Writing habits and telltale neighbors: analyzing clinical concept usage patterns with sublanguage embeddings

Abstract

Natural language processing techniques are being applied to increasingly diverse types of electronic health records, and can benefit from in-depth understanding of the distinguishing characteristics of medical document types. We present a method for characterizing the usage patterns of clinical concepts among different document types, in order to capture semantic differences beyond the lexical level. By training concept embeddings on clinical documents of different types and measuring the differences in their nearest neighborhood structures, we are able to measure divergences in concept usage while correcting for noise in embedding learning. Experiments on the MIMIC-III corpus demonstrate that our approach captures clinically-relevant differences in concept usage and provides an intuitive way to explore semantic characteristics of clinical document collections.

Medical Information Extraction in the Age of Deep Learning

“…Table 5). The best systems peak at 85% F 1 score for Advice (a distance of more than 13 percentage points to the top recognition results for medication-attributes), they slip to 78% 13 and 77% for Mechanism and Effect, respectively, and plummet to 59% for Interaction 14 . Differences between the first 13 Xu et al [86] even reach slightly more than 79% F 1 score for Mechanism (using UMLS-based concept embeddings with a Bi-LSTM approach), but substantially fall below the results for the other three relation types in comparison with all the systems mentioned in Table 6.…”

“…the seminal descriptive work distinguishing both these sublanguage types by Friedman et al [13]). Newman-Griffis and Fosler-Lussier [14] investigated different sublanguage patterns for the many varieties of clinical reports (pathology reports, discharge summaries, nurse and Intensive Care Unit notes, etc. ), while Nunez and Carenini [15] discussed the portability of embeddings across various fields of medicine reflecting characteristic sublanguage use patterns.…”

“…Differences between the first 13 Xu et al [86] even reach slightly more than 79% F 1 score for Mechanism (using UMLS-based concept embeddings with a Bi-LSTM approach), but substantially fall below the results for the other three relation types in comparison with all the systems mentioned in Table 6. 14 Dewi et al [87] and Sun et al [88] report on 86.3% and 84.5% F 1 scores, respectively, for the overall relation classification task both using a multi-layered CNN archi-and second-ranked systems are typically small, yet become larger on subsequent ranks (roughly between 3 to 4 percentage points relative to the top-ranked system).…”

Automated Coding of Under-Studied Medical Concept Domains: Linking Physical Activity Reports to the International Classification of Functioning, Disability, and Health

“…In addition, there is significant research into strategies for learning neural representations of entities in knowledge bases and coding systems. Past work has investigated diverse approaches, such as leveraging rich semantic information from knowledge base structure and web-scale annotated corpora (34,97,98), utilizing definitions of word senses (similar to our use of ICF definitions) (99,100), and combining terminologies with targeted selection of training corpora to learn applicationtailored concept representations (101,102). While most of the research on entity representations requires resources not yet available for FSI (e.g., large, annotated corpora; well-developed terminologies; robust and interconnected knowledge graph structure), all present significant opportunities to advance FSI coding technologies as more resources are developed.…”

TextEssence: A Tool for Interactive Analysis of Semantic Shifts Between Corpora

“…However, quantitative, vector-based comparison of embedding spaces faces significant conceptual challenges, such as a lack of appropriate alignment objectives and empirical instability (Gonen et al, 2020). While nearest neighbor-based change measurement has been proposed (Newman-Griffis and Fosler-Lussier, 2019;Gonen et al, 2020), its efficacy for small corpora with limited vocabularies remains to be determined. Our novel embedding confidence measure offers a step in this direction (see §6.3 for further discussion), but further research is needed.…”