Working Group

Predictive Big Data Analytics - Imaging-Genetics Fundamentals, Research Challenges, and Opportunities

Translational biomedical research relies on massive amounts of heterogeneous data and demands substantial transdisciplinary expertise to extract valuable information and acquire new knowledge about complex natural process. The Moore’s and Kryder’s laws guarantee exponential increase of computational power and information storage, respectively. These laws also dictate the rapid transdisciplinary advances, technological innovation and effective mechanisms for managing and interrogating Big Data.

Information Extraction from Clinical Notes for Secondary Analysis using Natural Language Processing

A significant amount of important clinical information is captured as free-text narratives. Clinical notes, radiology and pathology reports are examples of such unstructured clinical data. Information stored in unstructured data can be reused for a number of applications: clinical-decision support, evidence-based practice, and research.  However, extracting usable information from large datasets is difficult and time consuming. 

Natural Language Processing-assisted Text Annotation with BRAT

Natural Language Processing (NLP) methods are increasingly used to automate various tasks in the processing of medical and biological texts, and manually annotated text corpora are required for the development and evaluation of these methods. To assist in the creation of richly annotated biomedical corpora, text annotation tools need to integrate existing NLP methods, ontologies and database resources and provide users with an accessible interface for visualising and editing complex structured annotations.

V3NLP Framework and Leo – Tools to Build Applications that Extract Concepts from Clinical Text

Substantial amounts of clinically significant information such as symptoms and other personal details expressed by the patient to the provider are contained only within clinical electronic medical record notes. The field of clinical informatics is devoted to developing methods to accurately retrieve salient information from text for quality improvement, research, and population health. There are limited open source common platforms that perform these tasks. We present V3NLP Framework and LEO as an open source suite of functionalities to build such applications.

Biomedical Named-Entity Recognition: An Overview of MetaMap

Named-Entity Recognition (NER) is an important and difficult problem that is likely to remain an ongoing research area in biomedical informatics for the foreseeable future. The MetaMap system, developed by Dr. Alan (Lan) R. Aronson at the Lister Hill National Center for Biomedical Communications in the National Library of Medicine (NLM), is a well-known and often-cited NER application used the world over in academia, industry, government, and hospitals to analyze text drawn from both the biomedical literature (e.g., MEDLINE articles) and clinical reports.


Subscribe to RSS - Working Group