Knowledge Transfer with Medical Language Embeddings

TL;DR

This paper introduces a probabilistic model that combines structured medical knowledge graphs with unstructured clinical text to predict new relationships between medical concepts, addressing data sparsity issues.

Contribution

It presents a novel approach integrating semantic embeddings from knowledge graphs and text to improve medical relationship prediction and knowledge graph completion.

Findings

Successfully predicts relationships not in the database

Enhances knowledge graph completion accuracy

Addresses data scarcity in medical NLP

Abstract

Identifying relationships between concepts is a key aspect of scientific knowledge synthesis. Finding these links often requires a researcher to laboriously search through scien- tific papers and databases, as the size of these resources grows ever larger. In this paper we describe how distributional semantics can be used to unify structured knowledge graphs with unstructured text to predict new relationships between medical concepts, using a probabilistic generative model. Our approach is also designed to ameliorate data sparsity and scarcity issues in the medical domain, which make language modelling more challenging. Specifically, we integrate the medical relational database (SemMedDB) with text from electronic health records (EHRs) to perform knowledge graph completion. We further demonstrate the ability of our model to predict relationships between tokens not appearing in the relational database.