Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Carbon dot crosslinked hydrogel with microenvironment modulation for wound healing via synergistic antibacterial, antioxidant and hypoxia mitigation.

Colloids and surfaces. B, Biointerfaces·2026
Same author

Screening of immune adjuvants for an attenuated live vaccine derived from the Aeromonas veronii TH0426 strain with a hisJ gene deletion and evaluation of its immune protection in crucian carp.

Fish & shellfish immunology·2026
Same author

Peritectic engineering enhanced thermoelectrics for smart thermal messaging devices.

Nature communications·2026
Same author

Multifunctional 3D MXene@MDCF Nanoarchitectures via Hierarchical Networks Self-Assembled Strategy Achieving Superior Electromagnetic Wave Absorption and Flame Retardancy with Enhanced Thermal Management.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Discovery of Highly Potent and Selective IKZF2 Degraders for Cancer Immunotherapy.

Journal of medicinal chemistry·2026
Same author

Efficacy and safety of RC48 alone or combined with PD-1 inhibitors in high-risk non-muscle invasive bladder cancer.

Scientific reports·2026

Related Experiment Video

Updated: Jan 9, 2026

Author Spotlight: Assessing Brain Activity in Robotic-Assisted Lower Limb Rehabilitation Using fNIRS
05:25

Author Spotlight: Assessing Brain Activity in Robotic-Assisted Lower Limb Rehabilitation Using fNIRS

Published on: June 7, 2024

1.7K

A novel intelligent physiotherapy robot based on dynamic acupoint recognition method.

Yuhan Zhang1, Shiyang Sun1, Donghui Zhao1

  • 1School of Electrical Engineering, Shenyang University of Technology, Shenyang, China.

Frontiers in Neurorobotics
|December 10, 2025
PubMed
Summary
This summary is machine-generated.

This study introduces a novel two-stage method for precise acupoint recognition on the human back, enhancing physiotherapy robot accuracy. The system achieves high recall and low error, enabling reliable autonomous treatments.

Keywords:
RTMDet networkRTMPose networkacupoint recognitionphysiotherapy robotphysiotherapy task

More Related Videos

Visualizing Motion Patterns in Acupuncture Manipulation
08:18

Visualizing Motion Patterns in Acupuncture Manipulation

Published on: July 16, 2016

9.1K
Soft Pneumatic Robot Modulates Graph Theory Metrics of Brain Network for Hand Rehabilitation After Stroke
05:30

Soft Pneumatic Robot Modulates Graph Theory Metrics of Brain Network for Hand Rehabilitation After Stroke

Published on: October 10, 2025

393

Related Experiment Videos

Last Updated: Jan 9, 2026

Author Spotlight: Assessing Brain Activity in Robotic-Assisted Lower Limb Rehabilitation Using fNIRS
05:25

Author Spotlight: Assessing Brain Activity in Robotic-Assisted Lower Limb Rehabilitation Using fNIRS

Published on: June 7, 2024

1.7K
Visualizing Motion Patterns in Acupuncture Manipulation
08:18

Visualizing Motion Patterns in Acupuncture Manipulation

Published on: July 16, 2016

9.1K
Soft Pneumatic Robot Modulates Graph Theory Metrics of Brain Network for Hand Rehabilitation After Stroke
05:30

Soft Pneumatic Robot Modulates Graph Theory Metrics of Brain Network for Hand Rehabilitation After Stroke

Published on: October 10, 2025

393

Area of Science:

  • Robotics
  • Computer Vision
  • Biomedical Engineering

Background:

  • Physiotherapy robots require precise acupoint recognition for safe and effective treatment.
  • Accurate acupoint location on the human back is challenging due to limited external features.
  • Existing methods for hand and ear acupoint recognition are extensive but not directly applicable to the back.

Purpose of the Study:

  • To develop a robust and efficient acupoint recognition system for physiotherapy robots targeting the human back.
  • To improve the precision and reliability of autonomous acupoint identification.
  • To enable automated physiotherapy task path planning based on recognized acupoint coordinates.

Main Methods:

  • A two-stage recognition approach utilizing RTMDet for back region extraction and RTMPose for acupoint localization.
  • Conversion of acupoint coordinate regression to sub-pixel classification using the SimCC framework for enhanced speed and accuracy.
  • Integration of CSPNeXt for multi-layer feature fusion to improve feature extraction capabilities.
  • Development of a physiotherapy interaction interface for robot task path planning using 3D acupoint coordinates.

Main Results:

  • Achieved a recall of 90.17% on human datasets with an average detection error of 5.78 mm.
  • Demonstrated accurate identification of acupoints across different back postures.
  • Attained an inference speed of 30 FPS on a 4070Ti GPU, suitable for real-time applications.
  • Successfully conducted continuous physiotherapy tasks on multiple acupoints.

Conclusions:

  • The proposed two-stage acupoint recognition method significantly enhances the accuracy and reliability of physiotherapy robots.
  • This approach holds broad potential for advancing autonomous physiotherapy applications.
  • The system's effectiveness in real-world scenarios paves the way for improved patient treatment outcomes.