MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network
Abstract Background Medication recommendation based on electronic medical record (EMR) is a research hot spot in smart healthcare. For developing computational medication recommendation methods based on EMR, an important challenge is the lack of a large number of longitudinal EMR data with time corr...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-12-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-022-05102-1 |
| _version_ | 1797977228056723456 |
|---|---|
| author | Shaofu Lin Mengzhen Wang Chengyu Shi Zhe Xu Lihong Chen Qingcai Gao Jianhui Chen |
| author_facet | Shaofu Lin Mengzhen Wang Chengyu Shi Zhe Xu Lihong Chen Qingcai Gao Jianhui Chen |
| author_sort | Shaofu Lin |
| collection | DOAJ |
| description | Abstract Background Medication recommendation based on electronic medical record (EMR) is a research hot spot in smart healthcare. For developing computational medication recommendation methods based on EMR, an important challenge is the lack of a large number of longitudinal EMR data with time correlation. Faced with this challenge, this paper proposes a new EMR-based medication recommendation model called MR-KPA, which combines knowledge-enhanced pre-training with the deep adversarial network to improve medication recommendation from both feature representation and the fine-tuning process. Firstly, a knowledge-enhanced pre-training visit model is proposed to realize domain knowledge-based external feature fusion and pre-training-based internal feature mining for improving the feature representation. Secondly, a medication recommendation model based on the deep adversarial network is developed to optimize the fine-tuning process of pre-training visit model and alleviate over-fitting of model caused by the task gap between pre-training and recommendation. Result The experimental results on EMRs from medical and health institutions in Hainan Province, China show that the proposed MR-KPA model can effectively improve the accuracy of medication recommendation on small-scale longitudinal EMR data compared with existing representative methods. Conclusion The advantages of the proposed MR-KPA are mainly attributed to knowledge enhancement based on ontology embedding, the pre-training visit model and adversarial training. Each of these three optimizations is very effective for improving the capability of medication recommendation on small-scale longitudinal EMR data, and the pre-training visit model has the most significant improvement effect. These three optimizations are also complementary, and their integration makes the proposed MR-KPA model achieve the best recommendation effect. |
| first_indexed | 2024-04-11T05:03:36Z |
| format | Article |
| id | doaj.art-e20577f1df124511a190399ced888c6d |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-04-11T05:03:36Z |
| publishDate | 2022-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-e20577f1df124511a190399ced888c6d2022-12-25T12:32:04ZengBMCBMC Bioinformatics1471-21052022-12-0123111910.1186/s12859-022-05102-1MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial networkShaofu Lin0Mengzhen Wang1Chengyu Shi2Zhe Xu3Lihong Chen4Qingcai Gao5Jianhui Chen6Faculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyAbstract Background Medication recommendation based on electronic medical record (EMR) is a research hot spot in smart healthcare. For developing computational medication recommendation methods based on EMR, an important challenge is the lack of a large number of longitudinal EMR data with time correlation. Faced with this challenge, this paper proposes a new EMR-based medication recommendation model called MR-KPA, which combines knowledge-enhanced pre-training with the deep adversarial network to improve medication recommendation from both feature representation and the fine-tuning process. Firstly, a knowledge-enhanced pre-training visit model is proposed to realize domain knowledge-based external feature fusion and pre-training-based internal feature mining for improving the feature representation. Secondly, a medication recommendation model based on the deep adversarial network is developed to optimize the fine-tuning process of pre-training visit model and alleviate over-fitting of model caused by the task gap between pre-training and recommendation. Result The experimental results on EMRs from medical and health institutions in Hainan Province, China show that the proposed MR-KPA model can effectively improve the accuracy of medication recommendation on small-scale longitudinal EMR data compared with existing representative methods. Conclusion The advantages of the proposed MR-KPA are mainly attributed to knowledge enhancement based on ontology embedding, the pre-training visit model and adversarial training. Each of these three optimizations is very effective for improving the capability of medication recommendation on small-scale longitudinal EMR data, and the pre-training visit model has the most significant improvement effect. These three optimizations are also complementary, and their integration makes the proposed MR-KPA model achieve the best recommendation effect.https://doi.org/10.1186/s12859-022-05102-1Medication recommendationElectronic medical recordGraph attention networkPre-training modelAdversarial training |
| spellingShingle | Shaofu Lin Mengzhen Wang Chengyu Shi Zhe Xu Lihong Chen Qingcai Gao Jianhui Chen MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network BMC Bioinformatics Medication recommendation Electronic medical record Graph attention network Pre-training model Adversarial training |
| title | MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network |
| title_full | MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network |
| title_fullStr | MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network |
| title_full_unstemmed | MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network |
| title_short | MR-KPA: medication recommendation by combining knowledge-enhanced pre-training with a deep adversarial network |
| title_sort | mr kpa medication recommendation by combining knowledge enhanced pre training with a deep adversarial network |
| topic | Medication recommendation Electronic medical record Graph attention network Pre-training model Adversarial training |
| url | https://doi.org/10.1186/s12859-022-05102-1 |
| work_keys_str_mv | AT shaofulin mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork AT mengzhenwang mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork AT chengyushi mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork AT zhexu mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork AT lihongchen mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork AT qingcaigao mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork AT jianhuichen mrkpamedicationrecommendationbycombiningknowledgeenhancedpretrainingwithadeepadversarialnetwork |