Editorial Board
NCBI disease corpus
Graphical abstractDisplay Omitted NCBI disease corpus is built as a gold-standard resource for disease recognition.793 PubMed abstracts are annotated with disease mentions and concepts (MeSH/OMIM).14 Annotators produced high consistency level and inter-...
Pre-launch evaluation checklist for online health-promoting communities
Display Omitted A pre-launch evaluation checklist of online communities was developed.A delicate balance has to be meet between community autonomy and quality control.A comparison with the study context is needed before employing the checklist.Future ...
Semantic concept-enriched dependence model for medical information retrieval
Graphical abstractDisplay Omitted We incorporated semantic concept enriched dependence into the medical IR algorithm.The semantic concept-enriched dependence model showed robust performance.Performance gain was achieved independently from other previous ...
A biological continuum based approach for efficient clinical classification
A significant speedup of 4.73-fold was achieved with the employment of BCEN.BCEN provides a reusable framework for clinical feature pre-selection.Efficient redevelopment of up-to-date clinical classification models is possible with BCEN.Risk factors ...
Discovery of clinical pathway patterns from event logs using probabilistic topic models
Graphical AbstractDisplay Omitted We propose a probabilistic model to discover underlying patterns in clinical pathways.Our approach models treatment activities and that of occurrences in clinical pathways.Evaluation is done using a real clinical ...
Curbing false discovery rates in interpretation of genome-wide expression profiles
Display Omitted HighlightsFisher's exact test (FE) is widely used in genomics analysis.In most databases, the frequency distribution of genes is right skewed.This contradicts the null hypothesis of the FE test, thus leading to excessive false ...
FALCON or how to compute measures time efficiently on dynamically evolving dense complex networks?
Display Omitted A new time efficient complex network analysis library for dense networks is proposed.It is suited for the analysis, e.g. of several thousands of distributed brain sources.High performance is achieved by combined hard- and software ...
Understanding semantic mapping evolution by observing changes in biomedical ontologies
Display Omitted Correlation between the evolution of medical ontologies and their mappings.Quantitative analysis of medical ontologies and mappings evolution.Analysis of SNOMED CT and ICD-9-CM evolution. Knowledge Organization Systems (KOSs) are ...
Determining the difficulty of Word Sense Disambiguation
We explore estimating WSD performance on a range of ambiguous biomedical terms.We evaluate the difficulty predictions against the output of two WSD systems.Supervised methods are the best predictors but limited by labeled training data.Unsupervised ...
Supervised methods for symptom name recognition in free-text clinical records of traditional Chinese medicine
Symptom name recognition from free-text clinical records of TCM is well studied.Supervised methods are investigated and helpful strategies are proposed.Elaborate experiments are implemented and results are explained carefully.Effectiveness of the ...
Cross-domain targeted ontology subsets for annotation
Display Omitted We propose a computationally-efficient approach to ontology subset extraction.We use the SNOMED CORE subset as source; NDF-RT, RxNorm, and the UMLS Metathesaurus.We extract a targeted subset of drugs, RxNorm/CORE, 4% the size of ...
ReVeaLD
Display Omitted We develop a visual approach for intuitive query formulation by non-technical users.53 Linked biomedical data sources are queried in real time for data discovery.User-defined graphic rules dynamically assemble media-rich interfaces from ...
ProNormz - An integrated approach for human proteins and protein kinases normalization
Graphical abstractDisplay Omitted We present 'ProNormz' as a freely available web-based tool for human protein normalization.Besides normalization ProNormz distinguishes human protein kinases from other proteins.ProNormz is the first system to ...
Identifying non-elliptical entity mentions in a coordinated NP with ellipses
- Jeongmin Chae,
- Younghee Jung,
- Taemin Lee,
- Soonyoung Jung,
- Chan Huh,
- Gilhan Kim,
- Hyeoncheol Kim,
- Heungbum Oh
Display Omitted Our NER method resolves simple and complex ellipses in coordinated NPs.We presented two formal notations to express syntactic relationships between words.We model the process of making non-elliptical entity mentions into a coordinated ...
An SVM-based high-quality article classifier for systematic reviews
We classify high-quality articles using other topics' articles.Our classifiers by trained on articles having common properties improve the AUC.Our method can reduce the labor required for the new systematic reviews. ObjectiveTo determine whether SVM-...
Compass
Traditional hypotheses testing is not ideal for knowledge discovery in large data.We developed an approach for data mining intended as a hypothesis generating tool.Our approach is able to handle incomplete information.The model can handle both ...
Automatic signal extraction, prioritizing and filtering approaches in detecting post-marketing cardiovascular events associated with targeted cancer drugs from the FDA Adverse Event Reporting System (FAERS)
Graphical abstractDisplay Omitted Targeted cancer drugs are often associated with unexpectedly high cardiotoxicity.The Mechanisms underlying these drug-cardiovascular (CV) relationships are unclear.Systems approaches in studying targeted cancer drug-CV ...
Structural network analysis of biological networks for assessment of potential disease model organisms
Potential model organisms may be identified through the study of disease pathways.A graph mining technique can identify "fingerprints" from disease pathways.Potential model organism relevance can be determined based on these fingerprints.Non-obvious ...
Contrasting lexical similarity and formal definitions in SNOMED CT
SNOMED CT is a large and complex terminology and imperfections are inevitable.We propose an algorithmic quality assurance method to find inconsistent concepts.We formulate similarity sets: groups of concepts with similar lexical structure.Five different ...
