Deep learning with sentence embeddings pre-trained on biomedical corpora improves the performance of finding similar sentences in electronic medical records.

backgroundCapturing sentence semantics plays a vital role in a range of text mining applications. Despite continuous efforts on the development of related datasets and models in the general domain, both datasets and models are limited in biomedical and clinical domains. The BioCreative/OHNLP2018 organizers have made the first attempt to annotate 1068 sentence pairs from clinical notes and have called for a community effort to tackle the Semantic Textual Similarity (BioCreative/OHNLP STS) challenge.

methodsWe developed models using traditional machine learning and deep learning approaches. For the post challenge, we focused on two models: the Random Forest and the Encoder Network. We applied sentence embeddings pre-trained on PubMed abstracts and MIMIC-III clinical notes and updated the Random Forest and the Encoder Network accordingly.

resultsThe official results demonstrated our best submission was the ensemble of eight models. It achieved a Person correlation coefficient of 0.8328 - the highest performance among 13 submissions from 4 teams. For the post challenge, the performance of both Random Forest and the Encoder Network was improved; in particular, the correlation of the Encoder Network was improved by ~ 13%. During the challenge task, no end-to-end deep learning models had better performance than machine learning models that take manually-crafted features. In contrast, with the sentence embeddings pre-trained on biomedical corpora, the Encoder Network now achieves a correlation of ~ 0.84, which is higher than the original best model. The ensembled model taking the improved versions of the Random Forest and Encoder Network as inputs further increased performance to 0.8528.

conclusionsDeep learning models with sentence embeddings pre-trained on biomedical corpora achieve the highest performance on the test set. Through error analysis, we find that end-to-end deep learning models and traditional machine learning models with manually-crafted features complement each other by finding different types of sentences. We suggest a combination of these models can better find similar sentences in practice.