Integration of Domain Knowledge using Medical Knowledge Graph Deep Learning for Cancer Phenotyping

TL;DR

This paper introduces a method to incorporate medical knowledge graphs into word embeddings to improve NLP tasks in cancer pathology report analysis, leading to significant performance gains.

Contribution

It presents a novel approach to integrate domain-specific medical knowledge into word embeddings for enhanced cancer phenotyping from pathology reports.

Findings

Knowledge graph-informed embeddings outperform standard embeddings.

Improved F1 scores by 4.97% micro and 22.5% macro.

Effective extraction of six cancer characteristics.

Abstract

A key component of deep learning (DL) for natural language processing (NLP) is word embeddings. Word embeddings that effectively capture the meaning and context of the word that they represent can significantly improve the performance of downstream DL models for various NLP tasks. Many existing word embeddings techniques capture the context of words based on word co-occurrence in documents and text; however, they often cannot capture broader domain-specific relationships between concepts that may be crucial for the NLP task at hand. In this paper, we propose a method to integrate external knowledge from medical terminology ontologies into the context captured by word embeddings. Specifically, we use a medical knowledge graph, such as the unified medical language system (UMLS), to find connections between clinical terms in cancer pathology reports. This approach aims to minimize the distance between connected clinical concepts. We evaluate the proposed approach using a Multitask Convolutional Neural Network (MT-CNN) to extract six cancer characteristics -- site, subsite, laterality, behavior, histology, and grade -- from a dataset of ~900K cancer pathology reports. The results show that the MT-CNN model which uses our domain informed embeddings outperforms the same MT-CNN using standard word2vec embeddings across all tasks, with an improvement in the overall micro- and macro-F1 scores by 4.97\%and 22.5\%, respectively.