Adapting Abstract Meaning Representation Parsing to the Clinical Narrative -- the SPRING THYME parser

TL;DR

This paper presents the development and evaluation of the first AMR parser specifically designed for clinical notes, achieving high accuracy and demonstrating the importance of domain adaptation in clinical NLP.

Contribution

It introduces a novel clinical AMR parser trained with data augmentation and domain adaptation techniques, tailored for clinical narratives and evaluated on the THYME corpus.

Findings

Achieved an F1 score of 88% on clinical data

Demonstrated the effectiveness of data augmentation for domain-specific AMR parsing

Explored data requirements for clinical domain adaptation

Abstract

This paper is dedicated to the design and evaluation of the first AMR parser tailored for clinical notes. Our objective was to facilitate the precise transformation of the clinical notes into structured AMR expressions, thereby enhancing the interpretability and usability of clinical text data at scale. Leveraging the colon cancer dataset from the Temporal Histories of Your Medical Events (THYME) corpus, we adapted a state-of-the-art AMR parser utilizing continuous training. Our approach incorporates data augmentation techniques to enhance the accuracy of AMR structure predictions. Notably, through this learning strategy, our parser achieved an impressive F1 score of 88% on the THYME corpus's colon cancer dataset. Moreover, our research delved into the efficacy of data required for domain adaptation within the realm of clinical notes, presenting domain adaptation data requirements for AMR parsing. This exploration not only underscores the parser's robust performance but also highlights its potential in facilitating a deeper understanding of clinical narratives through structured semantic representations.