Biomedical Domain

This paper addresses the problem of extracting and processing relevant information from unstructured electronic documents of the biomedical domain. The documents are full scientific papers. This problem imposes several challenges, such as identifying text passages that contain relevant information, collecting the relevant information pieces, populating a database and a data warehouse, and mining these data. For this purpose, this paper proposes the IEDSS-Bio, an environment for Information Extraction and Decision Support System in Biomedical domain. In a case study, experiments with machine learning for identifying relevant text passages (disease and treatment effects, and patients number information on Sickle Cell Anemia papers) showed that the best results (95.9% accuracy) were obtained with a statistical method and the use of preprocessing techniques to resample the examples and to eliminate noise.

The complexity and scale of the knowledge in the biomedical domain has motivated research work towards mining heterogeneous data from structured and unstructured knowledge bases. Towards this direction, it is necessary to combine facts in order to formulate hypotheses or draw conclusions about the domain concepts. In this work we attempt to address this problem by using indirect knowledge connecting two concepts in a graph to identify hidden relations between them. The graph represents concepts as vertices and relations as edges, stemming from structured (ontologies) and unstructured (text) data. In this graph we attempt to mine path patterns which potentially characterize a biomedical relation. For our experimental evaluation we focus on two frequent relations, namely "has target", and "may treat". Our results suggest that relation discovery using indirect knowledge is possible, with an AUC that can reach up to 0.8. Finally, analysis of the results indicates tha...

Background

The complexity and scale of the knowledge in the biomedical domain has motivated research work towards mining heterogeneous data from both structured and unstructured knowledge bases. Towards this direction, it is necessary to combine facts in order to formulate hypotheses or draw conclusions about the domain concepts. This work addresses this problem by using indirect knowledge connecting two concepts in a knowledge graph to discover hidden relations between them. The graph represents concepts as vertices and relations as edges, stemming from structured (ontologies) and unstructured (textual) data. In this graph, path patterns, i.e. sequences of relations, are mined using distant supervision that potentially characterize a biomedical relation.

Results

It is possible to identify characteristic path patterns of biomedical relations from this representation using machine learning. For experimental evaluation two frequent biomedical relations, namely “has target”, and “may treat”, are chosen. Results suggest that relation discovery using indirect knowledge is possible, with an AUC that can reach up to 0.8, a result which is a great improvement compared to the random classification, and which shows that good predictions can be prioritized by following the suggested approach.

Conclusions

Analysis of the results indicates that the models can successfully learn expressive path patterns for the examined relations. Furthermore, this work demonstrates that the constructed graph allows for the easy integration of heterogeneous information and discovery of indirect connections between biomedical concepts.

The advent of the Semantic Web and, more recently, of the Linked Data initiative, has paved the way for new perspectives and opportunities in Terminology , namely regarding the operationalization of terminological products. Within the biomedical domain, changes have been substantial in the past decades and at their heart stand the current challenges regarding the production, use, storage and dissemination of medical data, information, and knowledge. In a context where biomedical terminological resources are becoming increasingly concept-oriented, terminology work should reflect a double dimension (both linguistic and conceptual) that may, in turn, support the aspired operationalization and in-teroperability in this field. Therefore, the purpose of this paper is to present a case study, based around the concept of , in which a methodology anchored in Terminology's double dimension aims to contribute to the enrichment of the Systematized Nomenclature of Medicine-Clinical Terms (SN...

This paper addresses the problem of extracting and processing relevant information from unstructured electronic documents of the biomedical domain. The documents are full scientific papers. This problem imposes several challenges, such as identifying text passages that contain relevant information, collecting the relevant information pieces, populating a database and a data warehouse, and mining these data. For this purpose, this paper proposes the IEDSS-Bio, an environment for Information Extraction and Decision Support System in Biomedical domain. In a case study, experiments with machine learning for identifying relevant text passages (disease and treatment effects, and patients number information on Sickle Cell Anemia papers) showed that the best results (95.9% accuracy) were obtained with a statistical method and the use of preprocessing techniques to resample the examples and to eliminate noise.