Visualization Engineering Platform for TCM Pulse Diagnosis - Pulse Diagnosis Based on Federated Learning to Diagnose Slippery and Choppy and Other Pulses Waveform Image Features to Assist in the Study of TCM Pathological Logic Analysis

The diagnoses processes of Traditional Chinese Medicine (TCM) focus on the following four main types of diagnoses methods consisting of inspection, olfaction, inquiry, and palpation. The most important one is palpation also called pulse diagnosis which is to measure wrist artery pulse by TCM doctor's fingers to detect patient's health state. The pulse diagnosis has three parts, namely 'Chun', 'Guan' and 'Chy', with the location. Wrist measurements correspond to different parts of the body's organs. In this project, it is to classify pulse types by using specialized pulse measuring instruments. The measured pulse wave (Measured Pulse Wave, MPW) was segmented into arterial pulse wave curves (APWC) by the image suggestion method. The research object of this project is to collect and group patients diagnosed by traditional Chinese medicine practitioners, namely slippery pulse, choppy pulse group and normal pulse control group, with at least 80 cases for each group. The research purpose of this project is mainly to carry out the visualization engineering platform of TCM pulse diagnosis - based on the pulse diagnosis of federated learning to diagnose the pulse waveform image features such as slippery pulse and choppy pulse to provide auxiliary TCM pathological logic analysis research and back-end cross-federal learning of TCM pulse diagnosis Implementation of the node system. In other words, it is expected that the pulse wave characteristics measured by TCM physicians who cooperate with experts in the field can be collected from many TCM pulse diagnosis federated learning nodes, and analyzed by the Multiple-Expert Repertory Grid Elicitation (MERGE) method. Finally, the artificial intelligence model based on FL is trained to carry out TCM pathological logic analysis and related research. The results will be provided to TCM physicians as an important reference to assist clinical diagnosis.

Study Overview

• Status: Recruiting
• Conditions: Slippery Pulse; Choppy Pulse
• Intervention / Treatment: Diagnostic test: Recurrent Neural Network

• Study Type: Observational
• Enrollment (Anticipated): 240

Contacts and Locations

Study Locations

• Taiwan
  • North District
    • Taichung, North District, Taiwan, 40447
      • Recruiting
      • China Medical University Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 20 years to 90 years (Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

outpatient

Description

Inclusion Criteria:

1. Already sign test consent permit.
2. More than 20-year-old.

Exclusion Criteria:

1. There is a wound or inflammation at the wrist skin measurement.

Study Plan

How is the study designed?

Design Details

• Observational Models: Other
• Time Perspectives: Prospective

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Slippery Pulse	The purpose of the study was to establish a scientific slippery pulse diagnosis system	30 minutes in duration

Collaborators and Investigators

• Sponsor: China Medical University Hospital

Study record dates

Study Major Dates

• Study Start (Anticipated): December 1, 2022
• Primary Completion (Anticipated): September 1, 2024
• Study Completion (Anticipated): September 1, 2024

Study Registration Dates

• First Submitted: November 20, 2022
• First Submitted That Met QC Criteria: November 20, 2022
• First Posted (Actual): November 29, 2022

Study Record Updates

• Last Update Posted (Estimate): December 6, 2022
• Last Update Submitted That Met QC Criteria: December 2, 2022
• Last Verified: December 1, 2022

More Information

Terms related to this study

• Keywords: Artificial intelligence; Recurrent Neural Network; Wrist artery pulse diagnosis instrument; TCM pulse diagnosis; Feature extraction
• Additional Relevant MeSH Terms: Pathologic Processes; Disease
• Other Study ID Numbers: CMUH111-REC2-168

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: No

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No