Disease knowledge graphs are a way to connect, organize, and access disparate information about diseases with numerous benefits for artificial intelligence (AI). To create knowledge graphs, it is necessary to extract knowledge from multimodal datasets in the form of relationships between disease concepts and normalize both concepts and relationship types.
We introduce REMAP, a multimodal approach for disease relation extraction and classification. The REMAP machine learning approach jointly embeds a partial, incomplete knowledge graph and a medical language dataset into a compact latent vector space, followed by aligning the multimodal embeddings for optimal disease relation extraction.
Systematized knowledge is becoming the backbone of AI, creating opportunities to inject semantics into AI and fully integrate it into machine learning algorithms. While prior semantic knowledge can assist in extracting disease relationships from text, existing methods can not fully leverage multimodal datasets. REMAP is a multimodal approach for extracting and classifying disease relationships by fusing structured knowledge and text information. REMAP provides a flexible neural architecture to easily find, access, and validate AI-driven relationships between disease concepts.
We introduce REMAP (Relation Extraction with Multimodal Alignment Penalty), a multimodal approach for extracting and classifying disease-disease relations. REMAP is a flexible multimodal algorithm that jointly learns over text and graphs with a unique capability to make predictions even when a disease concept exists in only one data type.
REMAP specifies graph-based and text-based deep transformation functions that embed each data type separately and optimize unimodal embedding spaces such that they capture the topology of a disease knowledge graph or the text semantics of disease concepts. To achieve data fusion, REMAP aligns unimodal embedding spaces through a novel alignment penalty loss using shared disease concepts as anchors. This way, REMAP can effectively model data type-specific distributions and diverse representations while also aligning embeddings of distinct data types. Further, REMAP can be jointly trained on both graph and text data types but evaluated on either of the two modalities alone.

In summary, the main contributions of this study are:
-
We develop REMAP, a flexible multimodal approach for extracting and classifying disease-disease relations. REMAP fuses knowledge graph embeddings with deep language models and can flexibly accommodate missing data types, which is necessary to facilitate REMAP’s validation and transition into biomedical implementation.
-
We rigorously evaluate REMAP for extraction and classification of disease-disease relations. To this end, we create a training dataset using distant supervision and a high-quality test dataset of gold-standard annotations provided by three domain experts, all medical doctors. Our evaluations show that REMAP achieves 88.6% micro-accuracy and 81.8% micro-F1 score on the human annotated dataset, outperforming text-based methods by 10 and 17.2 percentage points, respectively. Further, REMAP achieves the best performance, 89.8% micro-accuracy and 84.1% micro-F1 score, surpassing graph-based methods by 8.4 and 10.4 percentage points, respectively.
Examining the relation between hypogonadism and Goldberg-Maxwell
Hypogonadism and Goldberg-Maxwell syndrome are separate and distinct diseases. However, physicians find it challenging to differentiate in their diagnosis of these two diseases, especially in light of Goldberg-Maxwell syndrome being a rare disease. Differential diagnosis is a process wherein a doctor differentiates between two or more conditions that could be behind a person’s symptoms.
The figure below illustrates how a prediction of “differential diagnosis” (DDx) relationship between hypogonadism and Goldberg-Maxwell syndrome changes from incorrect to a correct prediction when using multimodal learning. REMAP correctly recognizes diseases from text and unifies them with disease concepts in the knowledge graph.
For example, hypogonadism and Goldberg-Maxwell syndrome have many outgoing edges of type differential diagnosis. Further, meta-paths connect these diseases in the graph, such as second-order “differential diagnosis → differential diagnosis” meta-path, and their outgoing degrees in the graph are relatively high. In the case of joint learning, the text-based model can extract part of the disease representation from the graph modality to update its internal representations and thus improve text-based classification of relations.

Publication
Multimodal Learning on Graphs for Disease Relation Extraction
Yucong Lin, Keming Lu, Sheng Yu, Tianxi Cai, and Marinka Zitnik
In Review 2022
@inproceedings{lin2022multimodal,
Title = {Multimodal Learning on Graphs for Disease Relation Extraction},
author = {Lin, Yucong and Lu, Keming and Yu, Sheng and Cai, Tianxi and Zitnik, Marinka},
booktitle = {In Review},
year = {2022}
}
Code
Datasets and Pytorch implementation of REMAP are available in the GitHub repository.