医学自然语言处理(NLP)相关论文汇总之 EMNLP 2020

【写在前面】最近一段时间在调研医疗方向的NLP论文，所以对于各个会议上的论文也顺手进行了分类整理，放在这里，希望对大家有一定的帮助吧。等到EMNLP2021、ACL2021等相关论文放榜后，还会继续更新。最近在阅读一篇文本分类综述，心生了一个复现里面代码和整理数据集的想法，目前正在稳步推荐，感兴趣的同学可以参考我的文本分类GitHub链接。汇总文本分类方法的目的是为了让大家更容易地在自己的数据集上跑baseline模型。

更多关于中文医疗自然语言处理的资源和论文汇总，请访问我的GitHub相关主页https://github.com/LovelyDayWang/Chinese_Medical_Natural_Language_Processing_Resources_and_Papers。
文本分类baseline仓库链接：https://github.com/LovelyDayWang/Summary-of-Text-Classification-Models-PyTorch，欢迎star～

机器翻译

机器翻译方法的评价应用于医学名词术语

Evaluation of Machine Translation Methods applied to Medical Terminologies

论文地址：https://www.aclweb.org/anthology/2020.louhi-1.7/

一个多语种的神经机器翻译模型的生物医学数据

A Multilingual Neural Machine Translation Model for Biomedical Data

论文地址：https://www.aclweb.org/anthology/2020.nlpcovid19-2.16/

WMT 2020生物医学翻译共同任务的发现：巴斯克语，意大利语和俄语作为新的附加语言

Findings of the WMT 2020 Biomedical Translation Shared Task: Basque, Italian and Russian as New Additional Languages

论文地址：https://www.aclweb.org/anthology/2020.wmt-1.76/

Elhuyar提交了有关术语和摘要翻译的2020年生物医学翻译任务

Elhuyar submission to the Biomedical Translation Task 2020 on terminology and abstracts translation

论文地址：https://www.aclweb.org/anthology/2020.wmt-1.87/

巴斯克生物医学神经机器翻译的预训练语言模型和回译

Pretrained Language Models and Backtranslation for English-Basque Biomedical Neural Machine Translation

论文地址：https://www.aclweb.org/anthology/2020.wmt-1.89/

机器阅读理解

具有结构知识和纯文本的owards医疗机器阅读理解

Towards Medical Machine Reading Comprehension with Structural Knowledge and Plain Text

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.111/

实体规范化

知识驱动的多涵义中医程序实体规范化生成模型

A Knowledge-driven Generative Model for Multi-implication Chinese Medical Procedure Entity Normalization

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.116/

用户嘈杂文本中的目标概念指导医学概念归一化

Target Concept Guided Medical Concept Normalization in Noisy User-Generated Texts

论文地址：https://www.aclweb.org/anthology/2020.deelio-1.8/

通过学习目标概念嵌入来在用户生成的文本中对医学概念进行规范化

Medical Concept Normalization in User-Generated Texts by Learning Target Concept Embeddings

论文地址：https://www.aclweb.org/anthology/2020.louhi-1.3/

命名实体识别

评估DistilBERT在命名实体识别任务上的性能，以检测受保护的健康信息和医疗概念

Assessment of DistilBERT performance on Named Entity Recognition task for the detection of Protected Health Information and medical concepts

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.18/

关系抽取

FedED：通过整体蒸馏进行联合学习以进行医学关系提取

FedED: Federated Learning via Ensemble Distillation for Medical Relation Extraction

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.165/

实体链接

**COMETA: A Corpus for Medical Entity Linking in the Social Media**

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.253/

生物医学文本的简单分层多任务神经网络端到端实体链接

Simple Hierarchical Multi-Task Neural End-To-End Entity Linking for Biomedical Text

论文地址：https://www.aclweb.org/anthology/2020.louhi-1.2/

语言模型

BioMegatron：更大的生物医学领域语言模型

BioMegatron: Larger Biomedical Domain Language Model

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.379/

用于生物医学和临床任务的预训练语言模型：理解和扩展最新技术

Pretrained Language Models for Biomedical and Clinical Tasks: Understanding and Extending the State-of-the-Art

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.17/

廉价的预适应语言模型的领域适应：生物医学NER和Covid-19 QA的案例研究

Inexpensive Domain Adaptation of Pretrained Language Models: Case Studies on Biomedical NER and Covid-19 QA

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.134/

关于小型，经过区分训练的语言表示模型对生物医学文本挖掘的有效性

On the effectiveness of small, discriminatively pre-trained language representation models for biomedical text mining

论文地址：https://www.aclweb.org/anthology/2020.sdp-1.12/

事件抽取

生物医学事件提取作为序列标记

Biomedical Event Extraction as Sequence Labeling

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.431/

带有层次知识图的生物医学事件提取

Biomedical Event Extraction with Hierarchical Knowledge Graphs

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.114/

数据集

COMETA：社交媒体中用于医疗实体链接的语料库

COMETA: A Corpus for Medical Entity Linking in the Social Media

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.253/

大型医学对话数据集

MedDialog: Large-scale Medical Dialogue Datasets

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.743/

MeDAL：用于自然语言理解预训练的医学缩写歧义消除数据集

MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretraining

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.15/

MedICaT：医学图像，标题和文字参考的数据集

MedICaT: A Dataset of Medical Images, Captions, and Textual References

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.191/

GGPONC：基于临床实践指南的具有丰富元数据的德国医学文本语料库

GGPONC: A Corpus of German Medical Text with Rich Metadata Based on Clinical Practice Guidelines

论文地址：https://www.aclweb.org/anthology/2020.louhi-1.5/

基于国外临床医学数据的NLP研究

使用Sig-Transformer编码器从瑞典医疗处方中提取信息

Information Extraction from Swedish Medical Prescriptions with Sig-Transformer Encoder

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.5/

挪威病历中晕厥病例的分类

Classification of Syncope Cases in Norwegian Medical Records

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.9/

对话

从医学对话中弱监督药物治疗方案的提取

Weakly Supervised Medication Regimen Extraction from Medical Conversations

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.20/

Summarize博士：利用局部结构对医学对话进行全球总结。

Dr. Summarize: Global Summarization of Medical Dialogue by Exploiting Local Structures.

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.335/

文本生成

强化学习与不平衡数据集的数据到文本医学报告生成

Reinforcement Learning with Imbalanced Dataset for Data-to-Text Medical Report Generation

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.202/

从医学图生成准确的电子健康评估？？推理还是文本生成？

Generating Accurate Electronic Health Assessment from Medical Graph

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.336/

问答

生物医学事件提取作为多回合问答

Biomedical Event Extraction as Multi-turn Question Answering

论文地址：https://www.aclweb.org/anthology/2020.louhi-1.10/

主题模型

为COVID-19医学研究文献开发策展主题模型

Developing a Curated Topic Model for COVID-19 Medical Research Literature

论文地址：https://www.aclweb.org/anthology/2020.nlpcovid19-2.30/

表示学习

**ERLKG: Entity Representation Learning and Knowledge Graph based association analysis of COVID-19 through mining of unstructured biomedical corpora**

论文地址：https://www.aclweb.org/anthology/2020.sdp-1.15/

用潜在变量模型学习生物医学关系的信息表示

Learning Informative Representations of Biomedical Relations with Latent Variable Models

论文地址：https://www.aclweb.org/anthology/2020.sustainlp-1.3/

Others

扩展卷积注意网络，用于从临床文本分配医学代码

Dilated Convolutional Attention Network for Medical Code Assignment from Clinical Text

论文地址：https://www.aclweb.org/anthology/2020.clinicalnlp-1.8/

图卷积网络和关注问题的双重注意总结中国医学答案

Summarizing Chinese Medical Answer with Graph Convolution Networks and Question-focused Dual Attention

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.2/

序列跨度分类与生物医学摘要的神经半马尔可夫CRFs

Sequential Span Classification with Neural Semi-Markov CRFs for Biomedical Abstracts

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.77/

表征医学笔记中信息的价值

Characterizing the Value of Information in Medical Notes

论文地址：https://www.aclweb.org/anthology/2020.findings-emnlp.187/

在新闻中查询各种类型的医疗索赔

Querying Across Genres for Medical Claims in News

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.139/

医学文本中时间事件的有效表示

An efficient representation of chronological events in medical texts

论文地址：https://www.aclweb.org/anthology/2020.louhi-1.11/

将疾病知识注入BERT以进行健康问答，医学推断和疾病名称识别

Infusing Disease Knowledge into BERT for Health Question Answering, Medical Inference and Disease Name Recognition

论文地址：https://www.aclweb.org/anthology/2020.emnlp-main.372/