Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers
As a key modifiable risk factor, alcohol consumption is clinically crucial information that allows medical professionals to further understand their patients’ medical conditions and suggest appropriate lifestyle modifying interventions. However, identifying alcohol-related information fro...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10044673/ |
| _version_ | 1797900210844729344 |
|---|---|
| author | Han Kyul Kim Yujin Park Yeju Park Eunji Choi Sodam Kim Hahyun You Ye Seul Bae |
| author_facet | Han Kyul Kim Yujin Park Yeju Park Eunji Choi Sodam Kim Hahyun You Ye Seul Bae |
| author_sort | Han Kyul Kim |
| collection | DOAJ |
| description | As a key modifiable risk factor, alcohol consumption is clinically crucial information that allows medical professionals to further understand their patients’ medical conditions and suggest appropriate lifestyle modifying interventions. However, identifying alcohol-related information from unstructured free-text clinical notes is often challenging. Not only are the formats of the notes inconsistent, but they also include a massive amount of non-alcohol-related information. Furthermore, for medical institutions outside of English-speaking countries, these clinical notes contain both a mixture of English and local languages, inducing additional difficulty in the extraction. Thanks to the increasing availability of electronic medical record (EMR), several previous works explored the idea of using natural language processing (NLP) to train machine learning models that automatically identify alcohol-related information from unstructured clinical notes. However, all these previous works are limited to English clinical notes, thereby able to leverage various large-scale external ontologies during the text preprocessing. Furthermore, they rely on simple NLP techniques such as the bag-of-words models that suffer from high dimensionality and out-of-vocabulary issues. Addressing these issues, we adopt fine-tuning multilingual transformers. By leveraging their linguistically rich contextual information learned during their pre-training, we are able to extract alcohol-related information from unstructured clinical notes without preprocessing the clinical notes on any external ontologies. Furthermore, our work is the first to explore the use of transformers in bilingual clinical notes to extract alcohol-related information. Even with minimal text preprocessing, we achieve extraction accuracy of 84.70% in terms of macro F-1 score. |
| first_indexed | 2024-04-10T08:42:25Z |
| format | Article |
| id | doaj.art-412d1c65c02e473286742f8f1bb21502 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T08:42:25Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-412d1c65c02e473286742f8f1bb215022023-02-23T00:00:27ZengIEEEIEEE Access2169-35362023-01-0111160661607510.1109/ACCESS.2023.324552310044673Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual TransformersHan Kyul Kim0https://orcid.org/0000-0002-4854-7211Yujin Park1https://orcid.org/0000-0002-7936-9307Yeju Park2Eunji Choi3Sodam Kim4Hahyun You5Ye Seul Bae6https://orcid.org/0000-0003-0763-5458Daniel J. Epstein Department of Industrial and Systems Engineering, University of Southern California, Los Angeles, CA, USADepartment of Biomedical Engineering, Seoul National University College of Medicine, Seoul, South KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul, South KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul, South KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul, South KoreaDepartment of Biomedical Engineering, Seoul National University College of Medicine, Seoul, South KoreaOffice of Hospital Information, Seoul National University Hospital, Seoul, South KoreaAs a key modifiable risk factor, alcohol consumption is clinically crucial information that allows medical professionals to further understand their patients’ medical conditions and suggest appropriate lifestyle modifying interventions. However, identifying alcohol-related information from unstructured free-text clinical notes is often challenging. Not only are the formats of the notes inconsistent, but they also include a massive amount of non-alcohol-related information. Furthermore, for medical institutions outside of English-speaking countries, these clinical notes contain both a mixture of English and local languages, inducing additional difficulty in the extraction. Thanks to the increasing availability of electronic medical record (EMR), several previous works explored the idea of using natural language processing (NLP) to train machine learning models that automatically identify alcohol-related information from unstructured clinical notes. However, all these previous works are limited to English clinical notes, thereby able to leverage various large-scale external ontologies during the text preprocessing. Furthermore, they rely on simple NLP techniques such as the bag-of-words models that suffer from high dimensionality and out-of-vocabulary issues. Addressing these issues, we adopt fine-tuning multilingual transformers. By leveraging their linguistically rich contextual information learned during their pre-training, we are able to extract alcohol-related information from unstructured clinical notes without preprocessing the clinical notes on any external ontologies. Furthermore, our work is the first to explore the use of transformers in bilingual clinical notes to extract alcohol-related information. Even with minimal text preprocessing, we achieve extraction accuracy of 84.70% in terms of macro F-1 score.https://ieeexplore.ieee.org/document/10044673/Clinical informaticsalcohol information extractionnatural language processinginformation extraction from clinical notesmultilingual transformers |
| spellingShingle | Han Kyul Kim Yujin Park Yeju Park Eunji Choi Sodam Kim Hahyun You Ye Seul Bae Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers IEEE Access Clinical informatics alcohol information extraction natural language processing information extraction from clinical notes multilingual transformers |
| title | Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers |
| title_full | Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers |
| title_fullStr | Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers |
| title_full_unstemmed | Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers |
| title_short | Identifying Alcohol-Related Information From Unstructured Bilingual Clinical Notes With Multilingual Transformers |
| title_sort | identifying alcohol related information from unstructured bilingual clinical notes with multilingual transformers |
| topic | Clinical informatics alcohol information extraction natural language processing information extraction from clinical notes multilingual transformers |
| url | https://ieeexplore.ieee.org/document/10044673/ |
| work_keys_str_mv | AT hankyulkim identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers AT yujinpark identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers AT yejupark identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers AT eunjichoi identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers AT sodamkim identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers AT hahyunyou identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers AT yeseulbae identifyingalcoholrelatedinformationfromunstructuredbilingualclinicalnoteswithmultilingualtransformers |