@inproceedings{alsaidi-etal-2025-comparing,
title = "Comparing representations of long clinical texts for the task of patient-note identification",
author = "Alsaidi, Safa and
Vincent, Marc and
Boyer, Olivia and
Garcelon, Nicolas and
Couceiro, Miguel and
Coulet, Adrien",
editor = "Ananiadou, Sophia and
Demner-Fushman, Dina and
Gupta, Deepak and
Thompson, Paul",
booktitle = "Proceedings of the Second Workshop on Patient-Oriented Language Processing (CL4Health)",
month = may,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.cl4health-1.9/",
doi = "10.18653/v1/2025.cl4health-1.9",
pages = "109--123",
ISBN = "979-8-89176-238-1",
abstract = "In this paper, we address the challenge of patient-note identification, which involves accurately matching an anonymized clinical note to its corresponding patient, represented by a set of related notes. This task has broad applications, including duplicate records detection and patient similarity analysis, which require robust patient-level representations. We explore various embedding methods, including Hierarchical Attention Networks (HAN), three-level Hierarchical Transformer Networks (HTN), LongFormer, and advanced BERT-based models, focusing on their ability to process medium-to-long clinical texts effectively. Additionally, we evaluate different pooling strategies (mean, max, and mean{\_}max) for aggregating word-level embeddings into patient-level representations and we examine the impact of sliding windows on model performance. Our results indicate that BERT-based embeddings outperform traditional and hierarchical models, particularly in processing lengthy clinical notes and capturing nuanced patient representations. Among the pooling strategies, mean{\_}max pooling consistently yields the best results, highlighting its ability to capture critical features from clinical notes. Furthermore, the reproduction of our results on both MIMIC dataset and Necker hospital data warehouse illustrates the generalizability of these approaches to real-world applications, emphasizing the importance of both embedding methods and aggregation strategies in optimizing patient-note identification and enhancing patient-level modeling."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="alsaidi-etal-2025-comparing">
<titleInfo>
<title>Comparing representations of long clinical texts for the task of patient-note identification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Safa</namePart>
<namePart type="family">Alsaidi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marc</namePart>
<namePart type="family">Vincent</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Olivia</namePart>
<namePart type="family">Boyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicolas</namePart>
<namePart type="family">Garcelon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Miguel</namePart>
<namePart type="family">Couceiro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adrien</namePart>
<namePart type="family">Coulet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Second Workshop on Patient-Oriented Language Processing (CL4Health)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sophia</namePart>
<namePart type="family">Ananiadou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dina</namePart>
<namePart type="family">Demner-Fushman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Deepak</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">Thompson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-238-1</identifier>
</relatedItem>
<abstract>In this paper, we address the challenge of patient-note identification, which involves accurately matching an anonymized clinical note to its corresponding patient, represented by a set of related notes. This task has broad applications, including duplicate records detection and patient similarity analysis, which require robust patient-level representations. We explore various embedding methods, including Hierarchical Attention Networks (HAN), three-level Hierarchical Transformer Networks (HTN), LongFormer, and advanced BERT-based models, focusing on their ability to process medium-to-long clinical texts effectively. Additionally, we evaluate different pooling strategies (mean, max, and mean_max) for aggregating word-level embeddings into patient-level representations and we examine the impact of sliding windows on model performance. Our results indicate that BERT-based embeddings outperform traditional and hierarchical models, particularly in processing lengthy clinical notes and capturing nuanced patient representations. Among the pooling strategies, mean_max pooling consistently yields the best results, highlighting its ability to capture critical features from clinical notes. Furthermore, the reproduction of our results on both MIMIC dataset and Necker hospital data warehouse illustrates the generalizability of these approaches to real-world applications, emphasizing the importance of both embedding methods and aggregation strategies in optimizing patient-note identification and enhancing patient-level modeling.</abstract>
<identifier type="citekey">alsaidi-etal-2025-comparing</identifier>
<identifier type="doi">10.18653/v1/2025.cl4health-1.9</identifier>
<location>
<url>https://aclanthology.org/2025.cl4health-1.9/</url>
</location>
<part>
<date>2025-05</date>
<extent unit="page">
<start>109</start>
<end>123</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Comparing representations of long clinical texts for the task of patient-note identification
%A Alsaidi, Safa
%A Vincent, Marc
%A Boyer, Olivia
%A Garcelon, Nicolas
%A Couceiro, Miguel
%A Coulet, Adrien
%Y Ananiadou, Sophia
%Y Demner-Fushman, Dina
%Y Gupta, Deepak
%Y Thompson, Paul
%S Proceedings of the Second Workshop on Patient-Oriented Language Processing (CL4Health)
%D 2025
%8 May
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-238-1
%F alsaidi-etal-2025-comparing
%X In this paper, we address the challenge of patient-note identification, which involves accurately matching an anonymized clinical note to its corresponding patient, represented by a set of related notes. This task has broad applications, including duplicate records detection and patient similarity analysis, which require robust patient-level representations. We explore various embedding methods, including Hierarchical Attention Networks (HAN), three-level Hierarchical Transformer Networks (HTN), LongFormer, and advanced BERT-based models, focusing on their ability to process medium-to-long clinical texts effectively. Additionally, we evaluate different pooling strategies (mean, max, and mean_max) for aggregating word-level embeddings into patient-level representations and we examine the impact of sliding windows on model performance. Our results indicate that BERT-based embeddings outperform traditional and hierarchical models, particularly in processing lengthy clinical notes and capturing nuanced patient representations. Among the pooling strategies, mean_max pooling consistently yields the best results, highlighting its ability to capture critical features from clinical notes. Furthermore, the reproduction of our results on both MIMIC dataset and Necker hospital data warehouse illustrates the generalizability of these approaches to real-world applications, emphasizing the importance of both embedding methods and aggregation strategies in optimizing patient-note identification and enhancing patient-level modeling.
%R 10.18653/v1/2025.cl4health-1.9
%U https://aclanthology.org/2025.cl4health-1.9/
%U https://doi.org/10.18653/v1/2025.cl4health-1.9
%P 109-123
Markdown (Informal)
[Comparing representations of long clinical texts for the task of patient-note identification](https://aclanthology.org/2025.cl4health-1.9/) (Alsaidi et al., CL4Health 2025)
ACL