A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search
Abstract Patient similarity search is an essential task in healthcare. Recent studies adopted electronic health records (EHRs) to learn patient representations for measuring the clinical similarities. These methods outperformed traditional methods, by capturing more information from various sources...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-08-01
|
| Series: | Data Science and Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s41019-023-00216-9 |
| _version_ | 1797675130072072192 |
|---|---|
| author | Yefan Huang Feng Luo Xiaoli Wang Zhu Di Bohan Li Bin Luo |
| author_facet | Yefan Huang Feng Luo Xiaoli Wang Zhu Di Bohan Li Bin Luo |
| author_sort | Yefan Huang |
| collection | DOAJ |
| description | Abstract Patient similarity search is an essential task in healthcare. Recent studies adopted electronic health records (EHRs) to learn patient representations for measuring the clinical similarities. These methods outperformed traditional methods, by capturing more information from various sources consisting of multi-modal EHRs, external knowledge and correlations among medical concepts. They often concerned certain type of data without taking full advantage of various information. We propose a graph representation learning framework, denoted by One-Size-Fits-Three (OSFT), that takes into account fusion-attention, neighbor-attention and global-attention from three types of information. Extensive experiments are conducted on two real datasets of MIMIC-III and MIMIC-IV, and the results verified the effectiveness and generality of our framework. When compared with baselines on patient similarity search, our framework achieved good effectiveness and comparative efficiency. The results provide new insights about whether the use of various information can better measure the patient similarity. The source codes are available at https://github.com/emmali808/ADDS/tree/master/EHRDeepHelper . |
| first_indexed | 2024-03-11T22:10:14Z |
| format | Article |
| id | doaj.art-c6d12be54eeb4bee939146c20a5adde8 |
| institution | Directory Open Access Journal |
| issn | 2364-1185 2364-1541 |
| language | English |
| last_indexed | 2024-03-11T22:10:14Z |
| publishDate | 2023-08-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Data Science and Engineering |
| spelling | doaj.art-c6d12be54eeb4bee939146c20a5adde82023-09-24T11:27:28ZengSpringerOpenData Science and Engineering2364-11852364-15412023-08-018330631710.1007/s41019-023-00216-9A One-Size-Fits-Three Representation Learning Framework for Patient Similarity SearchYefan Huang0Feng Luo1Xiaoli Wang2Zhu Di3Bohan Li4Bin Luo5School of Informatics, Xiamen UniversityByteDance Inc.School of Informatics, Xiamen UniversityCollege of Computer Science and Technology, Nanjing University of Aeronautics and AstronauticsCollege of Computer Science and Technology, Nanjing University of Aeronautics and AstronauticsSichuan Huhui Software Co. Ltd.Abstract Patient similarity search is an essential task in healthcare. Recent studies adopted electronic health records (EHRs) to learn patient representations for measuring the clinical similarities. These methods outperformed traditional methods, by capturing more information from various sources consisting of multi-modal EHRs, external knowledge and correlations among medical concepts. They often concerned certain type of data without taking full advantage of various information. We propose a graph representation learning framework, denoted by One-Size-Fits-Three (OSFT), that takes into account fusion-attention, neighbor-attention and global-attention from three types of information. Extensive experiments are conducted on two real datasets of MIMIC-III and MIMIC-IV, and the results verified the effectiveness and generality of our framework. When compared with baselines on patient similarity search, our framework achieved good effectiveness and comparative efficiency. The results provide new insights about whether the use of various information can better measure the patient similarity. The source codes are available at https://github.com/emmali808/ADDS/tree/master/EHRDeepHelper .https://doi.org/10.1007/s41019-023-00216-9Patient similarity searchMulti-modal EHRsMedical conceptsExternal knowledgeGraph representation learning |
| spellingShingle | Yefan Huang Feng Luo Xiaoli Wang Zhu Di Bohan Li Bin Luo A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search Data Science and Engineering Patient similarity search Multi-modal EHRs Medical concepts External knowledge Graph representation learning |
| title | A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search |
| title_full | A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search |
| title_fullStr | A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search |
| title_full_unstemmed | A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search |
| title_short | A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search |
| title_sort | one size fits three representation learning framework for patient similarity search |
| topic | Patient similarity search Multi-modal EHRs Medical concepts External knowledge Graph representation learning |
| url | https://doi.org/10.1007/s41019-023-00216-9 |
| work_keys_str_mv | AT yefanhuang aonesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT fengluo aonesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT xiaoliwang aonesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT zhudi aonesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT bohanli aonesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT binluo aonesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT yefanhuang onesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT fengluo onesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT xiaoliwang onesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT zhudi onesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT bohanli onesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch AT binluo onesizefitsthreerepresentationlearningframeworkforpatientsimilaritysearch |