Publication of scientific research presented at scientific meetings of the British Association of Oral and Maxillofacial Surgeons: 10 years on – have we published or perished?

In 2009 we evaluated the publication of research presented at annual scientific meetings of the British Association of Oral and Maxillofacial Surgeons (BAOMS) 2002–2006, inclusive. Since then, the format of these meetings has changed, there has been a rapid increase in the number of online-only journals, and restraints on time during training and consultant practice have continued. We have therefore investigated the pattern of publication after presentation at these meetings between 2010 and 2014. All abstracts accepted for oral presentations or posters were included, and publication had to follow no more than four years later. We searched PubMed for papers in peer-reviewed journals and compared the data with those from 2002–2006. A total of 975 abstracts were accepted (2010–2014) of which 221 (23%) went on to be published. The median (IQR) delay to publication was 13 (4–25) months. Most were clinical papers from groups based in the UK (p < 0.001) and most were published in BJOMS (p < 0.001). The rate of publication has not changed significantly between the two periods (23% compared with 24%), and patterns in the type of papers, delays, journals, and research groups, were similar. Despite consistent rates of publication within the specialty, OMFS produces fewer publications after presentation than other surgical specialties. Further research is required to evaluate this more fully.     

Recognition of medication information from discharge summaries using ensembles of classifiers

ABSTRACT: BACKGROUND: Extraction of clinical information such as medications or problems from clinical text is an important task of clinical natural language processing (NLP). Rule-based methods are often used in clinical NLP systems because they are easy to adapt and customize. Recently, supervised machine learning methods have proven to be effective in clinical NLP as well. However, combining different classifiers to further improve the performance of clinical entity recognition systems has not been investigated extensively. Combining classifiers into an ensemble classifier presents both challenges and opportunities to improve performance in such NLP tasks. METHODS: We investigated ensemble classifiers that used different voting strategies to combine outputs from three individual classifiers: a rule-based system, a support vector machine (SVM) based system, and a conditional random field (CRF) based system. Three voting methods were proposed and evaluated using the annotated data sets from the 2009 i2b2 NLP challenge: simple majority, local SVM-based voting, and local CRF-based voting. RESULTS: Evaluation on 268 manually annotated discharge summaries from the i2b2 challenge showed that the local CRF-based voting method achieved the best F-score of 90.84% (94.11% Precision, 87.81% Recall) for 10-fold cross-validation. We then compared our systems with the first-ranked system in the challenge by using the same training and test sets. Our system based on majority voting achieved a better F-score of 89.65% (93.91% Precision, 85.76% Recall) than the previously reported F-score of 89.19% (93.78% Precision, 85.03% Recall) by the first-ranked system in the challenge. CONCLUSIONS: Our experimental results using the 2009 i2b2 challenge datasets showed that ensemble classifiers that combine individual classifiers into a voting system could achieve better performance than a single classifier in recognizing medication information from clinical text. It suggests that simple strategies that can be easily implemented such as majority voting could have the potential to significantly improve clinical entity recognition.