基于真实世界研究的甲状腺疾病中医诊疗数据采集系统的构建研究

Research on Construction of Data Collection System for Real-world Thyroid Disease-based Traditional Chinese Medicine Diagnosis and Treatment

目的:构建基于真实世界研究的甲状腺疾病中医诊疗数据采集系统,为中医治疗甲状腺疾病的疗效评价和院前智慧分诊提供数据采集平台和算法模型。方法:采用文献研究法、自然语言处理法提取甲状腺疾病诊疗信息,动态量表自测法收集患者诊前四诊信息、系统测试法用来检测构建的系统性能,最终搭建集诊前筛查、诊中治疗、诊后随访于一体的甲状腺疾病中医诊疗数据采集系统。结果:构建了一套适用于甲状腺疾病的中医诊疗数据采集系统;最终从完整性、适用性、科学性3个方面进行了评价,发现系统操作需要点击次数比现有中医门诊系统减少6次(从平均56次减少到50次)、操作时间快2-3分钟(从平均10分钟减少到8分钟)、信息填写完整度高58%(从平均17%提高到75%)、内容包含度高22%(从平均58%提高到80%)。结论:研究构建的甲状腺疾病中医诊疗数据采集系统在数据采集和数据分析方面性能都较现有系统有所提高。系统充分整合了互联网、大数据以及人工智能等尖端技术,为患者提供了诊前筛查、诊中治疗、以及诊后随访等一系列全面的患者管理服务。这不仅在医院与患者之间建立了畅通的沟通桥梁,还实现了患者在离院后至再次入院的复诊全程智能化管理流程。

Objective:To establish a real-world TCM specialty data collection system for thyroid disorders and to design and construct a data prediction model to provide data evidence for the evaluation of the efficacy of TCM treatment of thyroid disorders,so as to provide a data collection platform and algorithmic model for pre-hospital smart triage and formation of outcomes in clinical departments.Methods:The integrated system for data collection of Traditional Chinese Medicine diagnosis and treatment of thyroid diseases,covering pre-screening,in-treatment,and post-treatment follow-up,was built by combining the literature research method and natural language processing method to extract information related to the diagnosis and treatment of thyroid diseases,the dynamic scale self-test method to collect the four diagnostic methods(inspection,auscultation and olfaction,inquiry,and palpation)information from patients before their visits,and the system testing method to assess and evaluate the performance of the constructed system.Results:Construction of a clinical information collection system for thyroid disorders in Chinese medicine.The final evaluation was conducted in three aspects:completeness,applicability,and scientificity,it was found that the number of operation clicks of this system was 6 times less than that of the existing TCM outpatient system(from an average of 56 clicks to 50 clicks),the operation time was 2-3 minutes faster(from an average of 10 minutes to 8 minutes),the degree of completeness of the information filled in was 58% higher(from an average of 17% to 75%),and the degree of content inclusion was 22% higher(from an average of 58% to 80%).Conclusion:The platform has improved performance in both data collection and data analysis compared to existing clinical systems.This clinical research evidence system makes full use of cutting-edge technologies such as the Internet,big data and artificial intelligence to provide a series of patient management services such as patient pre-consultation screening,in-consultation treatment and post-consultation follow-up,which builds a bridge of communication between hospitals and patients and realizes a closed-loop intelligent management process of patient follow-up from departure to admission.