Abstract:Pulse diagnosis is one of the most characteristic diagnostic methods in Traditional Chinese Medicine(TCM), as practiced and providing medical service in disease prevention, diagnosis and treatmentterritory. Lack of standardization and subjective of pulse diagnosis, is the main concern in modern Chinese medicine. There are problems in present pulse diagnosis equipment, such as low precision of pre-pressure control for pulse wave acquisition, weak anti-interference ability in the acquisition process and poor ability to adapt to different groups of people. In this study, we proposed an interactive model between pulse acquisition dexterity finger and flexible skin was identified by using nonlinear autoregressive networkwith exogenous inputs (NARX)neural network model, which was further used to simulate and test the entire control strategy. Secondly, an adaptive control strategy based on Radical Basis Function (RBF) neural network was proposed, and the stability of the closed-loop system was proved by Lyapunov stability theory. Finally, the simulation and laboratory experiments were conducted. The simulation experimentsshow that the mean square error(MSE)of the system identification model, which is based on NARX neural network model accurately describes the mathematical relationship between “Guan” wrist tissue and dexterous fingers, is 1.1878, and the percentage of relative mean square error is 0.36953%,and it could means that the model achieves the modeling accuracy needed for control purposes; The adaptive RBF neural network controller is at least 1.67 times faster than PID controller, 3.76 times smaller of maximum tracking negative error than PID controller, at the same time the laboratory experiments get the similar results. So it could be concluded that the adaptive RBF neural network controller has the characteristics of fast adjustment speed, small overshot and strong interference suppression ability, which achieves the precision and robustness of pre-pressure control in the process of pulse wave acquisition.

Pulse localization networks with infrared camera

Palpation Localization of Radial Artery Based on 3-Dimensional Convolutional Neural Networks

Accurate Targeting in Robot-Assisted TCM Pulse Diagnosis Using Adaptive Sensor Fusion

SFCN: Spoon Fully Convolutional Networks for Pulse Localization.

OCViTnet: Pulse Wave Learning for Cuffless Blood Pressure Estimation of Medical Service Robots

Full-field Wrist Pulse Signal Acquisition and Analysis by 3D Digital Image Correlation

Intelligent Detection of Pulse Position in Traditional Chinese Medicine Based on Bionic Pulse Diagnose Robotics

3Dcnn-Based Palpation Localization with Temporal Attention Module.

Research on the Best Measurement Situation Between Optical Probe and Tissue Surfaces in Non-Invasive Detection.

Robust Efficient Estimation of Heart Rate Pulse from Video.

Ballpoint-Pen Like Probes for Multipoint Dynamic Pulse Diagnosis System

Development and evaluation of a robotic system for lumbar puncture and epidural steroid injection

Remote Heart Rate Measurement From Near-Infrared Videos Based on Joint Blind Source Separation With Delay-Coordinate Transformation

PulseNet: A multitask learning network for remote heart rate estimation

An accurate tidal peak localization method in radial arterial pulse signals based on hybrid neural networks

A Feasibility Study of a Video-Based Heart Rate Estimation Method with Convolutional Neural Networks

Human Pulse Diagnosis For Medical Assessments Using A Wearable Piezoelectret Sensing System

Research on Haptic Feedback Control System of Bionic Pulse Diagnosis Equipment

Intelligent Wearable Wrist Pulse Detection System Based on Piezoelectric Sensor Array

3D Convolutional Neural Networks for Remote Pulse Rate Measurement and Mapping from Facial Video

Non-contact continuous blood pressure measurement based on imaging equipment