CTM Tongue Image Consulting System based on Deep Learning
Technology
FANG JINGXUAN1,2, LIU FEI2, FANG XIANG2*, SU LINGTAO2
1Yale University,
New Haven, CT 06520,
USA
2Lanjiang Qinghu (Beijing) Technology Co., Ltd,
Beijing 100084,
CHINA
* Corresponding Author
Abstract: - Internet is an important development step in information times. With the tide of Internet
development, internet plus health will become a trend of new times. The Chinese Traditional Medicine (CTM)
tongue image consulting system based on deep learning technology has created a more perfect and intelligent
personal health management system by connecting smart devices to mobile platforms. The system will serve
customers perfectly through health food therapy, medical consultation, etc.
Key-Words: - Chinese Traditional Medicine; tongue diagnosis; characteristic constitution; health monitoring;
dietary recommendations; deep learning; health management.
Received: March 2, 2023. Revised: October 21, 2023. Accepted: November 11, 2023. Published: January 26, 2024.
1 Introduction
In more than 1000 years, CTM diagnoses by tongue
image. [1], based on modern information
technology, especially high-end image processing
technique and machine learning methods,
personalized scientific diagnosis can be made by
utilizing tongue image. Thus, more ordinary people
can enjoy the health services brought about by
tongue image consultation. So the development of a
TCM tongue health consulting system based on
deep learning technology comes into being.
2 CTM Tongue Image Consulting
System
2.1 The Function of the CTM Tongue Image
Consulting System
The system has several functions as below. First,
tongue diagnosis is a very important step in TCM
observation. The doctor can understand the situation
of the patient through observation of the tongue.
The software of tongue diagnosis will integrate well
with the information technology method. Thus, the
diagnosis will be more comprehensive and
objective. The health of customer will be protected
effectively by setting up real-time updated profile
through long time auditing; second, APP, website,
Wechat, synthetic management of third-party
platform and effective building of personalized
health management system can truly meet the needs
of modern people to use anytime, anywhere. At the
same time, it has comprehensively realized
scientific, information, and entertainment; Thirdly,
the combination of health monitoring & dietary
recommendation and effective connection with the
medical institution can make the function more
comprehensive and the service more personalized.
2.2 The Product and Service of CTM Tongue
Image Consulting System
2.2.1 Background of the Product and Service
With the rapid popularization of mobile Internet
technology and the widely used smart phone, more
and more functions can be realized based on smart
mobile platform such as various APPs on Wechat
public platform. At the same time, due to people's
attention to health and the change of lifestyle,
various systems related to individual health records,
testing and diagnosis based on smart mobile
platform come into being. For various reasons
almost all of these systems or application programs
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
1
Volume 20, 2024
are based on western theories and relevant concepts.
The treasure TCM has not yet been carried forward
on mobile platforms.
Chinese civilization has a long history. TCM
has spread and developed with the process of
history. It has formed a set of solid and unique
TCM theory and clinical application system.
Compared with western medicine, TCM emphasizes
people-oriented. TCM comprehensively understands
various physical features of the human body and
systematically grasps the relevant information of the
occurrence and development. Thus TCM can guide
clinical treatment and health care. TCM forms a
multilevel and multi-angle complementarity with
the diagnosis and treatment system of western
medicine, [2], [3].
2.2.2 Tight Relationship with the Internet
Emerging industries formed by the "Internet +" and
integration and upgrading with traditional industries
will be the important economic growth point of
expanding consumption and promoting
employment. From the perspective of industrial
structure, "Internet +" drives the upgrading of
industrial structure by forming a great economic
vitality and wealth effect, rapid development of
emerging information industry, and fusion
optimization of internet technology and traditional
industries. Internet can reduce the transmission cost
of information to a very low level. After receiving a
large amount of information, the degree of
information asymmetry between consumers and
producers has decreased significantly. The choice
range of residents' consumption has been greatly
expanded. Residents can use social mobile networks
to unite. They begin to participate in all commercial
links of production and consumption after solving
the problems of dispersion and isolation to
interconnection. They can truly become the leading
force of production and consumption. [4], so the
system is developed based on the Internet.
2.3 The Product Line of the CTM Tongue
Image Consulting System
2.3.1 Tongue Image APP
2.3.1.1 Basic Principle
The diagnostic technology of traditional TCM is
well known around the world. The four diagnostic
methods including looking, listening, questioning,
and feeling pulse have been widely used for
thousands of years. Tongue diagnosis is an
important step to look for diagnosis. It is an
effective diagnosis method for the doctor to
understand the pathological changes in patients by
observation of the tongue. According to TCM
theory, the tongue is a mirror of the body's internal
physiology. The tongue can fully and flexibly
reflect the internal changes of the the internal organs
of human viscera. Therefore, tongue diagnosis has
become one of the important tools of TCM
diagnosis, [5].
Western medical research has also noticed that
tongue images can reflect physical conditions. The
British used a three-color colorimeter to check the
tongue of the patient in the 1970s. They believe that
any tongue containing blue is abnormal. However,
the diagnosis of the tongue has always been
analyzed and judged by subjective visual
observation. And doctors use language to
summarize the disease, lack of automatic and
quantitative objective measurement. Tongue
diagnosis is easily affected by external conditions
such as light, angle, and other factors, etc. Also, the
interference of doctors' human factors (such as
knowledge level, clinical diagnosis experience, etc.)
can affect tongue diagnosis. All these restrict the
wider application and modernization of tongue
diagnosis. With the continuous development and
technical promotion of the modernization of TCM
in recent years, people have higher requirements for
the quantification and objectification of tongue
diagnosis. The objective study of tongue image has
started since the 1990s. Researchers have made
many valuable explorations on extraction of tongue
image features and tongue image analysis, [6].
With rapid development of artificial intelligence
(AI) in past a few years, more and more AI
technologies combine with traditional applications.
Like tongue diagnosis, TCM doctors can examine
diseases of patients through tongue color, shape, and
texture. Many AI specialists want to simulate the
process that TCM doctors’ examination based on AI
technologies. Currently, most methods are focused
on comparison of the tongue photos themselves, [7].
These research methods propose higher
requirements for taking tongue photos.
On methodology our technology achieves a
breakthrough. The deep learning technology of the
convolutional neural network(CNN) is applied.
Through PCA Net model, tongue photos can be
analyzed starting from the underlying pixels for
feature purification and analysis. Because taking
tongue photos comes from mobile phone, the
method provides greater redundancy on different
mobile brand, photo color, taking angles etc. The
research method is pioneering.
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
2
Volume 20, 2024
The CTM tongue image consulting system
innovatively applies deep learning technology to
analyze and judge tongue photos. It is shown in
Figure 1, [8].
The main working mode of tongue image
diagnosis software is to take photos of users'
tongues through smart phones. Then the health of
the user can be judged through the softwares
analysis of photos. Users can be given
corresponding suggestions on health care, lifestyle,
and dietary choices by diagnosis and analysis of
tongue, and preliminary judgment and screening of
diseases. At the same time, users can choose to
answer some questions based on the theory and
practice of TCM to further improve the accuracy of
disease judgment and the pertinence of the
corresponding suggestions.
Fig. 1: Structure of the CTM tongue image
consulting system
2.3.1.2 Health Monitoring and Evaluation
Dynamic assessment of health status can be
obtained through recording the tongue image for a
long time and comparing the objective indexes
(color, texture, shape, etc.). For example, in patients
with acute ischemic stroke, the tongues are often red
and crooked, with greasy white moss [9]; red
tongues with greasy white moss are also observed in
patients with Helicobacter pylori infection [10].
Tongue shape assessment is also an important
component of tongue diagnosis. Such geometrical
shape information includes thickness, size, cracks,
and tooth marks. In patients with primary Sjögren
syndrome, the tongues are typically thin, reddish-
red, nonmossy or cracked, [11]. In patients with
primary insomnia, the tongues are predominantly
red and fat, with yellow and white greasy moss, and
tooth marks, [12] etc. The normal tongue is an
ellipse tongue, but there are also six other classes of
tongue shapes: square, rectangular, round, acute
triangular, obtusely triangular, and hammer, [13].
The user’s tongue can be better modeled through the
multiple use of the tongue image diagnosis software
by the same user. To find subtle but medically
meaningful changes in the tongue, it is necessary to
reduce the interference of various random errors.
Thus, it can effectively assist in assessing changes
in health status or physical recovery after treatment.
The process is shown in Figure 2.
Fig. 2: The change of tongue image
Figure 2 shows a comparison of the images of
the tongue of the same user at different times using
the software for detecting the image of the tongue.
The right side of Figure 2 is the release change of
the main seven features (red is the previous feature
distribution, and blue is the latter). It can be seen
from the picture that there are significant changes in
body state in the image above. However, there are
no big changes in the image low.
At the technical level of AI, an end-to-end
network called Tongue U-Net (TU-Net) can be
proposed, which combines the classical U-Net
structure with the squeeze-and-excitation (SE)
block, the dense atrous convolution (DAC) block,
and the residual multikernel pooling (RMP) block.
The method can be applied to a tongue dataset with
300 images, and it performed better than other
segmentation methods such asU-Net, Attention U-
Net, [14].
A tongue segmentation method using a
multitask, end-to-end learning model named
TongueNet, for supervised deep CNN training is
also presented. A feature pyramid network based on
the designed context-aware residual blocks to
extract multiscale tongue features is used. The
region of interest (ROI) of the feature maps was also
used for finer localization and segmentation. [15], in
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
3
Volume 20, 2024
a small-scale tongue dataset, they also applied U-
Net for fast tongue segmentation, and achieved the
highest accuracy of 98.45% and consumed 0.267 s
per picture on average, [16].
Table 1. Different tongue images versus health characteristics
Name
Tongue Image
Corresponding Symptoms
Normal
Light red and bright
Moist tongue
Moderate in size, soft &
flexible
The tongue coating is
even, thin, white and
moist
-----
Dentate
tongue
The edge of the tongue
looks like a bite mark
like a tooth
The tip of the tongue has
a papillary protrusion.
Poor renal function leads to the problem of
early aging of the kidney. There will be lower
limb edema and obesity.
The kidney reflects the main bone. People
with this tongue image are prone to low back
pain
Beware of the risk of latent glaucoma
Poor liver and gallbladder metabolism and
easy fatigue
Strawberry
tongue
Red dots can be seen on the
tongue image like strawberry
The function of the pancreas and spleen is
poor, and there are many obesity problems.
It is easy to have bad breath, poor skin
quality, and poor physical strength.
If you do not pay attention to maintenance,
be careful with diabetes
Blood
stasis
tongue
There is no tongue coating and
the tongue is dark red or
purplish red
There may be something wrong with the
circulatory system. Blood is acidic and the
metabolism is poor. Free radicals in the body
are too high, increasing the risk of cancer,
hypertension and stroke.
The heart is sluggish and the thinking is not
agile
Body
poison
tongue
The thick coating of the
tongue coating is clearly
visible
The toxin content in the body is too high and
necessary to do environmental protection in the
body & clean up some waste.
People with this tongue image have physical
problems, mainly the function of the lungs and
intestines. There will be constipation or
diarrhea
Pressure
tongue
After sticking the tongue, the
uncontrollable shaking will
occur on the surface of the
tongue will happen
Problem with autonomic nerves is indicated.
In severe cases, the hands will shake and the
body will also have the problem of shaking.
Having the tendency to neuroticism, stress,
insomnia, irritability, and depression
The problem of overwork death should be
paid attention to
Fissured
tongue
Longitudinal or transverse
cracks can be clearly seen on
the tongue surface
It indicates that there is something wrong
with body metabolism. There can be signs of
mineral deficiency or an imbalance in hormone
secretion
Women will have trouble with monthly affairs
and hair loss
Men fear signs of kidney failure
Sublingual
varicosis
Vascular sclerosis above the
neck, memory loss, migraine,
stroke insomnia, Alzheimer's
disease, neuroticism
Caused by trauma
Qi deficiency, qi depression, bad mood,
anger, and pressure may lead to congestion in
the body
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
4
Volume 20, 2024
Additionally, using the U-Net structure,
exploring to extract the ultrasound tongue contour
using the U-Net, and superior performance has been
obtained, [17].
The constitution and morphology related to
tongue image is listed in Table 1. [18], [19], TCM 9
tongue model classification can lead to different
health statuses. Through different classification,
more details could be discussed. For example,
tongue crack segmentation is also an essential
component of computer-aided diagnosis applied.
Tongue cracks refer to fissures with different
depths, and [7]: Digital tongue image analyzes for
health assessment 7shapes on the tongue’s surface
and muscle layer. In most cases, it can be seen as a
changeful curve structure on the surface of the
tongue, and its depth is determined by the severity
of the atrophy and lesions of the tongue mucosa.
Also, the quantitative value of the fissured tongue
reflects the health condition of the internal organs,
[20].
2.3.1.3 Dietary Recommendation System
According to the health situation reflected in the
tongue image of customers, diet recipes that can
improve your health can be recommended. The
system needs to consider several things. Firstly,
according to the results of daily health status,
personalized diet recommendations (diet therapy)
could be provided; Secondly, according to the
principle of health preservation of TCM,
geographical location, characteristics of solar terms,
etc., dishes could be arranged reasonably. Personal
order or comprehensive preparation of dishes will
be taken into account considering the image of all
diners will be took into consideration;
Thirdly, regarding chronic diseases, dietary
recommendations will be made based on
recommended guidelines combined with theory of
TCM. It focuses on cancer, hypertension, diabetes,
and digestive system diseases; Fourthly, dietary
recommendations will also be done base on main
manifestations of sub-health, including fatigue,
insomnia, tension, etc. The function of
monitoring dietary calories is also developed at the
same time. Automatic calorie estimation based on
recipes or photos of dishes is included. According to
the user’s selected dietary scheme (NIH-
recommended DASH guideline recommended by
NIH or popular light fasting banting), the user’s diet
can be guided and supervised. The process is shown
in Figure 3, [21].
The example shows ordering at a restaurant
based on tongue image. The customer can get the
hotel menu information by scanning QR code.
Depending on the situation of the tongue image,
dishes or side dishes will be automatically
recommended through tongue image diagnosis
software.
Fig. 3: Dietary recommendation system
2.3.2 Wechat Public Platform for TCM health
Wechat has become an integral part of people's
lives. Wechat has more than 1.2 billion users
because of the advantages including fast speed, free
expense, cross-platform, flexible, intelligent, etc.
Building a Wechat platform will not only
accumulate users and spread culture & information
but will also become the service interface of product
marketing, [22].
2.3.2.1 Release of TCM Health Information
Relevant seasonal and regional health knowledge
can be published regularly based on season and
region on the platform. General knowledge of health
care and family diet and medication health care
methods can also be introduced.
2.3.2.2 Online Classroom of TCM
The platform can regularly publish micro videos
related to TCM health care including TCM diet
regimen, TCM exercise regimen, TCM health
preservation etc. Online small-scale micro-
classroom teaching may be conducted by contacting
the relevant specialist. The ways can include online
real-time teaching by doctors and online real-time
communication.
2.3.2.3 TCM Online Health Consultation
Users can directly upload tongue image information
and send it to the background. The platform uses a
tongue image analysis program to automatically
analyze tongue image types and give diet
suggestions. The communication ways also include
user feedback questionnaire information and asking
questions to doctors, etc. The users can enjoy the
service of online doctor professional health
assessment, disease risk assessment, diet and
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
5
Volume 20, 2024
exercise suggestions, etc. User questions could also
be answered at the same time.
2.3.2.4 Health Micro-Store
The platform can become a platform for
professionals and buyers to communicate. Local
TCM clinics can not only sell relevant health care
products online, including health care equipment,
dietary supplement, and health books, etc. but also
bring online customers offline and provide real-time
diagnosis.
2.3.3 TCM Health Preservation Website
The developed tongue image APP, Wechat public
platform, and portal to develop will realize unified
background management and use the same database.
The website focuses on the health status of
individuals and families, customized private health
account for users and dynamic monitoring for
individual users. The main functions are shown
below.
2.3.3.1 Personal Health Management System
Users can log in to their own account on the website
and get relevant health information including long-
term tongue image analysis, long-term health
change trend of scientific system, health care
feedback, prediction & guideline received regularly,
and realizing the real-time update of users' health
records, etc. The users can look for people with
similar health and form their own circle of friends
through website.
2.3.3.2 TCM health Care Course
The website regularly publishes video courses on
TCM health preservation including TCM diet
regimen, TCM exercise regimen and TCM physical
health preservation etc. The website will realize
real-time communication between doctors and users
through the way of inviting experts and doctors to
give lectures online in real time on a regular basis.
2.3.3.3 TCM Health Care Information
Users can utilize the website easily and quickly to
obtain valuable information such as popular
knowledge including common sense of TCM, diet
health care of TCM and sports health care of TCM,
etc. The website will provide users with the main
information and communication channels of the
relevant medical institution and expert physicians
through an effective connection with relevant
medical institution.
3 Summary
The tongue image can be handled to understand the
health status of customers and give the
corresponding suggestions through the CTM tongue
image consulting system based on deep learning
technology.
3.1 Big Data System for Health Management
The big data system of health management is to
research and develop a personalized medical cloud
platform information system that integrates all data.
It can implement automatic health monitoring,
health guidance, and medical consultation. It can
provide customers with all-round health-related
information (diet, exercise, medicine, etc.). It can
also provide complete health detection, prevention
plan, and relevant health knowledge. The system is
shown in Figure 4.
The system has a functional interface that can
monitor dietary calories and physical exercise,
including setting and achieving fitness goals,
communicating with friends and doctors, learning,
etc.
3.2 Big Data System of Tongue Image
The big data system of tongue image can collect a
large number of tongue images and related health
information. The system searches for tongue images
to high similarity with users by means of big data
search. Then a more comprehensive physical
condition assessment can be performed for the
customers based on existing tongue images and their
diagnostic information. The system is shown in
Figure 5.
The customer tongue image can be searched for
in database. The health situation of the customer can
be evaluated using a similar tongue image. Health
advice can be provided and the information can be
stored in the electronic medical record.
Tongue classification tasks can be divided into
two main categories. The first is a classification
based on the clinical characteristics of the tongue,
such as the color of the tongue coating, the
distribution of the tongue coating, tooth marks,
tongue cracks, and tongue shape. The second is to
classify diseases that can be reflected in the clinical
attributes of the tongue, such as gastritis,
appendicitis, and breast cancer. In early research,
machine learning methods, such as support vector
machines were applied to classify tongue images.
Before training these classifiers, feature extraction is
an important step.
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
6
Volume 20, 2024
Fig. 4: Big data system for health management
Fig. 5: Big data system of tongue image
Many studies have tried to use different
methods to extract features such as the color,
texture, and shape to improve the accuracy of the
classifier, and some studies have optimized the
structure of the classifier, which greatly improves
the accuracy of tongue image classification. With
the introduction of deep learning, CNNs have
significantly increased the number of parameters in
the classifier and subsequently improved the
accuracy of the classifier. After that, some [7],
Digital tongue image analyses for health assessment
21 more complex networks such as VGG, Resnet,
Inception were proposed, and the performance of
tongue image classification was further improved.
However, more research is needed to further
improve the accuracy of tongue image
classification. In the next few years, tongue
classification also needs significant improvement
for broad applications. Tongue classification and
tongue diagnosis on mobile devices represent a
general trend of TCM-assisted tongue diagnosis, but
the quality of the data on mobile devices is limited.
It requires bigger datasets and better machine
learning models to address these issues. The
computerized tongue diagnosis system provides a
certain level of convenience for both physicians and
patients. Physicians can refer to the results of the
tongue diagnostic system to aid in their diagnostic
process; patients can upload their tongue pictures
via their cell phones to obtain an assessment of their
health status. Although many studies of
computerized TCM diagnostic systems have been
published, limited by hardware technologies and
computational methods, almost all of these
computational diagnostic systems can only simulate
one of the four diagnostic methods in TCM to assist
in disease detection. According to the four
diagnostic theories TCM, other data (e.g., pulse,
etc.) of the patient are also necessary to make a
more accurate assessment of health status.
Combining tongue diagnosis with other data, such
as pulse diagnosis, is an inevitable trend in
computerized TCM diagnosis.
In summary, with the gradual and wide
application of CTM tongue image consulting
system, the health level can be improved really
through the gradual transformation from the mode
of passive treatment to the mode of active
management and health prevention.
References:
[1] Yellow Emperor's Internal Classic Su Wen
proofreading and interpretation (Volume I
and II), (Second Edition), Beijing: People’s
Medical Publishing House, 2009
[2] Zhou Yunan. Correct understanding of
traditional Chinese medicine and Western
Medicine. [J]. Yunnan Journal of Traditional
Chinese Medicine and Materia Medica. 2008,
Vol. 29, No. 5: 81-83
[3] Ding Hong. The differences between TCM
and Western Medicine. [J], Clinical Journal
of Chinese Medicine, 2010. Vol. 2, No. 8:
119-122
[4] Yao Sufang. The role of Internet economy in
China's economic development. [J]. New
Finance World, 2019. Vol. 3: 19-20
[5] Yang Jie, Niu Xin, Xu Yuanjing, Niu
Shudong. Development of digital
informatization of TCM diagnosis, [J].
Archives Traditional Chinese Medicine, 2006.
Vol. 24, No. 5: 810-812
[6] Zhou Peizhuo, Lei Yang, Fan Xinyu.
Application of digital image processing
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
7
Volume 20, 2024
technology in tongue diagnosis of TCM. [J],
Henan TCM. 2020, Vol. 9: 1323-1326
[7] Jiacheng Xie, Congcong Jing, Ziyang Zhang,
Jiatuo Xu, Ye Duan and Dong Xu. Digital
tongue image analyses for health assessment.
Medical Review, DE GRUYTER. 2022: 1-27
[8] Kanawong, RatchadapornObafemi-Ajayi,
TayoTao MaDong XuShao LiYe
Duan. Automated Tongue Feature Extraction
for ZHENG Classification in Traditional
Chinese Medicine. Evidence-Based
Complementary and Alternative Medicine,
2012:912852, 2012
[9] Cao Hyb. Distribution characteristics of TCM
syndrome types inacute ischemic stroke and
correlation with tongue image. Clin JTradit
Chinese Med 2021;33:1312–6.
[10] Chen Y, Yuan H, Hui Y, Zhang XZ, Liu Y.
Distribution of TCM syndrome types and
tongue images in patients with failed
Helicobacter pylori eradication based on
propensity score matching. J Basic Chinese
Med 2021;27:986–9.
[11] Luo J, Zhang L, Chen J, Hu Q, Zhang Y, Tao
Q. Tongue appearances of patients with
primary Sjögren’s syndrome and their
correlations with syndromes. China J
Traditional Chinese Med Pharm
2021;36:3653–6.
[12] Chen H, Xu X, Zhou Y, Hu J. Discussion on
the characteristics and significance of tongue
manifestation in 8260 patients with primary
insomnia. China J Traditional Chinese Med
Pharm 2021;36:2971–3.
[13] Zhang Q, Zhou J, Zhang B. Computational
traditional Chinese medicine diagnosis: a
literature survey. Computer Biology Med
2021;133:104358.
[14] Huang Y, Lai Z, Wang W, editors. TU-Net: a
precise network for tongue segmentation.
Proceedings of the 2020 9th international
conference on computing and pattern
recognition; 2020.
[15] Zhou C, Fan H, Li Z. Tonguenet: accurate
localization and segmentation for tongue
images using deep neural networks. IEEE
Access 2019;7:148779–89.
[16] Zhou J, Zhang Q, Zhang B, Chen X.
TongueNet: a precise and fast tongue
segmentation system using U-Net with a
morphological processing layer. Appl Sci
2019;9:3128.
[17] Zhu J, Styler W, Calloway I. A CNN-based
tool for automatic tongue contour tracking in
ultrasound images. ArXiv preprint
arXiv:190710210 2019.
[18] Zhang Ke, Ding Chenghua, Zheng Shaoyong,
Fang Fang, Ning Lun, Lin Lin, Gao Weiwei,
Chen Xuejiao, Wang Yuchen. Discussion on
the correlation between tongue image and
constitution. [C]. Proceedings of the 13th
Annual Conference of traditional Chinese
medicine diagnosis. 2012: 86-91
[19] Wang Yanhui. Clinical practical tongue
Atlas. [M]. Beijing: Chemical Industry Press,
2012
[20] Zhang H-K, Hu Y-Y, Wang L-J, Zhang W-Q,
Li F-F. Computer identification and
quantification of fissured tongue diagnosis.
2018 IEEE international conference on
bioinformatics and biomedicine (BIBM).
[21] Wang Lu. The world's Healthiest Diet:
DASH. [J]. Food and Health. 2018. Vol. 10
[22] Dong Wenxin. [J]. Public Communication of
Science & Technology. 2021, No.3:153-155.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The authors equally contributed in the present
research, at all stages from the formulation of the
problem to the final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The authors have no conflicts of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on SIGNAL PROCESSING
DOI: 10.37394/232014.2024.20.1
Fang Jingxuan, Liu Fei, Fang Xiang, Su Lingtao
E-ISSN: 2224-3488
8
Volume 20, 2024