Using Convolutional Neural Networks to Support Insertion of New Concepts into SNOMED CT

Abstract

Many major medical ontologies go through a regular (bi-annual, monthly, etc.) release cycle. A new release will contain corrections to the previous release, as well as genuinely new concepts that are the result of either user requests or new developments in the domain. New concepts need to be placed at the correct place in the ontology hierarchy. Traditionally, this is done by an expert modeling a new concept and running a classifier algorithm. We propose an alternative approach that is based on providing only the name of a new concept and using a Convolutional Neural Network-based machine learning method. We first tested this approach within one version of SNOMED CT and achieved an average 88.5% precision and an F1 score of 0.793. In comparing the July 2017 release with the January 2018 release, limiting ourselves to predicting one out of two or more parents, our average F1 score was 0.701.

Figure 1:

Serializing the hierarchical structure of one concept into one document

Figure 2:

Data flow during vectorization phase. D = 336,893 is the total number of concepts.

Figure 3:

Data flow during training the CNN model. Test data is also generated “on the fly.”

Figure 4:

Data flow during testing CNN mode phase

Figure 5:

Data flow during testing CNN model phase with new data (use case)