Classifying medical relations in clinical text via convolutional neural networks

TL;DR

This paper introduces a CNN-based approach with multi-pooling and a category-level constraint loss for classifying medical relations in clinical texts, outperforming previous models without external features.

Contribution

It presents a novel CNN architecture with multi-pooling and a category-level constraint loss for improved medical relation classification.

Findings

Outperforms previous single-model methods

Competitive with ensemble-based approaches

Does not rely on external features

Abstract

Deep learning research on relation classification has achieved solid performance in the general domain. This study proposes a convolutional neural network (CNN) architecture with a multi-pooling operation for medical relation classification on clinical records and explores a loss function with a category-level constraint matrix. Experiments using the 2010 i2b2/VA relation corpus demonstrate these models, which do not depend on any external features, outperform previous single-model methods and our best model is competitive with the existing ensemble-based method.