A pediatric emergency prediction model using natural language process in the pediatric emergency department
Abstract This study developed a predictive model using deep learning (DL) and natural language processing (NLP) to identify emergency cases in pediatric emergency departments. It analyzed 87,759 pediatric cases from a South Korean tertiary hospital (2012–2021) using electronic medical records. Vario...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-87161-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571856970842112 |
|---|---|
| author | Arum Choi Chohee Kim Jisu Ryoo Jangyeong Jeon Sangyeon Cho Dongjoon Lee Junyeong Kim Changhee Lee Woori Bae |
| author_facet | Arum Choi Chohee Kim Jisu Ryoo Jangyeong Jeon Sangyeon Cho Dongjoon Lee Junyeong Kim Changhee Lee Woori Bae |
| author_sort | Arum Choi |
| collection | DOAJ |
| description | Abstract This study developed a predictive model using deep learning (DL) and natural language processing (NLP) to identify emergency cases in pediatric emergency departments. It analyzed 87,759 pediatric cases from a South Korean tertiary hospital (2012–2021) using electronic medical records. Various NLP models, including four machine learning (ML) models with Term Frequency-Inverse Document Frequency (TF-IDF) and two DL models based on the KM-BERT framework, were trained to differentiate emergency cases using clinician transcripts. Gradient Boosting, among the ML models, performed best with an AUROC of 0.715, AUPRC of 0.778, and F1-score of 0.677. DL models, especially the fine-tuned KM-BERT model, showed superior performance, achieving an AUROC of 0.839, AUPRC of 0.879, and F1-score of 0.773. Shapley-based explanations provided insights into model predictions, underlining the potential of these technologies in medical decision-making. This study demonstrates the potential of advanced DL techniques for NLP in emergency medical settings, offering a more precise and efficient approach to managing healthcare resources and improving patient outcomes. |
| format | Article |
| id | doaj-art-5e1add1368c048b39eb2ff15172c1b44 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-5e1add1368c048b39eb2ff15172c1b442025-02-02T12:16:15ZengNature PortfolioScientific Reports2045-23222025-01-0115111110.1038/s41598-025-87161-xA pediatric emergency prediction model using natural language process in the pediatric emergency departmentArum Choi0Chohee Kim1Jisu Ryoo2Jangyeong Jeon3Sangyeon Cho4Dongjoon Lee5Junyeong Kim6Changhee Lee7Woori Bae8Department of Radiology, College of Medicine, The Catholic University of KoreaVUNODepartment of Emergency Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of KoreaDepartment of Artificial Intelligence, Chung-Ang UniversityDepartment of Artificial Intelligence, Chung-Ang UniversityDepartment of Artificial Intelligence, Chung-Ang UniversityDepartment of Artificial Intelligence, Chung-Ang UniversityDepartment of Artificial Intelligence, Korea UniversityDepartment of Emergency Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of KoreaAbstract This study developed a predictive model using deep learning (DL) and natural language processing (NLP) to identify emergency cases in pediatric emergency departments. It analyzed 87,759 pediatric cases from a South Korean tertiary hospital (2012–2021) using electronic medical records. Various NLP models, including four machine learning (ML) models with Term Frequency-Inverse Document Frequency (TF-IDF) and two DL models based on the KM-BERT framework, were trained to differentiate emergency cases using clinician transcripts. Gradient Boosting, among the ML models, performed best with an AUROC of 0.715, AUPRC of 0.778, and F1-score of 0.677. DL models, especially the fine-tuned KM-BERT model, showed superior performance, achieving an AUROC of 0.839, AUPRC of 0.879, and F1-score of 0.773. Shapley-based explanations provided insights into model predictions, underlining the potential of these technologies in medical decision-making. This study demonstrates the potential of advanced DL techniques for NLP in emergency medical settings, offering a more precise and efficient approach to managing healthcare resources and improving patient outcomes.https://doi.org/10.1038/s41598-025-87161-xPediatric emergency departmentEmergency room visitsPrediction modelNatural language processLanguage model |
| spellingShingle | Arum Choi Chohee Kim Jisu Ryoo Jangyeong Jeon Sangyeon Cho Dongjoon Lee Junyeong Kim Changhee Lee Woori Bae A pediatric emergency prediction model using natural language process in the pediatric emergency department Scientific Reports Pediatric emergency department Emergency room visits Prediction model Natural language process Language model |
| title | A pediatric emergency prediction model using natural language process in the pediatric emergency department |
| title_full | A pediatric emergency prediction model using natural language process in the pediatric emergency department |
| title_fullStr | A pediatric emergency prediction model using natural language process in the pediatric emergency department |
| title_full_unstemmed | A pediatric emergency prediction model using natural language process in the pediatric emergency department |
| title_short | A pediatric emergency prediction model using natural language process in the pediatric emergency department |
| title_sort | pediatric emergency prediction model using natural language process in the pediatric emergency department |
| topic | Pediatric emergency department Emergency room visits Prediction model Natural language process Language model |
| url | https://doi.org/10.1038/s41598-025-87161-x |
| work_keys_str_mv | AT arumchoi apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT choheekim apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT jisuryoo apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT jangyeongjeon apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT sangyeoncho apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT dongjoonlee apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT junyeongkim apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT changheelee apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT wooribae apediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT arumchoi pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT choheekim pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT jisuryoo pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT jangyeongjeon pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT sangyeoncho pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT dongjoonlee pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT junyeongkim pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT changheelee pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment AT wooribae pediatricemergencypredictionmodelusingnaturallanguageprocessinthepediatricemergencydepartment |