Relation extraction from clinical texts using domain invariant convolutional neural network

Abstract

In recent years extracting relevant information from biomedical and clinical texts such as research articles, discharge summaries, or electronic health records have been a subject of many research efforts and shared challenges. Relation extraction is the process of detecting and classifying the semantic relation among entities in a given piece of texts. Existing models for this task in biomedical domain use either manually engineered features or kernel methods to create feature vector. These features are then fed to classifier for the prediction of the correct class. It turns out that the results of these methods are highly dependent on quality of user designed features and also suffer from curse of dimensionality. In this work we focus on extracting relations from clinical discharge summaries. Our main objective is to exploit the power of convolution neural network (CNN) to learn features automatically and thus reduce the dependency on manual feature engineering. We evaluate performance of the proposed model on i2b2-2010 clinical relation extraction challenge dataset. Our results indicate that convolution neural network can be a good model for relation exaction in clinical text without being dependent on expert's knowledge on defining quality features.

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

“…Based on the dynamic range attention mechanism, we propose two kinds of MASK rc denoted as Eq. (13) and (14), respectively.…”

“…Neural network methods can extract the relation features without complicated feature engineering. e.g., recurrent capsule network [13] and domain invariant convolutional neural network [14]. However, These methods cannot utilize joint features between entity and relation, resulting in lower generalization performance when compared with joint learning methods.…”

Data-driven analysis of biomedical literature suggests broad-spectrum benevolence of culinary herbs and spices

“…After tagging spice/herb and disease entities, the task of relationship extraction can be treated as a multi-label classification problem with three classes of associations: positive, negative, and no associations. We used a Convolutional Neural Network (CNN) classifier with word, position, part of speech and chunk embedding [24][25][26] for extracting intra-sentence spicedisease relationships. It obtained an accuracy of 86.7% and macro-averaged precision, recall and F1 score of 90.7%, 80% and 84.2% respectively on an external test set.…”

“…Starting with preprocessed and manually labeled sentences, we developed a machine learning classifier for extracting relations from unlabeled sentences. We tested the following models [24][25][26] For both the CNN models, the word embedding was initialized using pre-trained weights from…”

Detecting Named Entities and Relations in German Clinical Reports

“…In context of detecting medical concepts (named entity recognition; NER) and their relations (relation extraction; RE) conditional random fields (CRF) (Lafferty et al 2001) and support vector machines (SVM) (Joachims 1999) have been very popular supervised methods that were frequently used for the last decade. In recent years neural network based supervised learning has gained popularity (see, e.g., Nguyen and Grishman (2015); Sahu et al (2016); Zeng et al (2014)). …”