
浏览全部资源
扫码关注微信
1.季华实验室智能机器人工程研究中心(广东 佛山 528200)
2.上海中医药大学附属岳阳中西医结合医院(上海 200437)
3.广州中医药大学基础医学院(广东 广州 510006)
4.复旦大学工程与应用技术研究院(上海 200433)
苗春龙,男,在读博士生,主要从事基于图计算的中医临床决策支持研究
沈融,副主任医师,硕士生导师;E-mail:shenrong@shyueyanghospital.com
罗静静,研究员,硕士生导师;E-mail: luojingjing@fudan.edu.cn
收稿日期:2023-06-01,
修回日期:2023-08-01,
纸质出版日期:2024-11-25
移动端阅览
苗春龙,沈融,兰杰等.基于软阈值领域聚合图网络的《伤寒论》方剂推荐方法[J].上海中医药大学学报,2024,38(06):8-18.
MIAO Chunlong,SHEN Rong,LAN Jie,et al.Prescription recommendation method of Treatise on Febrile Diseases based on soft threshold domain aggregation graph network[J].Academic Journal of Shanghai University of Traditional Chinese Medicine,2024,38(06):8-18.
苗春龙,沈融,兰杰等.基于软阈值领域聚合图网络的《伤寒论》方剂推荐方法[J].上海中医药大学学报,2024,38(06):8-18. DOI: 10.16306/j.1008-861x.2024.06.002.
MIAO Chunlong,SHEN Rong,LAN Jie,et al.Prescription recommendation method of Treatise on Febrile Diseases based on soft threshold domain aggregation graph network[J].Academic Journal of Shanghai University of Traditional Chinese Medicine,2024,38(06):8-18. DOI: 10.16306/j.1008-861x.2024.06.002.
目的
2
针对《伤寒论》辅助诊断和决策支持缺乏等问题,以《伤寒论》名家医案为基础,提出一种基于深度学习的《伤寒论》医案数据建模方法和《伤寒论》方剂推荐模型。
方法
2
本研究采用基于Levenshtein距离的同义词识别算法对医案资料中的症状描述进行聚类,减小症状数据集维度;以“患者”-“症状”为本体构建《伤寒论》异质图数据集,并采用Sentence-BERT模型获得具有深层语义的节点嵌入向量;提出融合自适应软阈值滤波的领域聚合ST-GraphSAGE图卷积深度学习模型,并将其应用于《伤寒论》方剂推荐任务。
结果
2
最终将非结构化的《伤寒论》医案建模为图结构数据集,并在该数据集上进行了实验,在F1-Score等4项评价指标下,提出的ST-GraphSAGE取得了优于机器学习及其他主流图深度学习模型的成绩,方剂推荐准确率可达76%,且相较于基础模型,ST-GraphSAGE在处理含噪数据时准确率提高0.51%~3.92%。
结论
2
提出《伤寒论》医案建模方案及ST-GraphSAGE图深度学习模型,其方剂推荐准确率较高且泛化能力强,能充分利用《伤寒论》医案的个体及关系信息,为中医临床的决策支持提供新的研究思路。
Objective
2
In view of the problems such as the lack of auxiliary diagnosis and decision support in the field of Treatise on Febrile Diseases (TFD), this paper proposes a data modeling scheme for TFD medical records and a TFD prescription recommendation model based on deep learning per the medical records of famous experts in TFD.
Methods
2
In this study, the synonym recognition algorithm based on Levenshtein distance is used to cluster the symptom descriptions in TFD medical records to reduce the dimension of the symptom data set. The "patient"-"symptom" is used as the ontology to construct a heterogeneous graph data set of TFD and Sentence-BERT model was adopted to obtain node embedding vectors with deep semantics. A domain aggregation ST-GraphSAGE deep learning model fused with adaptive soft threshold filtering was proposed and applied to the TFD prescription recommendation task.
Results
2
Unstructured TFD medical records were modeled as a graph-structured data set, and experiments were conducted on this data set. Through four evaluation indicators such as F1-Score, the performance of ST-GraphSAGE proposed in this paper achieved better results than machine learning and other mainstream graph deep learning models. The accuracy rate of prescription recommendation reached 76%, and compared to the basic model, the accuracy rate was increased by 0.51%~3.90% when ST-GraphSAGE was used to deal with noisy data.
Conclusion
2
The TFD medical records modeling scheme and ST-GraphSAGE graph deep learning model proposed in this paper have higher accuracy rate in prescription recommendation and strong generalization ability, which can make full use of the individual and relational information of TFD medical records to provide new research ideas for decision support in clinical practice of traditional Chinese medicine.
胡晓娟 , 洪燕龙 , 张磊 , 等 . 一种经方智能推荐方法及应用该方法的设备 [P]. 中国 : CN112233804A , 2021-01-15 .
HU X J , HONG Y L , ZHANG L , et al . A Method for Intelligent Recommendation of Classic Prescriptions and Equipment for Applying the Method [P]. China : CN112233804A , 2021-01-15 .
汤尔群 , 任廷革 , 陈明 , 等 . 基于数据挖掘方法的《伤寒论》方证知识挖掘研究 [J]. 中国中医药信息杂志 , 2012 , 19 ( 4 ): 31 - 34 .
TANG E Q , REN T G , CHEN M , et al . Knowledge Mining of Prescriptions and Syndromes Knowledge in Treatise on Febrile Dise-ases Based on Data Mining Method [J]. Chinese Journal of Information on Traditional Chinese Medicine , 2012 , 19 ( 4 ): 31 - 34 .
洪燕龙 , 张林 , 胡晓娟 , 等 . 互联网与人工智能背景下的中医药诊疗服务新模式设想与研究 [J]. 上海中医药大学学报 , 2020 , 34 ( 6 ): 1 - 8 .
HONG Y L , ZHANG L , HU X J , et al . Conception and research on new model of traditional Chinese medicine diagnosis and treatment service under the background of internet and artificial intelligence [J]. Academic Journal of Shanghai University of Traditional Chinese Medicine , 2020 , 34 ( 6 ): 1 - 8 .
LI W , YANG Z , SUN X . Exploration on Generating Traditional Chinese Medicine Prescription from Symptoms with an End-to-End Method [C]// Natural Language Processing and Chinese Computing . Proceedings of the 8th Chinese Conference, Dunhuang, China . Berlin : Springer International Publishing , 2019 : 486 - 498 .
许帆 . 基于临床数据的中医处方推荐方法研究 [D]. 北京 : 北京交通大学 , 2019 : 27 - 31 .
XU F . Study on TCM Prescription Recommendation Based on Clinical Data [D]. Beijing : Beijing Jiaotong University , 2019 : 27 - 31 .
秦尧 . 基于机器学习的中医方剂推荐方法的研究 [D]. 北京 : 北京工业大学 , 2020 : 42 - 56 .
QIN Y . Research on Recommendation Method of Traditional Chinese Medicine Prescription Based on Machine Learning [D]. Beijing : Beijing University of Technology , 2020 : 42 - 56 .
梁恒肇 . 中医处方辅助推荐系统的设计与实现 [D]. 成都 : 电子科技大学 , 2018 : 39 - 49 .
LIANG H Z . Research and Application of Auxiliary TCM Prescription Recommendation System [D]. Chengdu : University of Electronic Science and Technology of China , 2018 : 39 - 49 .
KIPF T N , WELLING M . Semi-supervised classification with graph convolutional networks [C]// International Conference on Learning Representations (ICLR 2017). https:// OpenReview.net , 2017
BRESSON X , LAURENT T . Residual Gated Graph ConvNets [C]// 6th International Conference on Learning Representations (ICLR 2018 . Vancouver, Canada, 2018 .
VELIČKOVIĆ P , CUCURULL G , CASANOVA A , et al . Graph attention networks [C]// International Conference on Learning Representations (ICLR 2018 . Vancouver, Canada, 2018 .
BRODY S , ALON U , YAHAV E . How attentive are graph attention networks? [C]// International Conference on Learning Representations (ICLR 2021 ), https:// OpenReview.net , 2021 .
SHI Y , HUANG Z , FENG S , et al . Masked label prediction: Unified message passing model for semi-supervised classification [EB/OL]. ( 2020 . 10 . 05 ) [ 2023-01-10 ]. https://arxiv.org/abs/2009.03509 https://arxiv.org/abs/2009.03509 .
DUVENAUD D K , MACLAURIN D , IPARRAGUIRRE J , et al . Convolutional Networks on Graphs for Learning Molecular Fingerprints [C]// Advances in Neural Information Processing Systems (NIPS) , Montréal, Canada , 2015 . Red Hook, Newyork : Curran Associates, Inc. , 2015: 2215 - 2223 .
BROCKSCHMIDT M . GNN-Film: Graph Neural Networks with Feature-Wise Linear Modulation [C]// Proceedings of the 37th International Conference on Machine Learning (ICML) . Vienna, Austria , 2020 . PMLR, 2020 : 1144 - 1152 .
WANG X , JI H Y , SHI C , et al . Heterogeneous Graph Attention Network [C]// The World Wide Web Conference (WWW). San Francisco, California : Association for Computing Machinery , 2019 : 2022 - 2032 .
HAMILTON W , LESKOVEC J . Inductive Representation Learning on Large Graphs [C]// Advances in Neural Information Processing Systems (NeurIPS) , Long Beach, USA , 2017 : 1024 - 1034 .
俞飞蝶 . 基于图嵌入和强化学习的诊疗方案推荐方法研究 [D]. 北京 : 北京交通大学 , 2021 .
YU F D . Research on Recommendation Method of Treatment Regime based onGraph Embedding and Reinforcement Learning [D]. Beijing : BeijingJiaotong University , 2021 .
李升圆 . 融合个体特征的中医药方推荐技术 [D]. 上海 : 华东师范大学 , 2022 : 24 - 42 .
LI S Y . Traditional Chinese Medicine Prescription Recommendation Technology Using Individual Characteristics [D]. Shanghai : East China Normal University , 2022 : 24 - 42 .
RUAN C Y , MA J G , WANG Y , et al . Discovering Regularities from Traditional Chinese Medicine Prescriptions via Bipartite Embedding Model [C]// International Joint Conference on Artificial Intelligence (IJCAI) . San Francisco, CA : Morgan Kaufmann , 2019 : 3346 - 3352 .
RUAN C Y , WANG Y , ZHANG Y C , et al . Exploring Regularity in Traditional Chinese Medicine Clinical Data Using Heterogeneous Weighted Networks Embedding [C]// International Conference on Database Systems for Advanced Applications (DASFAA) . Berlin : Springer-Cham , 2019 : 310 - 313 .
靳远远 . 基于图神经网络的中医药方推荐技术研究 [D]. 上海 : 华东师范大学 , 2022 : 31 - 98 .
JIN Y Y . Research on TCM Prescription Recommendation based on Graph Neural Network [D]. Shanghai : East China Normal University , 2022 : 31 - 98 .
LEVENSHTEIN V I . Binary codes capable of correcting deletions, insertions, and Reversals [J]. Soviet Physics Doklady , 1966 , 10 ( 8 ): 707 - 710 .
DEVLIN J , CHANG M W , LEE K , et al . BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding [C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) . East Stroudsburg, PA : Association for Computational Linguistics , 2019 : 4171 - 4186 .
LIU Y , OTT M , GOYAL N , et al . RoBERTa: A Robustly Optimized BERT Pretraining Approach [C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP) . Hong Kong , 2019 : 2450 - 2461 .
BROWN T , MANN B , RYDER N , et al . Language Models are Few-Shot Learners [C]// Advances in Neural Information Processing Systems (NeurIPS) , Cambridge, MA : MIT Press , 2020 : 1877 - 1901 .
REIMERS N , GUREVYCH I . Sentence-BERT: Sentence Embeddings Using Siamese BERT-Networks [C]// International Joint Conference on Natural Language Processing (IJCNLP) . Hong Kong , 2019 : 3980 - 3990 .
SHUMAN D I , NARANG S K , FROSSARD P , et al . The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains [J]. IEEE Signal Proc Mag , 2013 , 30 ( 3 ): 83 - 98 .
NT H , MAEHARA T . Revisiting graph neural networks: All we have is low-pass filters [EB/OL]. ( 2019-05-26 ) [ 2023-01-10 ]. https://arxiv.org/abs/1905.09550 https://arxiv.org/abs/1905.09550 .
ZHAO M , ZHONG S , FU X , et al . Deep residual shrinkage networks for fault diagnosis [J]. IEEE Trans Industr Inform , 2019 , 16 ( 7 ): 4681 - 4690 .
邢玉瑞 . 中医方证辨证概念争鸣探讨 [J]. 中华中医药杂志 , 2018 , 33 ( 7 ): 2747 - 2750 .
XING Y R . Discussion on the concept of traditional medical formula and syndrome differentiation [J]. China Journal of Traditional Chinese Medicine and Pharmacy , 2018 , 33 ( 7 ): 2747 - 2750 .
ISOGAWA K , IDA T , SHIODERA T , et al . Deep shrinkage convolutional neural network for adaptive noise reduction [J]. IEEE Signal Proc Let , 2017 , 25 ( 2 ): 224 - 228 .
0
浏览量
0
下载量
0
CSCD
0
CNKI被引量
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621