Research on Human Acupoint Detection by Integrating Key Point Information and Acupoint Theory
IEEE Access · 2024 · 12 · 181889-181898
DOI: 10.1109/access.2024.3509026 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2024. Verfasst von Fengshou Zhang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Based on the construction of the human body key point recognition model and the study of Chinese medicine meridian and acupuncture point theory, the detection framework for human body acupuncture points was realized by combining the relationship between key points and acupuncture points. In the human body acupuncture point detection framework, deep learning is first used to construct the human body key point recognition model, and the model is trained with a human body image dataset. The target detection model is built using the Faster R-CNN algorithm to perform human target detection, while the HRNet key point recognition algorithm is employed to detect human key points through thermal regression. Based on the “acupoint-located method by bone proportional Cun” from the standards of Chinese medicine for acupuncture points, this study proposes a calculation method for converting pixel values to ‘cun’ values. By examining the positional relationships of ‘cun’ values between different acupoints in the theory of channels and meridians in Chinese medicine, the pixel coordinates of the acupoints are calculated after conversion. The key point detection model is then utilized to detect the pixel coordinates of multiple acupoints in the input human image, enabling the visualization of these acupoints on the image. The experimental results successfully detected all 21 acupoints in the Bladder Meridian of Foot-Taiyang, and the moxibustion efficacy was not affected within the error range. Through information fusion, the combination of the human body key point detection algorithm with Chinese medicine meridian and acupoint theory enabled successful detection of acupoints in human body images, providing improved support for intelligent moxibustion point recognition.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhang, F., An, Q., Song, W., & Bai, M. (2024). Research on Human Acupoint Detection by Integrating Key Point Information and Acupoint Theory. IEEE Access, 12, 181889-181898. https://doi.org/10.1109/access.2024.3509026
Vancouver
Zhang F, An Q, Song W, Bai M. Research on Human Acupoint Detection by Integrating Key Point Information and Acupoint Theory. IEEE Access. 2024;12:181889-181898. doi:10.1109/access.2024.3509026
BibTeX
@article{Zhang2024,
author = {Zhang, Fengshou and An, Qingqin and Song, Weidong and Bai, Mengxuan},
title = {Research on Human Acupoint Detection by Integrating Key Point Information and Acupoint Theory},
journal = {IEEE Access},
year = {2024},
volume = {12},
pages = {181889--181898},
doi = {10.1109/access.2024.3509026}
}
EUVIMED