Extracting seizure frequency from epilepsy clinic notes: a machine reading approach to natural language processing

• Published in: Journal of the American Medical Informatics Association : JAMIA Vol. 29; no. 5; pp. 873 - 881
• Main Authors: Xie, Kevin, Gallagher, Ryan S, Conrad, Erin C, Garrick, Chadric O, Baldassano, Steven N, Bernabei, John M, Galer, Peter D, Ghosn, Nina J, Greenblatt, Adam S, Jennings, Tara, Kornspun, Alana, Kulick-Soper, Catherine V, Panchal, Jal M, Pattnaik, Akash R, Scheid, Brittany H, Wei, Danmeng, Weitzman, Micah, Muthukrishnan, Ramya, Kim, Joongwon, Litt, Brian, Ellis, Colin A, Roth, Dan
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 13-04-2022
• Subjects: Electronic Health Records; Epilepsy; Humans; Natural Language Processing; Research and Applications; Retrospective Studies; Seizures; epilepsy; electronic medical record; question-answering; natural language processing
• DOI: 10.1093/jamia/ocac018
• EISSN: 1527-974X
• ISSN: 1527-974X; 1067-5027

Abstract
Objective
Seizure frequency and seizure freedom are among the most important outcome measures for patients with epilepsy. In this study, we aimed to automatically extract this clinical information from unstructured text in clinical notes. If successful, this could improve clinical decision-making in epilepsy patients and allow for rapid, large-scale retrospective research.
Materials and Methods
We developed a finetuning pipeline for pretrained neural models to classify patients as being seizure-free and to extract text containing their seizure frequency and date of last seizure from clinical notes. We annotated 1000 notes for use as training and testing data and determined how well 3 pretrained neural models, BERT, RoBERTa, and Bio_ClinicalBERT, could identify and extract the desired information after finetuning.
Results
The finetuned models (BERTFT, Bio_ClinicalBERTFT, and RoBERTaFT) achieved near-human performance when classifying patients as seizure free, with BERTFT and Bio_ClinicalBERTFT achieving accuracy scores over 80%. All 3 models also achieved human performance when extracting seizure frequency and date of last seizure, with overall F1 scores over 0.80. The best combination of models was Bio_ClinicalBERTFT for classification, and RoBERTaFT for text extraction. Most of the gains in performance due to finetuning required roughly 70 annotated notes.
Discussion and Conclusion
Our novel machine reading approach to extracting important clinical outcomes performed at or near human performance on several tasks. This approach opens new possibilities to support clinical practice and conduct large-scale retrospective clinical research. Future studies can use our finetuning pipeline with minimal training annotations to answer new clinical questions.