Literature Retrieval for Precision Medicine with Neural Matching and Faceted Summarization

Findings of the Association for Computational Linguistics: EMNLP 2020, pages 3389–3399 November 16 - 20, 2020. c⃝2020 Association for Computational Linguistics 3389 Literature Retrieval for Precision Medicine with Neural Matching and Faceted Summarization Jiho Noh Department of Computer Science University of Kentucky Kentucky, USA jiho.noh@uky.edu Ramakanth Kavuluru Division of Biomedical Informatics University of Kentucky Kentucky, USA ramakanth.kavuluru@uky.edu Abstract Information retrieval (IR) for precision medicine (PM) often involves looking for multiple pieces of evidence that characterize a patient case. This typically includes at least the name of a condition and a genetic variation that applies to the patient. Other factors such as demographic attributes, comorbidities, and social determinants may also be pertinent. As such, the retrieval problem is often formulated as ad hoc search but with multiple facets (e.g., disease, mutation) that may need to be incorporated. In this paper, we present a document reranking approach that combines neural query-document matching and text summarization toward such retrieval scenarios. Our architecture builds on the basic BERT model with three specific components for reranking: (a). document-query matching (b). keyword extraction and (c). facet-conditioned abstractive summarization. The outcomes of (b) and (c) are used to essentially transform a candidate document into a concise summary that can be compared with the query at hand to compute a relevance score. Component (a) directly generates a matching score of a candidate document for a query. The full architecture benefits from the complementary potential of document-query matching and the novel document transformation approach based on summarization along PM facets. Evaluations using NIST’s TREC-PM track datasets (2017–2019) show that our model achieves state-of-the-art performance. To fos- ter reproducibility, our code is made available here: https://github.com/bionlproc/ text-summ-for-doc-retrieval. 1 Introduction The U.S. NIH’s precision medicine (PM) initia- tive (Collins and Varmus, 2015) calls for designing treatment and preventative interventions consider- ing genetic, clinical, social, behavioral, and en- vironmental exposure variability among patients. The initiative rests on the widely understood find- ing that considering individual variability is critical in tailoring healthcare interventions to achieve sub- stantial progress in reducing disease burden world- wide. Cancer was chosen as its near term focus with the eventual aim of expanding to other condi- tions. As the biomedical research enterprise strives to fulfill the initiative’s goals, computing needs are also on the rise in drug discovery, predictive modeling for disease onset and progression, and in building NLP tools to curate information from the evidence base being generated. 1.1 TREC Precision Medicine Series Facet Input Disease Melanoma Genetic variation BRAF (E586K) Demographics 64-year-old female Disease Gastric cancer Genetic variation ERBB2 amplification Demographics 64-year-old male Table 1: Example cases from 2019 TREC-PM dataset In a dovetailing move, the U.S. NIST’s TREC (Text REtrieval Conference) has been running a PM track since 2017 with a focus on cancer (Roberts et al., 2020). The goal of the TREC-PM task is to identify the most relevant biomedical articles and clinical trials for an input patient case. Each case is composed of (1) a disease name, (2) a gene name and genetic …

Content truncated to 500 words