A neuro-symbolic method for understanding free-text medical evidence. (6th May 2021)

• Record Type: Journal Article
• Authors: Kang, Tian; Turfah, Ali; Kim, Jaehyun; Perotte, Adler; Weng, Chunhua
• Is Part Of: Journal of the American Medical Informatics Association. Volume 28:Number 8(2021)
• Page Start: 1703
• Page End: 1711
• Publication Date: 2021-05-06
• Subjects: natural language understanding -- machine reading comprehension -- transformer -- medical evidence computing; Medical informatics -- Periodicals; Information Services -- Periodicals; Medical Informatics -- Periodicals; Médecine -- Informatique -- Périodiques; Informatica; Geneeskunde; Informatique médicale; Computer network resources; Electronic journals; 610.285
• DOI: 10.1093/jamia/ocab077
• Languages: English
• ISSNs: 1067-5027

Abstract

Abstract: Objective: We introduce Medical evidence Dependency ( MD )–informed attention, a novel neuro-symbolic model for understanding free-text clinical trial publications with generalizability and interpretability. Materials and Methods: We trained one head in the multi-head self-attention model to attend to the Medical evidence Ddependency ( MD ) and to pass linguistic and domain knowledge on to later layers ( MD informed). This MD-informed attention model was integrated into BioBERT and tested on 2 public machine reading comprehension benchmarks for clinical trial publications: Evidence Inference 2.0 and PubMedQA. We also curated a small set of recently published articles reporting randomized controlled trials on COVID-19 (coronavirus disease 2019) following the Evidence Inference 2.0 guidelines to evaluate the model's robustness to unseen data. Results: The integration of MD-informed attention head improves BioBERT substantially in both benchmark tasks—as large as an increase of +30% in the F1 score—and achieves the new state-of-the-art performance on the Evidence Inference 2.0. It achieves 84% and 82% in overall accuracy and F1 score, respectively, on the unseen COVID-19 data. Conclusions: MD-informed attention empowers neural reading comprehension models with interpretability and generalizability via reusable domain knowledge. Its compositionality can benefit any transformer-based architecture for machine reading comprehension of free-text medical evidence.