OBJECTIVE:To examine the feasibility of using statistical text classification to automatically identify health information technology (HIT) incidents in the USA Food and Drug Administration (FDA) Manufacturer and User Facility Device Experience (MAUDE) database.DESIGN:We used a subset of 570 272 incidents including 1534 HIT incidents reported to MAUDE between 1 January 2008 and 1 July 2010. Text classifiers using regularized logistic regression were evaluated with both 'balanced' (50% HIT) and 'stratified' (0.297% HIT) datasets for training, validation, and testing. Dataset preparation, feature extraction, feature selection, cross-validation, classification, performance evaluation, and error analysis were performed iteratively to further improve the classifiers. Feature-selection techniques such as removing short words and stop words, stemming, lemmatization, and principal component analysis were examined.MEASUREMENTS:κ statistic, F1 score, precision and recall.RESULTS:Classification performance was similar on both the stratified (0.954 F1 score) and balanced (0.995 F1 score) datasets. Stemming was the most effective technique, reducing the feature set size to 79% while maintaining comparable performance. Training with balanced datasets improved recall (0.989) but reduced precision (0.165).CONCLUSIONS:Statistical text classification appears to be a feasible method for identifying HIT reports within large databases of incidents. Automated identification should enable more HIT problems to be detected, analyzed, and addressed in a timely manner. Semi-supervised learning may be necessary when applying machine learning to big data analysis of patient safety incidents and requires further investigation.

BACKGROUND:The identification of drug characteristics is a clinically important task, but it requires much expert knowledge and consumes substantial resources. We have developed a statistical text-mining approach (BInary Characteristics Extractor and biomedical Properties Predictor: BICEPP) to help experts screen drugs that may have important clinical characteristics of interest.RESULTS:BICEPP first retrieves MEDLINE abstracts containing drug names, then selects tokens that best predict the list of drugs which represents the characteristic of interest. Machine learning is then used to classify drugs using a document frequency-based measure. Evaluation experiments were performed to validate BICEPP's performance on 484 characteristics of 857 drugs, identified from the Australian Medicines Handbook (AMH) and the PharmacoKinetic Interaction Screening (PKIS) database. Stratified cross-validations revealed that BICEPP was able to classify drugs into all 20 major therapeutic classes (100%) and 157 (of 197) minor drug classes (80%) with areas under the receiver operating characteristic curve (AUC) > 0.80. Similarly, AUC > 0.80 could be obtained in the classification of 173 (of 238) adverse events (73%), up to 12 (of 15) groups of clinically significant cytochrome P450 enzyme (CYP) inducers or inhibitors (80%), and up to 11 (of 14) groups of narrow therapeutic index drugs (79%). Interestingly, it was observed that the keywords used to describe a drug characteristic were not necessarily the most predictive ones for the classification task.CONCLUSIONS:BICEPP has sufficient classification power to automatically distinguish a wide range of clinical properties of drugs. This may be used in pharmacovigilance applications to assist with rapid screening of large drug databases to identify important characteristics for further evaluation.

This study aimed to develop a new informatics platform for the discovery, recovery and multi-level analysis of the effects of individual genes and multiple gene combinations on pathophenotypes of bacteria. Natural language processing algorithms were employed to extract gene-disease associations from PubMed literature and annotated genomes of bacteria with epidemic potential. From these associations gene virulence profiles were generated enabling the comparison of gene signatures within and across genomes. It allowed the identification of virulence genes and construction of their association networks as well as the detection of knowledge gaps. This proof-of-concept study confirmed the feasibility of our original approach for integrating bacterial genome level knowledge with published observations from clinical settings.

Background Computational discovery is playing an ever-greater role in supporting the processes of knowledge synthesis. A significant proportion of the more than 18 million manuscripts indexed in the PubMed database describe infectious disease syndromes and various infectious agents. This study is the first attempt to integrate online repositories of text-based publications and microbial genome databases in order to explore the dynamics of relationships between pathogens and infectious diseases. Methodology/Principal Findings Herein we demonstrate how the knowledge space of infectious diseases can be computationally represented and quantified, and tracked over time. The knowledge space is explored by mapping of the infectious disease literature, looking at dynamics of literature deposition, zooming in from pathogen to genome level and searching for new associations. Syndromic signatures for different pathogens can be created to enable a new and clinically focussed reclassification of the microbial world. Examples of syndrome and pathogen networks illustrate how multilevel network representations of the relationships between infectious syndromes, pathogens and pathogen genomes can illuminate unexpected biological similarities in disease pathogenesis and epidemiology. Conclusions/Significance This new approach based on text and data mining can support the discovery of previously hidden associations between diseases and microbial pathogens, clinically relevant reclassification of pathogenic microorganisms and accelerate the translational research enterprise.

This paper describes a system developed for the transformation of English sentences into a first order logical form representation. The metho dology is centered on the use of a dependency grammar based parser . We demonstrate the suitability of applying a dependency parser based solution to the given task and in turn explain some of the limitations and challenges involved when using such an approach. The efficiencies and deficiencies of our approach are discussed as well as considerations for further enhanc e-