Model Establishment of Cross-Disease Course Prediction Using Transfer Learning
In recent years, the development and application of artificial intelligence have both been topics of concern. In the medical field, an important direction of medical technology development is the extraction and use of applicable information from existing medical records to provide more accurate and...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/10/4907 |
| _version_ | 1797501864023949312 |
|---|---|
| author | Josh Jia-Ching Ying Yen-Ting Chang Hsin-Hua Chen Wen-Cheng Chao |
| author_facet | Josh Jia-Ching Ying Yen-Ting Chang Hsin-Hua Chen Wen-Cheng Chao |
| author_sort | Josh Jia-Ching Ying |
| collection | DOAJ |
| description | In recent years, the development and application of artificial intelligence have both been topics of concern. In the medical field, an important direction of medical technology development is the extraction and use of applicable information from existing medical records to provide more accurate and helpful diagnosis suggestions. Therefore, this paper proposes using the development of diseases with easily discernible symptoms to predict the development of other medically related but distinct diseases that lack similar data. The aim of this study is to improve the ease of assessing the development of diseases in which symptoms are difficult to detect, and to improve the utilization of medical data. First, a time series model was used to capture the continuous manifestations of diseases with symptoms that could be easily found at different time intervals. Then, through transfer learning and attention mechanism, the general features captured were applied to the predictive model of the development of diseases with insufficient data and symptoms that are difficult to detect. Finally, we conducted a comprehensive experimental study based on a dataset collected from the National Health Insurance Research Database in Taiwan. The results demonstrate that the effectiveness of our transfer learning approach outperforms state-of-the-art deep learning prediction models for disease course prediction. |
| first_indexed | 2024-03-10T03:24:44Z |
| format | Article |
| id | doaj.art-c390e0bcc26c4fdd9a63f6180a45efbf |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T03:24:44Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-c390e0bcc26c4fdd9a63f6180a45efbf2023-11-23T09:54:56ZengMDPI AGApplied Sciences2076-34172022-05-011210490710.3390/app12104907Model Establishment of Cross-Disease Course Prediction Using Transfer LearningJosh Jia-Ching Ying0Yen-Ting Chang1Hsin-Hua Chen2Wen-Cheng Chao3Department of Management Information Systems, National Chung Hsing University, Taichung 402, TaiwanDepartment of Management Information Systems, National Chung Hsing University, Taichung 402, TaiwanDepartment of Medical Research, Taichung Veterans General Hospital, Taichung 402, TaiwanDepartment of Critical Care Medicine, Taichung Veterans General Hospital, Taichung 402, TaiwanIn recent years, the development and application of artificial intelligence have both been topics of concern. In the medical field, an important direction of medical technology development is the extraction and use of applicable information from existing medical records to provide more accurate and helpful diagnosis suggestions. Therefore, this paper proposes using the development of diseases with easily discernible symptoms to predict the development of other medically related but distinct diseases that lack similar data. The aim of this study is to improve the ease of assessing the development of diseases in which symptoms are difficult to detect, and to improve the utilization of medical data. First, a time series model was used to capture the continuous manifestations of diseases with symptoms that could be easily found at different time intervals. Then, through transfer learning and attention mechanism, the general features captured were applied to the predictive model of the development of diseases with insufficient data and symptoms that are difficult to detect. Finally, we conducted a comprehensive experimental study based on a dataset collected from the National Health Insurance Research Database in Taiwan. The results demonstrate that the effectiveness of our transfer learning approach outperforms state-of-the-art deep learning prediction models for disease course prediction.https://www.mdpi.com/2076-3417/12/10/4907deep learningtime series modelstransfer learningelectronic health records |
| spellingShingle | Josh Jia-Ching Ying Yen-Ting Chang Hsin-Hua Chen Wen-Cheng Chao Model Establishment of Cross-Disease Course Prediction Using Transfer Learning Applied Sciences deep learning time series models transfer learning electronic health records |
| title | Model Establishment of Cross-Disease Course Prediction Using Transfer Learning |
| title_full | Model Establishment of Cross-Disease Course Prediction Using Transfer Learning |
| title_fullStr | Model Establishment of Cross-Disease Course Prediction Using Transfer Learning |
| title_full_unstemmed | Model Establishment of Cross-Disease Course Prediction Using Transfer Learning |
| title_short | Model Establishment of Cross-Disease Course Prediction Using Transfer Learning |
| title_sort | model establishment of cross disease course prediction using transfer learning |
| topic | deep learning time series models transfer learning electronic health records |
| url | https://www.mdpi.com/2076-3417/12/10/4907 |
| work_keys_str_mv | AT joshjiachingying modelestablishmentofcrossdiseasecoursepredictionusingtransferlearning AT yentingchang modelestablishmentofcrossdiseasecoursepredictionusingtransferlearning AT hsinhuachen modelestablishmentofcrossdiseasecoursepredictionusingtransferlearning AT wenchengchao modelestablishmentofcrossdiseasecoursepredictionusingtransferlearning |