Building a Traditional Chinese Medicine Clinical Diagnosis and Treatment Database

Building a Traditional Chinese Medicine Clinical Diagnosis and Treatment Database: A Prospective Multicenter Cross-Sectional Study

Collecting Traditional Chinese Medicine (TCM) clinical diagnosis and treatment data, including doctor-patient dialogues, tongue diagnosis, facial diagnosis, and TCM constitution information, to construct databases for tongue diagnosis, TCM constitution, and doctor-patient dialogues. Based on artificial intelligence technology, engage in research related to the standardization and intelligentization of TCM.

Study Overview

• Status: Not yet recruiting
• Conditions: Artificial Intelligence; Medicine, Chinese Traditional
• Intervention / Treatment: Other: Observational study, non intervention

Detailed Description

The technological principles of large language models align with the empirical medical principles of Traditional Chinese Medicine (TCM), and the rise of large model technology can greatly promote the progress of TCM. However, there is currently a lack of clinical diagnosis and treatment databases with TCM characteristics for training TCM artificial intelligence(AI) large models.

At present, a large-scale tongue image database has not yet been established for modeling common TCM tongue appearances, thereby ensuring the accuracy and consistency of TCM diagnosis and promoting the objective standardization of TCM diagnostic development.

Considering the feedback from the subjects in clinical work that the TCM constitution survey questionnaire has a large volume, takes a long time, and has certain subjective issues, we plan to carry out a large-scale clinical observational study to optimize the process of TCM constitution identification.

Traditional Chinese Medicine (TCM) doctor-patient dialogues and medical record writing are essential entities generated during the TCM diagnosis and treatment process. Assisting in consultation, medical record generation, and treatment plan recommendations based on doctor-patient dialogues have significant clinical and research value. Therefore, we plan to collect a large number of doctor-patient dialogues and outpatient medical records to construct a doctor-patient dialogue database, preparing in advance for optimizing interactive large-scale TCM models.

In summary, the research on constructing a TCM clinical diagnosis and treatment database has important clinical and scientific research value. This will help to improve the standardization and normalization of TCM diagnosis and treatment, and also support the modernization and internationalization of TCM. By applying big data analysis and artificial intelligence technology, it is possible to delve deeper into TCM diagnosis and treatment information, providing richer and more accurate data resources for clinical decision-making and scientific research exploration in TCM.

• Study Type: Observational
• Enrollment (Estimated): 80000

Contacts and Locations

• Study Contact: Name: Yulong Zhang, Doctor; Phone Number: 18810550602; Email: zhongxiyi1101@163.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

There are no specific restrictions on the disease, gender, or health status of the enrolled population.

Description

Inclusion Criteria:

• People who come to the hospital for physical examination and medical treatment；
• Participants voluntarily participate in the study.

Exclusion Criteria:

• Subjects with difficulty in tongue extension, communication, etc. who cannot cooperate with data collection；
• The researchers determined that there were other factors that may have forced the subjects to terminate the study.

Study Plan

How is the study designed?

Number of groups / cohorts

3

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Traditional Chinese Medicine Tongue Image Group; Internally, using random allocation, divided into training group and validation group	Other: Observational study, non intervention; Observational study, non intervention
Traditional Chinese Medicine Constitution Data Group; Internally, using random allocation, divided into training group and validation group	Other: Observational study, non intervention; Observational study, non intervention
Traditional Chinese Medicine Doctor Patient Dialogue Data Group; Data used for fine-tuning traditional Chinese medicine models	Other: Observational study, non intervention; Observational study, non intervention

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Development of a tongue image-based machine learning tool	It is anticipated to enroll 50,000 samples to establish a Traditional Chinese Medicine (TCM) tongue appearance database.; The tongue images will undergo quality selection and preprocessing.; The tongue images will be manually annotated, with 40% allocated to the training group and 60% to the testing group.; For the training group: A TCM tongue appearance model will be constructed based on the manually annotated tongue images.; For the testing group: The TCM tongue appearance model will be used to interpret the tongue images.; Analyze the consistency between the tongue appearance interpretations by the model built from the training group and the readings by TCM physicians for the testing group's tongue images.	20 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
TCM Constitution Multimodal Model	It is projected to include 20,000 samples to construct a Traditional Chinese Medicine (TCM) constitution database.; Of these, 15,000 samples will complete the TCM constitution survey questionnaire and have their tongue and facial diagnosis data collected.; Based on the data from step 2, a multimodal TCM constitution assessment model will be constructed.; An additional 5,000 samples will be tested, first by completing the multimodal TCM constitution assessment model, and then the TCM constitution survey questionnaire.; Analyze the consistency between the readings of the TCM constitution assessment model and the results from the TCM constitution survey questionnaire generation, and treatment plan recommendations, prepare in advance for optimizing interactive large-scale TCM models.	20 months

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Traditional Chinese Medicine (TCM) Doctor-Patient Dialogue Database	It is anticipated to include 10,000 samples to construct a Traditional Chinese Medicine (TCM) doctor-patient dialogue database.; Optimize the TCM large model, which mainly includes the following aspects: a Q&A system, assistance in medical record generation, and recommendation of diagnosis and treatment plans.	20 months

Collaborators and Investigators

• Sponsor: Fifth Affiliated Hospital, Sun Yat-Sen University
• Investigators: Principal Investigator: Qi Zeng, Doctor, Fifth Affiliated Hospital, Sun Yat-Sen University

Publications and helpful links

General Publications

• Yuan L, Yang L, Zhang S, Xu Z, Qin J, Shi Y, Yu P, Wang Y, Bao Z, Xia Y, Sun J, He W, Chen T, Chen X, Hu C, Zhang Y, Dong C, Zhao P, Wang Y, Jiang N, Lv B, Xue Y, Jiao B, Gao H, Chai K, Li J, Wang H, Wang X, Guan X, Liu X, Zhao G, Zheng Z, Yan J, Yu H, Chen L, Ye Z, You H, Bao Y, Cheng X, Zhao P, Wang L, Zeng W, Tian Y, Chen M, You Y, Yuan G, Ruan H, Gao X, Xu J, Xu H, Du L, Zhang S, Fu H, Cheng X. Development of a tongue image-based machine learning tool for the diagnosis of gastric cancer: a prospective multicentre clinical cohort study. EClinicalMedicine. 2023 Feb 6;57:101834. doi: 10.1016/j.eclinm.2023.101834. eCollection 2023 Mar.
• Tian F, Liu D, Wei N, Fu Q, Sun L, Liu W, Sui X, Tian K, Nemeth G, Feng J, Xu J, Xiao L, Han J, Fu J, Shi Y, Yang Y, Liu J, Hu C, Feng B, Sun Y, Wang Y, Yu G, Kong D, Wang M, Li W, Chen K, Li X. Prediction of tumor origin in cancers of unknown primary origin with cytology-based deep learning. Nat Med. 2024 May;30(5):1309-1319. doi: 10.1038/s41591-024-02915-w. Epub 2024 Apr 16.
• Tan Y, Zhang Z, Li M, Pan F, Duan H, Huang Z, Deng H, Yu Z, Yang C, Shen G, Qi P, Yue C, Liu Y, Hong L, Yu H, Fan G, Tang Y. MedChatZH: A tuning LLM for traditional Chinese medicine consultations. Comput Biol Med. 2024 Apr;172:108290. doi: 10.1016/j.compbiomed.2024.108290. Epub 2024 Mar 13.

Study record dates

Study Major Dates

• Study Start (Estimated): August 1, 2024
• Primary Completion (Estimated): August 15, 2024
• Study Completion (Estimated): August 15, 2026

Study Registration Dates

• First Submitted: July 18, 2024
• First Submitted That Met QC Criteria: July 24, 2024
• First Posted (Actual): July 29, 2024

Study Record Updates

• Last Update Posted (Actual): July 29, 2024
• Last Update Submitted That Met QC Criteria: July 24, 2024
• Last Verified: July 1, 2024

More Information

Terms related to this study

• Other Study ID Numbers: ZDWY.ZYZLK.009

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No