@inproceedings{yang-etal-2022-knowledge-injected,
title = "Knowledge Injected Prompt Based Fine-tuning for Multi-label Few-shot {ICD} Coding",
author = "Yang, Zhichao and
Wang, Shufan and
Rawat, Bhanu Pratap Singh and
Mitra, Avijit and
Yu, Hong",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.127",
doi = "10.18653/v1/2022.findings-emnlp.127",
pages = "1767--1781",
abstract = "Automatic International Classification of Diseases (ICD) coding aims to assign multiple ICD codes to a medical note with average length of 3,000+ tokens. This task is challenging due to a high-dimensional space of multi-label assignment (tens of thousands of ICD codes) and the long-tail challenge: only a few codes (common diseases) are frequently assigned while most codes (rare diseases) are infrequently assigned. This study addresses the long-tail challenge by adapting a prompt-based fine-tuning technique with label semantics, which has been shown to be effective under few-shot setting. To further enhance the performance in medical domain, we propose a knowledge-enhanced longformer by injecting three domain-specific knowledge: hierarchy, synonym, and abbreviation with additional pretraining using contrastive learning. Experiments on MIMIC-III-full, a benchmark dataset of code assignment, show that our proposed method outperforms previous state-of-the-art method in 14.5{\%} in marco F1 (from 10.3 to 11.8, P{\textless}0.001). To further test our model on few-shot setting, we created a new rare diseases coding dataset, MIMIC-III-rare50, on which our model improves marco F1 from 17.1 to 30.4 and micro F1 from 17.2 to 32.6 compared to previous method.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2022-knowledge-injected">
<titleInfo>
<title>Knowledge Injected Prompt Based Fine-tuning for Multi-label Few-shot ICD Coding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhichao</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shufan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bhanu</namePart>
<namePart type="given">Pratap</namePart>
<namePart type="given">Singh</namePart>
<namePart type="family">Rawat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Avijit</namePart>
<namePart type="family">Mitra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hong</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Automatic International Classification of Diseases (ICD) coding aims to assign multiple ICD codes to a medical note with average length of 3,000+ tokens. This task is challenging due to a high-dimensional space of multi-label assignment (tens of thousands of ICD codes) and the long-tail challenge: only a few codes (common diseases) are frequently assigned while most codes (rare diseases) are infrequently assigned. This study addresses the long-tail challenge by adapting a prompt-based fine-tuning technique with label semantics, which has been shown to be effective under few-shot setting. To further enhance the performance in medical domain, we propose a knowledge-enhanced longformer by injecting three domain-specific knowledge: hierarchy, synonym, and abbreviation with additional pretraining using contrastive learning. Experiments on MIMIC-III-full, a benchmark dataset of code assignment, show that our proposed method outperforms previous state-of-the-art method in 14.5% in marco F1 (from 10.3 to 11.8, P\textless0.001). To further test our model on few-shot setting, we created a new rare diseases coding dataset, MIMIC-III-rare50, on which our model improves marco F1 from 17.1 to 30.4 and micro F1 from 17.2 to 32.6 compared to previous method.</abstract>
<identifier type="citekey">yang-etal-2022-knowledge-injected</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.127</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.127</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>1767</start>
<end>1781</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Knowledge Injected Prompt Based Fine-tuning for Multi-label Few-shot ICD Coding
%A Yang, Zhichao
%A Wang, Shufan
%A Rawat, Bhanu Pratap Singh
%A Mitra, Avijit
%A Yu, Hong
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F yang-etal-2022-knowledge-injected
%X Automatic International Classification of Diseases (ICD) coding aims to assign multiple ICD codes to a medical note with average length of 3,000+ tokens. This task is challenging due to a high-dimensional space of multi-label assignment (tens of thousands of ICD codes) and the long-tail challenge: only a few codes (common diseases) are frequently assigned while most codes (rare diseases) are infrequently assigned. This study addresses the long-tail challenge by adapting a prompt-based fine-tuning technique with label semantics, which has been shown to be effective under few-shot setting. To further enhance the performance in medical domain, we propose a knowledge-enhanced longformer by injecting three domain-specific knowledge: hierarchy, synonym, and abbreviation with additional pretraining using contrastive learning. Experiments on MIMIC-III-full, a benchmark dataset of code assignment, show that our proposed method outperforms previous state-of-the-art method in 14.5% in marco F1 (from 10.3 to 11.8, P\textless0.001). To further test our model on few-shot setting, we created a new rare diseases coding dataset, MIMIC-III-rare50, on which our model improves marco F1 from 17.1 to 30.4 and micro F1 from 17.2 to 32.6 compared to previous method.
%R 10.18653/v1/2022.findings-emnlp.127
%U https://aclanthology.org/2022.findings-emnlp.127
%U https://doi.org/10.18653/v1/2022.findings-emnlp.127
%P 1767-1781
Markdown (Informal)
[Knowledge Injected Prompt Based Fine-tuning for Multi-label Few-shot ICD Coding](https://aclanthology.org/2022.findings-emnlp.127) (Yang et al., Findings 2022)
ACL