JMIR Publications

ABSTRACT

Background:

Obtaining and describing the semiology efficiently and classifying seizures types correctly are crucial for the diagnosis and treatment of epilepsy. Nevertheless, there exists an inadequacy in related informatics resources and decision-support tools.

Objective:

We developed an ontology-based symptom extraction tool and employed machine learning to achieve automated binary classification of epilepsy in this study.

Methods:

Using present history data of electronic health record (EHR) from the Southwest Epilepsy Centre in China, we constructed a epilepsy semiology ontology and a symptom-entity extraction tool to extract seizure duration, seizure symptoms, and seizure frequency from the unstructured text by combining manual annotation with NLP techniques. Additionally, we achieved automatic classification of patients in the study cohort with high accuracy based on the extracted seizure feature data using multiple machine learning methods.

Results:

Data include present history from 10,925 cases between 2010 and 2020. Six annotators labelled a total of 2,500 texts to obtain 5844 words of semiology and construct a epilepsy semiology ontology(ESO) with 702 terms. Based on the ontology, the extraction tool achievd an accuracy rate of 85% in symptom extraction. Furthermore, We trained a Stacking ensemble learning model combining XGBoost and Random Forest with a F1 score of 75.03%. And the Random Forest model had the highest area under the curve (AUC) of 0.984.

Conclusions:

This work demonstrated the feasibility of NLP-assisted structural extraction of epilepsy medical record texts and downstream tasks, providing open ontology resources for subsequent related work.