A Strategy on Selecting Performance Metrics for Classifier Evaluation

2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

et al. 2019

Self Cite

Entity and relation extraction is the necessary step in structuring medical text. However, the feature extraction ability of the bidirectional long short term memory network in the existing model does not achieve the best effect. At the same time, the language model has achieved excellent results in more and more natural language processing tasks. In this paper, we present a focused attention model for the joint entity and relation extraction task. Our model integrates well-known BERT language model into joint learning through dynamic range attention mechanism, thus improving the feature representation ability of shared parameter layer. Experimental results on coronary angiography texts collected from Shuguang Hospital show that the F1-scores of named entity recognition and relation classification tasks reach 96.89% and 88.51%, which outperform state-of-the-art methods by 1.65% and 1.22%, respectively.

Section: Experimental Studies a Dataset And Evaluation Metricsmentioning

confidence: 99%

Fine-tuning BERT for Joint Entity and Relation Extraction in Chinese Medical Text

Kui

2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

et al. 2019

Self Cite

“…Clinical text structuring is an important problem which is highly related to practical applications. Considerable efforts arXiv:1908.06606v2 [cs.CL] 22 Oct 2019 have been made on CTS task. These studies can be roughly divided into three categories, namely rule and dictionary based methods, task-specific end-to-end methods and pipeline methods.…”

Section: Related Work a Clinical Text Structuringmentioning

confidence: 99%

Question Answering based Clinical Text Structuring Using Pre-trained Language Model

Qiu

2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

et al. 2019

Self Cite

Clinical text structuring is a critical and fundamental task for clinical research. Traditional methods such as taskspecific end-to-end models and pipeline models usually suffer from the lack of dataset and error propagation. In this paper, we present a question answering based clinical text structuring (QA-CTS) task to unify different specific CTS tasks and make dataset shareable. A novel model that aims to introduce domainspecific features (e.g., clinical named entity information) into pre-trained language model is also proposed for QA-CTS task. Experimental results on Chinese pathology reports collected from Ruijing Hospital demonstrate our presented QA-CTS task is very effective to improve the performance on specific tasks. Our proposed model also competes favorably with strong baseline models in specific tasks.

“…We further split these sentences into clauses by commas. Detailed statistics of different types of entities are listed in Table II. In the following experiments, widely-used performance measures such as precision, recall, and F 1 -score [40], [41] are used to evaluate the methods.…”

Section: A Dataset and Evaluation Metricsmentioning

confidence: 99%

Fast and Accurate Recognition of Chinese Clinical Named Entities with Residual Dilated Convolutions

Qiu

Wang

2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

et al. 2018

Self Cite

Clinical Named Entity Recognition (CNER) aims to identify and classify clinical terms such as diseases, symptoms, treatments, exams, and body parts in electronic health records, which is a fundamental and crucial task for clinical and translation research. In recent years, deep learning methods have achieved significant success in CNER tasks. However, these methods depend greatly on Recurrent Neural Networks (RNNs), which maintain a vector of hidden activations that are propagated through time, thus causing too much time to train models. In this paper, we propose a Residual Dilated Convolutional Neural Network with Conditional Random Field (RD-CNN-CRF) to solve it. Specifically, Chinese characters and dictionary features are first projected into dense vector representations, then they are fed into the residual dilated convolutional neural network to capture contextual features. Finally, a conditional random field is employed to capture dependencies between neighboring tags. Computational results on the CCKS-2017 Task 2 benchmark dataset show that our proposed RD-CNN-CRF method competes favorably with state-of-the-art RNN-based methods both in terms of computational performance and training time.