PICO element detection in medical text without metadata: Are first sentences enough?

Efficient identification of patient, intervention, comparison, and outcome (PICO) components in medical articles is helpful in evidence-based medicine. The purpose of this study is to clarify whether first sentences of these components are good enough to train naive Bayes classifiers for sentence-level PICO element detection. We extracted 19,854 structured abstracts of randomized controlled trials with any P/I/O label from PubMed for naive Bayes classifiers training. Performances of classifiers trained by first sentences of each section ( CF) and those trained by all sentences ( CA) were compared using all sentences by ten-fold cross-validation. The results measured by recall, precision, and F-measures show that there are no significant differences in performance between CF and CA for detection of O-element ( F-measure. = 0.731. ±. 0.009 vs. 0.738. ±. 0.010, p= 0.123). However, CA perform better for I-elements, in terms of recall (0.752. ±. 0.012 vs. 0.620. ±. 0.007, p<. 0.001) and F-measures (0.728. ±. 0.006 vs. 0.662. ±. 0.007, p<. 0.001). For P-elements, CF have higher precision (0.714. ±. 0.009 vs. 0.665. ±. 0.010, p<. 0.001), but lower recall (0.766. ±. 0.013 vs. 0.811. ±. 0.012, p<. 0.001). CF are not always better than CA in sentence-level PICO element detection. Their performance varies in detecting different elements.