Research of Acupoint Location Methods in Traditional Chinese

Medicine

Zhaozhao Fang

College of Optoelectronics and Information Engineering, Fujian Normal University, 350000, Fuzhou, China

Keywords: TCM Acupoints, Acupoint Location, Acupoint Recognition.

Abstract: The advancement of science and technology has brought new opportunities and challenges to the

development of Chinese medicine. In recent years, people have conducted extensive research on the

methods of applying modern science and technology to acupoint positioning in Chinese medicine. Based on

the current research status, the report summarizes and categorizes the general methods of the past 10 years.

The methods are roughly divided into four categories, which are based on the direction of vision

technology, based on algorithms, based on the direction of the electrical impedance characteristics of

acupoints, and other methods. The methods are roughly divided into four categories, which are basedd on

the direction of vision, algorithms, the direction of the electrical impedance characteristics of acupoint, and

other methods. This paper makes a simple comparison of its advantages and disadvantages, and

looksforward to the future research direction based on the integration of various other methods of vision

technology. Thus, it could provide a certain reference for the future development of acupuncture point

positioning technology.

1 INTRODUCTION

Nowadays, people's pressure is increasing, and more

and more people are paying attention to medical

care. Traditional Chinese medicine plays a very

important role in health care. Our human body has

many acupuncture points. Acupoints can not only

treat diseases clinically, but also have great health

benefits in daily life. Therefore, the accuracy of

finding and positioning acupoints plays a decisive

role in traditional Chinese medicine (Chang 2017).

Failure to find accurate acupuncture points will not

only affect the effect of treatment and health care,

but also impair health. It takes a lot of energy and

time to find and locate acupuncture points, which is

even more difficult for non-professionals. With the

development of science and technology, people

began to try to combine the search and positioning

of acupoints with modern technology, so as to

achieve higher accuracy in finding and positioning

acupoints and the liberation of human labor.

Nowadays, there are more and more researches on

acupoint location. This report categorizes the general

research methods into 4 categories, and summarizes

some of the specific methods based on these 4

categories, and then compares the advantages and

disadvantages of the 4 categories of methods, and

puts forward the prospects for future research

directions.

2 RESEARCH CLASSIFICATION

2.1 Positioning based on the Direction

of Vision Technology

The machine vision is used to extract and organize

the data of the collected images, so as to realize the

accurate positioning of the acupuncture points. Ma

Zhewen and Yu Haoguang established a massage

robot acupoint tracking system based on visual

positioning, and realized real-time detection and

dynamic tracking of acupoints through image

acquisition, image processing, coordinate

conversion, etc. (Ma & Yu. 2010). Huanbing Gao et

al. judge whether the human body is moving by

comparing two adjacent images, and use methods

such as sharpening using morphology to process the

image using multiple morphological structural

elements to corrode the prior edges, and recognize

that the values in the coordinates should be

Fang, Z.

Research of Acupoint Location Methods in Traditional Chinese Medicine.

DOI: 10.5220/0011373100003438

In Proceedings of the 1st International Conference on Health Big Data and Intelligent Healthcare (ICHIH 2022), pages 513-517

ISBN: 978-989-758-596-8

513

converted. It is the movement coordinates, so as to

realize the dynamic tracking of the acupoints (Gao,

Lu, & Wang 2014).

Zhang Huakai designed a monocular visual servo

system in 2012, proposed an acupoint location

method based on artificial signs, and compared the

acupoint location based on fast normalized

correlation gray-scale matching algorithm with the

second time based on Mahalanobis distance.

Matching the stability of the SIFT feature to match

the acupoint location algorithm, it is concluded that

the latter has better stability (Zhang 2012). Sheng

Lin et al. further considered that based on computer

binocular vision technology, collecting images of

different human limbs, obtaining three-dimensional

coordinates in space, through technical analysis, to

realize the recognition and positioning of human

acupoints, which is more accurate and has lower

error (Lin, & Yi 2019).

Figure 1: The basic block diagram of the positioning research of binocular vision (Lin, & Yi 2019).

Zhao Yang et al. applied infrared image

processing to TCM diagnosis and treatment in 2012,

and proposed an automatic facial acupoint location

algorithm, which combines Minimum Eigenvalue

corner detection and Log edge detection to locate the

position of basic facial features. On this basis,

automatic acupuncture points are realized.

Positioning (Zhao, Zhang & Lian 2012). In 2020,

Wang Cong carried out research on the vision-based

acupoint-finding method in the field of medical

acupuncture, mainly from the overall recognition of

human posture and limbs, and then local acupoint

positioning, and obtained preliminary research

results (Wang 2020). In 2020, in order to improve

the accuracy of human acupoint recognition and

shorten the time of human acupoint recognition, Fu

Yangyang and Gao Zhiyu used Matlab 9.0 digital

image processing technology, gradient descent, and

sliding window hybrid algorithm to identify and

localize human acupoints. This method can

effectively improve the efficiency of human body

acupoint positioning (Fu & Gao 2020).

2.2 Orientation based on Algorithm

Direction

The acupoint location using algorithm mainly relies

on the learning function of the neural network to

predict the acupoint coordinates and realize the

function of automatically finding acupuncture

points. Wang Hongwei proposed an automatic

acupoint positioning method based on the fusion of

CMAC network and Q learning algorithm for human

foot acupuncture and acupoint positioning skills

learning (Wang 2013). More research relies on BP

neural network. Based on the learning function of

BP network in neural network theory, Du Guangyue

et al. took several acupoint coordinate data as

samples, and used MATLAB for data programming

training to realize the function of massage robots in

human acupoint search (Du, Lu & Zhang 2011).

Zhang Qiuyun optimized it and proposed a point

coordinate prediction method based on a genetic

algorithm to optimize BP neural network, which

made up for some shortcomings of BP neural

network. From the simulation results, it can be seen

that the learning speed of this method is faster and

faster. The accuracy is high and the prediction effect

is better (Zhang, Zhang & Li 2017). Yang

Xiangping and Wu Yudan designed a prediction

model of acupoint relative coordinates based on

particle swarm optimization optimized BP neural

network (PSO-BP), which combined with ARM to

form a system that can be used for human acupoint

positioning. The results show that the system can

predict the location of acupoints well (Yang & Wu

2018).

ICHIH 2022 - International Conference on Health Big Data and Intelligent Healthcare

514

Figure 2: Neural network three-layer topology (Yang & Wu 2018).

2.3 Positioning based on Electrical

Impedance Characteristics of

Acupoints

In terms of electrical properties, acupoints have low

electrical resistance compared to non-acupoint

tissues (Chen 2010). In 2010, according to the low

impedance characteristics of acupuncture points,

Chen Zhengliang designed and manufactured an

integrated dot matrix measurement electrode based

on a flexible circuit board, stimulated the human

body by excitation current, and compared the results

to determine the spatial location of acupuncture

points (Chen 2010). Yang Xiangping and Xia

Zhiyuan in 2018, by improving the two-electrode

method of acupoint resistance detection technology,

combined with the mechanical structure design and

embedded control technology, designed a new

embedded acupoint recognition device (Yang & Xia

2018). In 2018, based on the two-electrode method

of acupoint resistance detection technology, Yang

Xiangping and Xia Zhiyuan designed a new

embedded acupoint recognition device, which

combined the mechanical structure design with the

embedded control technology.

2.4 Other Methods

The use of many other methods also provides a lot

of new ideas for the location of acupuncture points.

In 2013, Gao Dongwen, Xiao Husheng and others

conducted a study on acupuncture positioning of

pork using three-dimensional and two-dimensional

high-frequency ultrasound technology. The results

show that high-frequency ultrasound guidance is

accurate, intuitive, dynamic, and accurate for the

anatomical positioning of the acupoints of the living

body. The feature of low price can be used as the

conclusion of the standard of living body acupoint

positioning (Gao, Xiao, Xu, Zhang, Xu, Yin &

Wang 2013).

In 2018, Dong Shihui and Wang Xu were based

on the intelligent functions of ABB robots.

According to the characteristics of the distribution of

acupoints in the main parts of the human body, the

human acupoints were classified and processed.

According to the spatial movement instructions of

their manipulators, they designed their coordinate

displacements on the surface of the human body.

Localization, so as to identify the acupoints of the

meridians of the human body (Dong & Wang 2018).

In 2019, Kun-Chan Lan and Gerhard Litscher

proposed a system that uses augmented reality to

locate acupuncture points. Compared with

traditional acupuncture point probe devices that

work by measuring skin impedance, this system

does not require any additional hardware, but only

based on software. In the case of mild symptoms,

through this system, patients can quickly find the

corresponding acupuncture points for acupuncture or

massage (Lan, & Litscher 2019).

Research of Acupoint Location Methods in Traditional Chinese Medicine

515

3 COMPARISON AND OUTLOOK

Algorithm-based acupoint positioning mainly uses

the learning function of neural network theory.

Taking the coordinate data of several acupoints on

the body as a sample, other acupoint coordinates are

derived through training to realize the function of

finding acupuncture points. Because the positioning

of a neural network depends on physiological

characteristics, and some acupuncture points only

rely on the physiological structure of the body,

because the proportions of the human body are

different, it is impossible to find the characteristics

according to the image based on the principle of

vision, so it is difficult to locate by visual means,

using neural network The characteristics of this type

of acupuncture points can be realized. However, the

algorithm-based approach has strict requirements on

data samples, and the number of samples needs to be

large. Acupuncture point location based on vision

method uses visual measurement combined with

image processing to manually mark acupoint

location. In actual massage or acupuncture, it is

impossible for the patient to remain motionless. If

the patient moves, the coordinates of the acupoints

need to be changed. The real-time image collection

can be achieved based on vision, so as to achieve the

effect of accurately positioning the acupuncture

points. However, because the entire system is too

large, the fast real-time performance of positioning

is reduced (Zhang, Zhang & Li 2017). Based on the

electrical impedance characteristics of acupoints, the

acupoints and their surrounding non-acupoint tissues

have low electrical resistance and high potential in

electrical properties to locate acupoints. This method

is very safe and accurate, so it is often used as a way

to verify the accuracy of data in algorithms and

visual methods. Like many other methods, it is also

difficult to popularize because of the high equipment

requirements.

Many acupuncture point positioning techniques

now have a strong purpose. For example, according

to needs, different methods of acupuncture point

positioning on the face and feet are selected.

Because different methods have different

applicability to different parts of acupuncture points,

before conducting research, you should clarify your

needs and choose the appropriate method. In

general, vision-based acupuncture point positioning

methods are more common, with more application

scenarios and more complex. The future research

direction can be based on the acupuncture point

positioning in the visual direction, combined with

other methods to make up for the shortcomings, to

improve the accuracy and respond to various

scenarios and needs.

4 CONCLUSIONS

This report introduces the background of the

combination of TCM acupoint location and modern

technology, and summarizes and categorizes the

current technology research. The current method of

acupoint location is developing steadily, and more

and more people are trying to use new methods to

improve and improve the research on accurate

acupoint positioning system. After a simple analysis

of the advantages and disadvantages of each method,

it can be seen that each method has its own special

applications. Among them, the vision-based method

is more comprehensive. In the future, the vision

method can be combined with other methods to

solve specific problems and make up for the

shortcomings, so as to achieve more accurate results.

ACKNOWLEDGMENTS

This paper was completed with the patient help of

the teaching assistant Zhu Jiahui and the thesis

teacher Han Min. The assistant teacher helped me

choose the topic of the thesis from the very

beginning, and helped me find a lot of learning

resources. The thesis teacher helped me revise the

thesis. The conception of the paper, and helped me

correct the mistakes. Here, I would like to express

my sincere thanks to the teachers.

REFERENCES

Chang Menglong, (2017). Research on facial acupoint

location method based on feature point location

algorithm (Master's thesis, Beijing University of

Technology).

Chen Zhengliang, (2010). Human meridian electrical

impedance detection technology and preliminary

experimental research (Master's thesis, Zhejiang

University).

Dong Shihui & Wang Xu, (2018). Human body meridian

and acupoint recognition based on ABB robot.

Modern Manufacturing Technology and Equipment

(02), 187-188+190.

doi:10.16107/j.cnki.mmte.2018.0288.

Du Guangyue, Lu Shouyin & Zhang Huakai, (2011). The

design of human body acupoint finding function of

traditional Chinese medicine massage robot. Bulletin

ICHIH 2022 - International Conference on Health Big Data and Intelligent Healthcare

516

of Science and Technology (05), 637-640.

doi:10.13774/j.cnki.kjtb.2011.05.008.

Fu Yangyang & Gao Zhiyu, (2020). Research on human

acupoint location based on gradient descent and

sliding window hybrid algorithm. Information on

Traditional Chinese Medicine (02), 106-110.

doi:10.19656/j.cnki.1002-2406.200054 .

Gao Dongwen, Xiao Husheng, Xu Zhizhang, Zhang

Aihong, Xu Fang, Yin Haoqiang... & Wang Qi,

(2013). Feasibility study of three-dimensional and

two-dimensional high-frequency ultrasound

technology in acupuncture point positioning. China

Journal of Integrated Traditional Chinese and Western

Medicine Imaging (03), 231-234.

doi:CNKI:SUN:JHYX.0.2013-03-002.

Gao, H. B., Lu, S. Y., & Wang, T., (2014). A Dual-Arm

Cooperating Physiotherapy Service Robot Based on

Visual Position. In Applied Mechanics and Materials

(Vol. 459, pp. 222-227). Trans Tech Publications Ltd.

Lan, K. C., & Litscher, G., (2019). Robot-controlled

acupuncture—an innovative step towards

modernization of the ancient traditional medical

treatment method. Medicines, 6(3), 87.

Lin, S., & Yi, P., (2019, July). Human acupoint

positioning system based on binocular vision. In IOP

Conference Series: Materials Science and Engineering

(Vol. 569, No. 4, p. 042029). IOP Publishing.

Ma Zhewen & Yu Haoguang, (2010). Massage robot

acupoint tracking system based on visual positioning.

Robot Technology and Application (06), 33-35. doi:

CNKI: SUN: JIQI.0.2010-06-011.

Wang Cong, (2020). Research on the method of laser

acupuncture and moxibustion robot visual acupoint

finding (Master's thesis, Beijing University of Posts

and Telecommunications).

Wang Hongwei, (2013). Research on skill learning and

fusion method of foot massage robot (Master's thesis,

Shenyang Ligong University).

Yang Xiangping & Wu Yudan, (2018). Design of human

acupoint positioning system based on PSO-BP neural

network. Application of Electronic Technology (09),

75-78. doi:10.16157/j.issn.0258-7998.180900.

Yang Xiangping & Xia Zhiyuan, (2018). Design of a new

embedded acupoint recognition device. Chinese

Acupuncture (06), 675-678.

doi:10.13703/j.0255-2930.2018.06.031.

Zhang Huakai, (2012). Vision-based TCM massage robot

acupoint location and tracking system (Master's thesis,

Shandong Jianzhu University).

Zhang Qiuyun, Zhang Ying & Li Chen, (2017).

Application of genetic algorithm to optimize BP

neural network in traditional Chinese medicine

massage robot. Applied Science and Technology (02),

73-77.

Zhao Yang, Zhang Tao & Lian Qiusheng, (2012). Bell

facial paralysis acupuncture point location algorithm

based on infrared facial image. Small microcomputer

system (07), 1613-1619.

doi:CNKI:SUN:XXWX.0.2012-07- 044.

Research of Acupoint Location Methods in Traditional Chinese Medicine

517