Large-Scale Application of Named Entity Recognition to Biomedicine and Epidemiology

Abstract

Background: Despite significant advancements in biomedical named entity recognition methods, the clinical application of these systems continues to face many challenges: (1) most of the methods are trained on a limited set of clinical entities; (2) these methods are heavily reliant on a large amount of data for both pretraining and prediction, making their use in production impractical; (3) they do not consider non-clinical entities, which are also related to patient's health, such as social, economic or demographic factors. Methods: In this paper, we develop Bio-Epidemiology-NER (https://pypi.org/project/Bio-Epidemiology-NER/) an open-source Python package for detecting biomedical named entities from the text. This approach is based on Transformer-based approach and trained on a dataset that is annotated with many named entities (medical, clinical, biomedical and epidemiological). This approach improves on previous efforts in three ways: (1) it recognizes many clinical entity types, such as medical risk factors, vital signs, drugs, and biological functions; (2) it is easily configurable, reusable and can scale up for training and inference; (3) it also considers non-clinical factors (age and gender, race and social history and so) that influence health outcomes. At a high level, it consists of the phases: preprocessing, data parsing, named entity recognition and named entities enhancement. Results: Experimental results show that our pipeline outperforms other methods on three benchmark datasets with macro-and micro average F1 scores around 90 percent and above.

Proceedings of the 3rd Workshop for Natural Language Processing Open Source Software (NLP-OSS 2023)

“…Fine-tuned, transformer-based encoder models are both SOTA in research 1 and commonly used by developers to solve real-world problems, see e.g. (Raza et al, 2022;. Popular open source frameworks, like the ones provided by HuggingFace , greatly facilitate the use of such models.…”

Improving Clinical Decision Making with a Two-Stage Recommender System: A Case Study on MIMIC-III Dataset

“…RS [2], [42] have been widely studied and applied in various domains, including ecommerce, social networks, and healthcare. In the medical domain, several studies [21], [37], [43], [44], [44], [45] have investigated the development of clinical decision support systems that assist healthcare professionals in making diagnostic or therapeutic decisions. These systems typically rely on rule-based or ML-based models that analyze patient data and provide recommendations based on clinical guidelines or past patient outcomes.…”

“…-NoteEvents: This table contains free-text clinical notes written by healthcare professionals during a patient's hospitalization. We used the BERT-named entity recognition model[21] to extract relevant information, such as medical history, diagnoses, and treatment plans. -Labevents: This table includes laboratory measurements.…”

nerblackbox: A High-level Library for Named Entity Recognition in Python

“…Fine-tuned, transformer-based encoder models are both SOTA in research 1 and commonly used by developers to solve real-world problems, see e.g. (Raza et al, 2022;. Popular open source frameworks, like the ones provided by HuggingFace (Wolf et al, 2020;Lhoest et al, 2021;Von Werra et al, 2022), greatly facilitate the use of such models.…”