多特征融合的远程会诊智能分诊研究

doi:10.3778/j.issn.1002-8331.2309-0080

摘要/Abstract

摘要： 为解决调研中发现的现阶段大多数远程会诊平台仍依赖于人工分诊的问题，探索提出一种融合多特征的BTB智能分诊模型。假设数据中所安排的会诊科室为实际最优会诊科室，选取患者年龄、性别、基层医生初步诊断作为分诊依据。针对远程会诊申请信息复杂，重点分布不均的特点，使用BERT模型训练词向量，在此基础上，利用文本卷积神经网络（TextCNN）和双层双向门控循环神经网络（double-layer BiGRU）捕捉远程会诊申请的局部特征信息和上下文信息。将BERT特征与TextCNN特征、双层BiGRU特征相融合，最后输入到全连接层中，为基层医生推荐多个合适的会诊科室。模型在国家远程医疗中心大数据上进行多次训练和测试，当推荐科室为5时，模型的准确率可达90.77%，MRR达到76.45%。对比实验表明，提出的BTB智能分诊模型的性能比传统的机器学习模型更优，可为远程医疗平台实现智能化提供理论支撑和实践指导。

关键词: 智能分诊, 多特征融合, 远程会诊, 神经网络

Abstract: A multi-feature BTB intelligent triage model is proposed to address the problem that most teleconsultation platforms still rely on manual triage. This model selects the actual optimal consultation departments from the data, and uses age, gender, and initial diagnosis of primary doctors as the basis for triage. In response to the complexity and uneven distribution of teleconsultation application information, the BERT model is used to train word vectors. On this basis, text convolutional neural network (TextCNN) and double-layer bidirectional gated recurrent unit (BiGRU) are utilized to capture local feature information and context information of teleconsultation application. Additionally, the BERT features are combined with TextCNN features and double-layer BiGRU features. Finally, these features are fed into the fully connected layer to recommend multiple suitable consultation departments for primary care doctors. The proposed BTB intelligent triage model is trained and tested on a large data set from the National Telemedicine Center of China, and when the recommended department is 5, the model can reach 90.77% in accuracy and 76.45% in MRR. Comparative experiments demonstrate that the proposed model outperforms traditional machine learning models and can serve as a useful theoretical reference and practical guide to enhance the intelligence of telemedicine platforms.

Key words: intelligent triage, multi-feature fusion, teleconsultation, neural networks

翟运开, 李俊珂, 李金林, 乔岩. 多特征融合的远程会诊智能分诊研究[J]. 计算机工程与应用, 2025, 61(3): 326-335.

ZHAI Yunkai, LI Junke, LI Jinlin, QIAO Yan. Intelligent Triage of Teleconsultation Based on Multi-Feature Fusion[J]. Computer Engineering and Applications, 2025, 61(3): 326-335.

参考文献

[1] ZHANG H P, ZHOU W H, SUN Y J. Joint allocation of emergency medical resources with time-lag correlation during cross-regional epidemic outbreaks[J]. Computers & Industrial Engineering, 2022, 164: 107895.
[2] DORSEY E R, TOPOL E J. State of telehealth[J]. New England Journal of Medicine, 2016, 375(2): 154-161.
[3] BAUDIER P, KONDRATEVA G, AMMI C, et al. Digital transformation of healthcare during the COVID-19 pandemic: patients’ teleconsultation acceptance and trusting beliefs[J]. Technovation, 2023, 120: 102547.
[4] 董天舒, 张梅奎, 艾雪伟. 远程会诊自动化分诊处理系统的设计与研发[J]. 中国数字医学, 2016, 11(12): 53-55.
DONG T S, ZHANG M K, AI X W. The consideration and design on automatic triage and and processing system of telemedicine consultation[J]. China Digital Medicine, 2016, 11(12): 53-55.
[5] 董天舒, 张梅奎. 医院预约挂号模式在远程会诊调度环节的运用与思考[J]. 中国医院管理, 2017, 37(1): 40-41.
DONG T S, ZHANG M K. Practice and thinking of using appointment service of hospital outpatient in telemedicine consultation scheduling work[J]. Chinese Hospital Management, 2017, 37(1): 40-41.
[6] ISERSON K V, MOSKOP J C. Triage in medicine, part I: concept, history, and types[J]. Annals of Emergency Medicine, 2006, 49(3): 275-281.
[7] HAMID R A, ALBAHRI A S, ALBAHRI O S, et al. Dempster-shafer theory for classification and hybridised models of multi-criteria decision analysis for prioritisation: a telemedicine framework for patients with heart diseases[J]. Journal of Ambient Intelligence and Humanized Computing, 2022, 13: 4333-4367.
[8] TSCHANDL P, RINNER C, APALLA Z. et al. Human-computer collaboration for skin cancer recognition[J]. Nature Medicine, 2020, 26: 1229-1234.
[9] XIE Y C, QUANG D N, HASLINA H, et al. Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: an economic analysis modelling study[J]. The Lancet Digital Health, 2020, 2(5): e240-e249.
[10] SAITEJA P C, GAHANGIR H, AYUSH G, et al. Smart home health monitoring system for predicting type 2 diabetes and hypertension[J]. Journal of King Saud University-Computer and Information Sciences, 2022, 34(3): 862-870.
[11] 史嘉兴, 唐锐, 何雨昆, 等. 基于疫情防控的医院业务流程再造研究[J]. 中国数字医学, 2020, 15(5): 67-69.
SHI J X, TANG R, HE Y K, et al. The consideration and design on automatic triage and processing system of telemedicine consultation[J]. China Digital Medicine, 2020, 15(5): 67-69.
[12] 王若佳, 王继民. 用户认知视角下在线问诊平台医生推荐研究[J]. 图书情报工作, 2023, 67(10): 128-138.
WANG R J, WANG J M. Research on doctor recommendation of online “ask the doctor” platforms based on the perspective of users recognition[J]. Library and Information Service, 2023, 67(10): 128-138.
[13] 周鑫, 熊回香, 肖兵. 一种融合标签和患者咨询文本的医生推荐算法[J]. 情报科学, 2023, 41(3): 145-154.
ZHOU X, XIONG H X, XIAO B. A physician recommendation algorithm based on the fusion of label and patient consultation text[J]. Information Science, 2023, 41(3): 145-154.
[14] 路薇, 高盼, 翟运开. 带有反馈调节的远程医疗专家自适化推荐[J]. 系统管理学报, 2023, 32(5): 960-975.
LU W, GAO P, ZHAI Y K. An adaptive recommendation method for telemedicine specialists with feedback adjustment[J]. Journal of Systems & Management, 2023, 32(5): 960-975.
[15] BELLO A, NG S, LEUNG M. A BERT framework to sentiment analysis of tweets[J]. Sensors, 2023, 23(1): 506.
[16] 李文亮, 杨秋翔, 秦权. 多特征混合模型文本情感分析方法[J]. 计算机工程与应用, 2021, 57(19): 205-213.
LI W L, YANG Q X, QIN Q. Multi-feature mixed model text sentiment analysis method[J]. Computer Engineering and Applications, 2021, 57(19): 205-213.
[17] ZHOU Y J, LI J L, CHI J H, et al. Set-CNN: a text convolutional neural network based on semantic extension for short text classification[J]. Knowledge-Based Systems, 2022, 257(5): 109948.
[18] TONG B, NI R, BAOJIA W, et al. A BERT-based hybrid short text classification model incorporating CNN and attention-based BiGRU[J]. Journal of Organizational and End User Computing (JOEUC), 2021, 33(6): 1-21.
[19] 杨文涛, 雷雨琦, 李星月, 等. 融合汉字输入法的BERT与BLCG的长文本分类研究[J]. 计算机工程与应用, 2024, 60(9): 196-202.
YANG W T, LEI Y Q, LI X Y, et al. Chinese long text classification model based on BERT fused Chinese input methods and BLCG[J]. Computer Engineering and Applications, 2024, 60(9): 196-202.
[20] JIANG X, SONG C, XU Y, et al. Research on sentiment classification for netizens based on the BERT-BiLSTM-TextCNN model[J]. PeerJ Computer Science, 2022, 8(3): e1005.
[21] 赵宏, 傅兆阳, 王乐. 基于特征融合的中文文本情感分析方法[J]. 兰州理工大学学报, 2022, 48(3): 94-102.
ZHAO H, FU Z Y, WANG L. Sentiment analysis of Chinese text based on feature fusion[J]. Journal of Lanzhou University of Technology, 2022, 48(3): 94-102.
[22] ZHOU P, SHI W, TIAN J, et al. Attention-based bidirectional long short-term memory networks for relation classification[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics. Berlin, Germany: ACL, 2016: 207-212.
[23] 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, 2019: 4171-4186.
[24] BAO G, ZHANG C X, LIU J M, et al. Improving text classification with weighted word embeddings via a multi-channel TextCNN model[J]. Neurocomputing, 2019, 363: 366-374.
[25] CHEN H, ZHANG Z P, HUANG S T, et al. TextCNN-based ensemble learning model for Japanese text multi-classification[J]. Computers and Electrical Engineering, 2023, 109(B): 108751.
[26] ZHANG B, XIONG D, XIE J, et al. Neural machine translation with GRU-gated attention model[J]. IEEE Transactions on Neural Networks and Learning Systems, 2020, 31(11): 4688-4698.
[27] 国家卫生和计划生育委员会. 远程医疗信息系统建设技术指南(2014 年版)[Z]. 2014.
National Health and Family Planning Commission. Technical guidelines for the construction of telemedicine information systems (2014 edition) [Z]. 2014.