F-score is achieved by the proposed framework when compared with the state-of-the-art approach, demonstrating the effectiveness of the proposed framework. © 2014 The Author 2014. The source code for the proposed framework is freely available and can be downloaded at http://cse.seu.edu.cn/people/zhoudeyu/ETI_Sourcecode.zip.

Educational Model of Computer as a Base for Informatics Learning

It is proposed to use one common model of computer for teaching different parts of the informatics course, connected with both hardware and software subjects. Reasoning of such slant is presented; the most suitable themes of the course, where it is practical, are enumerated. The own author's development (including software support) – the educational model of virtual computer "E97" and compiler from Pascal language for it – are described. It is accented, that the discussed ideas are helpful for any other similar model.

Faculty of Mathematics and Informatics, Sofia University

The Faculty of Mathematics and Informatics (FMI) of Sofia University “St. Kliment Ohridski” is briefly presented as an educational and research institution. The possible contribution of FMI to KT-DigiCULT-BG project is analyzed.

A self-referenced optical intensity sensor network using POFBGs for biomedical applications

This work bridges the gap between the remote interrogation of multiple optical sensors and the advantages of using inherently biocompatible low-cost polymer optical fiber (POF)-based photonic sensing. A novel hybrid sensor network combining both silica fiber Bragg gratings (FBG) and polymer FBGs (POFBG) is analyzed. The topology is compatible with WDM networks so multiple remote sensors can be addressed providing high scalability. A central monitoring unit with virtual data processing is implemented, which could be remotely located up to units of km away. The feasibility of the proposed solution for potential medical environments and biomedical applications is shown.

Business Informatics

A definition of the concept "business informatics" based on the General Information Theory is discussed in the paper.

One Application of the Informatics in the Descriptive Geometry

This work acquaints with a program for interactive computer training to students on the subject "Mutual intersecting of pyramids in axonometry ”. Our software is a set of three modules, which we call "student", "teacher" and "autopilot". It gives the final solution of the problem, the traceability of various significant moments in its solution and 3D-image of the finished composition of the two intersecting polyhedra, stripped of the working lines and subjected to rotation and translation.

Biomedical relation extraction:from binary to complex

Biomedical relation extraction aims to uncover high-quality relations from life science literature with high accuracy and efficiency. Early biomedical relation extraction tasks focused on capturing binary relations, such as protein-protein interactions, which are crucial for virtually every process in a living cell. Information about these interactions provides the foundations for new therapeutic approaches. In recent years, more interests have been shifted to the extraction of complex relations such as biomolecular events. While complex relations go beyond binary relations and involve more than two arguments, they might also take another relation as an argument. In the paper, we conduct a thorough survey on the research in biomedical relation extraction. We first present a general framework for biomedical relation extraction and then discuss the approaches proposed for binary and complex relation extraction with focus on the latter since it is a much more difficult task compared to binary relation extraction. Finally, we discuss challenges that we are facing with complex relation extraction and outline possible solutions and future directions.

Predictability of unplanned extubations

Acute life threatening events such as cardiac/respiratory arrests are often predictable in adults and children. However critical events such as unplanned extubations are considered as not predictable. This paper seeks to evaluate the ability of automated prediction systems based on feature space embedding and time series methods to predict unplanned extubations in paediatric intensive care patients. We try to exploit the trends in the physiological signals such as Heart Rate, Respiratory Rate, Systolic Blood Pressure and Oxygen saturation levels in the blood using signal processing aspects of a frame-based approach of expanding signals using a nonorthogonal basis derived from the data. We investigate the significance of the trends in a computerised prediction system. The results are compared with clinical observations of predictability. We will conclude by investigating whether the prediction capability of the system could be exploited to prevent future unplanned extubations. © 2014 IEEE.

Assessment Criteria for Non-auditorial Educational Forms in Informatics

Report published in the Proceedings of the National Conference on "Education in the Information Society", Plovdiv, May, 2013

Digitization of Old Mathematical Periodicals Published by the Institute of Mathematics and Informatics, Bulgarian Academy of Sciences

The digitization practice for retro-converting of the mathematical periodicals, published by the Institute of Mathematics and Informatics, Bulgarian Academy of Sciences (IMI-BAS) and the followed benefits for long-term preserving and assuring open access to these materials is discussed in the article.

Special issue on '6th CIRP-Sponsored International Conference of Digital Enterprise Technology (DET2009) - Enterprise Informatics'

Editorial

Educational Objectives and Evaluation of Student's Internship and Practice in Informatics

Report published in the Proceedings of the National Conference on "Education and Research in the Information Society", Plovdiv, May, 2014

Vacuum brazing of alumina ceramic to titanium for biomedical implants using pure gold as the filler metal

One of the many promising applications of metal/ceramic joining is in biomedical implantable devices. This work is focused on vacuum brazing of C.P titanium to 96% alumina ceramic using pure gold as the filler metal. A novel method of brazing is developed where resistance heating of C.P titanium is done inside a thermal evaporator using a Ta heating electrode. The design of electrode is optimized using Ansys resistive heating simulations. The materials chosen in this study are biocompatible and have prior history in implantable devices approved by FDA. This research is part of Boston Retinal implant project to make a biocompatible implantable device (www.bostonretina.org). ^ Pure gold braze has been used in the construction of single terminal feedthrough in low density hermetic packages utilizing a single platinum pin brazed to an alumina or sapphire ceramic donut (brazed to a titanium case or ferrule for many years in implantable pacemakers. Pure gold (99.99%) brazing of 96% alumina ceramic with CP titanium has been performed and evaluated in this dissertation. Brazing has been done by using electrical resistance heating. The 96% alumina ceramic disk was manufactured by high temperature cofired ceramic (HTCC) processing while the Ti ferrule and gold performs were purchased from outside. Hermetic joints having leak rate of the order of 1.6 × 10-8 atm-cc/ sec on a helium leak detector were measured. ^ Alumina ceramics made by HTCC processing were centreless grounded utilizing 800 grit diamond wheel to provide a smooth surface for sputtering of a thin film of Nb. Since pure alumina demonstrates no adhesion or wetting to gold, an adhesion layer must be used on the alumina surface. Niobium (Nb), Tantalum (Ta) and Tungsten (W) were chosen for evaluation since all are refractory (less dissolution into molten gold), all form stable oxides (necessary for adhesion to alumina) and all are readily thin film deposited as metals. Wetting studies are also performed to determine the wetting angle of pure gold to Ti, Ta, Nb and W substrates. Nano tribological scratch testing of thin film of Nb (which demonstrated the best wetting properties towards gold) on polished 96% alumina ceramic is performed to determine the adhesion strength of thin film to the substrate. The wetting studies also determined the thickness of the intermetallic compounds layers formed between Ti and gold, reaction microstructure and the dissolution of the metal into the molten gold.^