Professional Documents
Culture Documents
Syndrome Differentiation and Treatment Algorithm Model in Traditional Chinese Medicine Based On Disease Cause Location Characteristics and Conditions
Syndrome Differentiation and Treatment Algorithm Model in Traditional Chinese Medicine Based On Disease Cause Location Characteristics and Conditions
Received October 20, 2018, accepted November 9, 2018, date of publication November 16, 2018,
date of current version December 18, 2018.
Digital Object Identifier 10.1109/ACCESS.2018.2881535
ABSTRACT Traditional Chinese medicine (TCM) is based on a unique disease diagnosis and treatment
system that has been developed over the last 2,300 years. In the TCM, ‘‘syndrome differentiation and
treatment’’ (SDAT) is a core method for doctors to deal with diseases. This diagnostic and therapeutic
technique that infer the occurrence and the development of diseases by observing symptoms as a whole,
not only has its own uniqueness but also has been recognized by the public in oriented medical fields for
its clinical efficacy. With recent developments in computer science, the Internet, big data, and artificial
intelligence, a study based on the SDAT algorithm has aroused much attention. This paper encompasses
three stages spanning 30 years to accomplish the following: 1) the TCM data and the modern SDAT system
were collated and summarized based on 35,706 reference data on the TCM, starting from the syndrome
differentiation of four aspects, such as the cause, location, characteristics, and conditions of the disease
(CLCC), we constructed a quantitative model of the TCM SDAT regarding the CLCC of the disease,
collected the symptom information on the diagnosed subject, and transferred them to the SDAT assistant
algorithm for calculation and analysis, to determine the CLCC, Based on the therapy recommended by
the differentiation results in the knowledge base and the prescription and traditional Chinese medicines
recommended by the therapy, any stage of all diseases could determine a syndrome type by differentiating
the CLCC, we constructed the basic SDAT algorithm integrating theory, method, prescription, and medicine
and realized the calculability in the TCM diagnosis and treatment process; 2) based on the SDAT algorithm,
we developed the TCM doctor’s workstation software and introduced it to more than 80 TCM institutions
in Sichuan province, China, we collated a large-scale trove of samples of the TCM data platform that
was established with more than 2.9 million TCM electronic medical records (EMRs) and reference data,
and had the compliance tested and algorithm verified on the 9,300 EMRs of the common diseases in the
TCM; and 3) based on the dimension reduction and degree elevation optimization of the technology with
a directed graph to the basic algorithm, the algorithm complexity was reduced and the accuracy of the
algorithm was improved. It was demonstrated that the coincidence rate of the basic model was 80.47%
and the basic coincidence rate was 96.19%. After optimizing the basic algorithm (for example, for gastric
abscess), the coincidence rate increased by 7.04%. The test results demonstrated the efficacy of the model
study. This model realized a computable SDAT to specify and assist in the differentiation diagnosis and in
the treatment processes of the TCM and improve the service quality of the TCM diagnosis and treatment.
INDEX TERMS Cause, location, characteristics, conditions of diseases, syndrome differentiation and
treatment, algorithm, traditional Chinese medicine.
2169-3536 2018 IEEE. Translations and content mining are permitted for academic research only.
VOLUME 6, 2018 Personal use is also permitted, but republication/redistribution requires IEEE permission. 71801
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
J. Chen et al.: SDAT Algorithm Model in TCM Based on Disease CLCC
TABLE 1. 46-bit quantitative code for the cause, location, characteristics, and conditions of disease.
1) DISEASE LOCATION OF THE SYNDROME (13 BITS) and bowel, but no effect on the heart, liver, gallbladder, etc.
Heart, liver, gallbladder, spleen, stomach, bowel, lung, kid- As shown in Table 1, a total of 1,923 symptom codes are put
ney, bladder, bladder, uterus, thoroughfare and concept, in the system knowledge base.
essence chamber and surface. It should be noted that the Method 3: We use quantitative code (D) of TCM efficacy,
disease location of the syndrome is specific to the TCM which is a numerical string comprising a 46-bit decimal
medication rather than an exact anatomic site. number (as shown in Table 2, in which the 46-bit quantitative
code corresponding to the medicinal materials is listed).
2) DISEASE CAUSE AND CHARACTERISTICS OF THE It is assumed that the quality of the medicinal material and
SYNDROME (33 BITS) the technological process of the preparation are controllable,
Fu-organ, irregularity, heat, heat deficiency, blood heat, body and the 46-bit quantitative code of TCM efficacy can be con-
fluid deficiency, toxicity, pus, accumulation, cold, heatstroke, structed according to the quantification for the TCM efficacy
deficiency of vital energy, blood deficiency, yin deficiency, qi in the known ingredients of TCM. A total of 491 quantitative
instability, vexation, coma (abnormal consciousness), blood codes of medication efficacy are completed in the system
stasis, pulse obstruction, bleeding, qi stagnation, cough, knowledge base:
wind, stirring wind, dryness, dampness, water, sputum, diet,
D = d1 d2 . . . . . . dn−1 dn , dn ∈ {0, 1, 2, 3}, n = 46. (2)
and parasitosis.
B. MATHEMATICAL MODEL OF SDAT
3) MILD, MODERATE, AND SEVERE CONDITIONS
Method 4: We use B = (O, S, C, T , P, G, Fj∈[1,8] ) as the
Take the numerical value as the basis for determining the
TCM doctor’s workstation, where B indicates TCM doctor’s
severity of the conditions upon assignment for the numeric
workstation; O indicates the patient’s chief complaint col-
coding of each symptom.
lected by the application system; S indicates an assemblage
Method 2: We propose to use quantitative code (V ) for
of symptom names, covering all the patient’s symptoms;
CLCC regarding the TCM symptoms, which is a numerical
i indicates the number of symptoms; and Si ⇔ V si indicates
string composed of 46 decimal numbers as shown in Table 1:
the unique 46-bit quantitative code (V si ) for the CLCC corre-
V = v1 v2 . . . . . . vn−1 vn , vn ∈ {0, 1, 2, 3}, n = 46. (1) sponding to each symptom name (Si ):
n
Among the numbers, vn indicates the degrees of influ- X
ence of current symptoms and the n-th differentiation ele- S= Si , Si ⇔ V si . (3)
ment, ranging from strong, medium, weak to zero. These are i=1
expressed in the decimal numbers 3, 2, 1, and 0, respectively. C indicates the name of syndrome type, which is
For example, ‘‘bad breath’’ has a mild effect on the stomach comprehensively composed of several symptoms, i.e., the
recombination of symptoms, an equivalent to the syndrome TABLE 3. Total quantitative code of symptoms.
in Western medicine. For instance, fever, aversion to cold,
headache, and floating pulse are all external symptoms and
the syndrome type is characterized by ‘‘wind–cold exterior
symptom.’’
T indicates the therapeutic method. It elects a specific
therapeutic rule after determining the syndrome and clearly
investigating the disease cause and pathogenesis. The method
of treatment is also called a basic therapeutic method, which
summarizes the common features of many specific therapeu-
tic methods, and thus has universal significance in clinical
practice. It encompasses the eight therapeutic methods: per-
spiration, vomiting, descending, moderation, warming, clear- FIGURE 1. Corresponding diagram for the total quantitative code of
ing, tonifying, and digesting. The specific therapeutic method symptoms and differentiation elements.
the therapy and prescription would be determined accord- divided into seven syndrome types including ‘‘hyposomnia of
ingly. Therefore, the recommended therapies and prescrip- liver fire harassing the heart’’, ‘‘hyposomnia of phlegm-heat
tions used for the knowledge base were based on govern- harassing the heart’’, and ‘‘stomach gas estrangement’’. Addi-
ment documents, such as Government Document 9 [2017] tionally, the syndromes in each syndrome type are described
and 18 [2016], State Administration of Traditional Chinese in detail. Take ‘‘hyposomnia of liver fire harassing the heart’’
Medicine of P. R. China; Memo No. 83, 114, 93, 9, 141, 147, as an example. The syndromes are sudden insomnia, short
State Administration of Chinese Medicine and Pharmacol- temper, difficulty in falling asleep, and sleep easily disturbed
ogy, Bureau of Medical Administration [2015–2010]. The by dreams, fullness and distension in chest and hypochon-
algorithm was constructed by the ‘‘13th Five-Year’’ projected drium, proneness to sighing, bitter mouth, dry throat,
teaching materials (10th version), including Chinese Inter- dizziness and distention in the head, hot eyes and tinnitus,
nal Medicine, Gynecology of Traditional Chinese Medicine, constipation and scanty dark urine, red tongue with yel-
Pediatrics of Traditional Chinese Medicine, and Surgery of low fur and pulse string.’’ Each syndrome corresponds to a
Traditional Chinese Medicine. unique therapy. Take ‘‘hyposomnia of phlegm-heat harassing
For example, the disease name ‘‘insomnia’’ in Memo the heart’’ as an example. The therapy is ‘‘soothing liver
No. 148, State Administration of Chinese Medicine and Phar- and purging fire. Each therapy corresponds to one or two
macology, Bureau of Medical Administration [2010] was prescriptions. In this case, the recommended prescriptions
are Huanglian Wendan decoction, including rhizoma pinel- TABLE 8. Units for magnetic properties.
liae preparata, Chen pi, zhu ru, Zhishi, Gardenia jasmi-
noides, goldthread, fuling, polygala root, platycladi seed, and
liquorice.
In conclusion, after determining the ‘‘syndrome type
name’’ with the differentiation results, we can pinpoint a ther-
apy in the knowledge base, with at most two recommended
prescriptions. This is why differentiation is precisely at the
core of disease diagnosis and treatment in TCM. A 46-bit
quantitative code is presented in this paper for the purpose
of differentiation, with reference data combined with clinical
medical cases being introduced later to construct a knowl-
edge base, including an introduction to the corresponding
regulations.
F7 : Function A-custom-action (Fj∈[1,6] ) refers to support
for manual intervention of the algorithm by doctors on their
own. For example, A-custom-action (Refining (S)) indicates
that doctors can extract the primary syndromes based on the
primary syndrome information automatically provided by the
system, to achieve a more accurate primary syndrome by
deleting and adding the primary syndrome information.
F8 : Function Validation:V × P uses the primary syndrome
quantitative code and prescription efficacy to compare the
quantitative code and drug, and verify the anastomosis for
the 46-bit quantitative code regarding the differentiation and
diagnosis and treatment.
C. SDAT ALGORITHM
Input: Patient’s information on four diagnostic methods (O)
Output: Patient’s syndrome type (C), therapy (T ), and
prescription (P).
In Table 8, EMR QTY refers to the total quantity of TCM for epigastric pain, 2,096 cases for dysmenorrhea, 2,639 cases
EMRs generated during the diagnosis and treatment using for diabetes, 1,793 cases for pneumonia, and 520 cases for
the SDAT system; proportion refers to the proportion of eczema. The test was conducted according to the following
TCM doctors that use the platform in a certain hospital to test standards of differentiation:
all TCM doctors in this hospital. The average percentage 1) The result will be accepted if the first two items at
of doctors using the system in Tables 8 and 9 is 63.12%; the disease location are fully consistent. In the case of
the total number of disease entities, symptoms, syndrome deficiency of heart and spleen, it is accepted if there are
types, prescriptions, and traditional Chinese medicines only heart and spleen for the first two superimposed numer-
involves the five tertiary hospitals. ical values (achieved from Function Superposition) at
the primary syndrome site. It is basically accepted if
B. PREPROCESSING OF REFERENCES AND CLINICAL DATA there are heart and spleen in the first four bits. It is
There may be input errors, deficiencies, repetitions, irregular unaccepted if either heart or spleen is absent from the
titles, or other problems for the data that cause a reduction of disease location.
data quality and make it difficult to carry out direct analysis. 2) It is accepted if the first three are consistent in the
Therefore, it is necessary to preprocess the data. We retained cause, characteristics, and conditions of the disease.
the patient information, such as patient ID, admission date, In the case of deficiency of heart and spleen, it is
symptoms, syndrome type, disease entities, disease loca- accepted if there are qi deficiency and blood deficiency
tion, disease characteristics, therapeutic method, and pre- in the first three bits according to the superimposed
scription. Additionally, the information is normalized and numerical values (achieved from Function Superposi-
unified in accordance with Basic Theory Nomenclature of tion) of the cause, characteristics, and conditions of the
Traditional Chinese Medicine (GB/T20348-2006) and Clin- disease. It is basically accepted if there are qi defi-
ical Diagnosis and Treatment of Traditional Chinese ciency and blood deficiency in the individual cause,
Medicine–Syndromes (GB/T16751.2-1997) issued by the characteristics, and conditions of the disease. It is unac-
State Administration of Traditional Chinese Medicine of the cepted if either qi deficiency or blood deficiency is
P. R. China, and data falling short of the national TCM absent.
terminology standards were dealt with by means of amend- 3) It is accepted if the name of the syndrome type is
ment, splitting, combination, and removal of repetitions. consistent. In the case of deficiency of heart and spleen,
After pretreatment, 422 disease entities, 318 syndrome types, it is accepted. It is basically accepted if the name of the
15,105 symptoms, 2,896 prescriptions, and 2,914 drugs were syndrome type is not the deficiency of heart and spleen,
achieved. but the features of the syndrome are consistent accord-
ing to CLCC (i.e., there are heart and spleen in the
IV. TEST ON COMMON CLINICAL DISEASES first two bits of disease location, and qi deficiency and
Five common TCM diseases (epigastric pain, dysmenor- blood deficiency in the first three bits of the CLCC).
rhea, diabetes, pneumonia, and eczema) were selected for It is unaccepted if the syndrome and the CLCC are
testing. Five-star EMR (referring to all contents of the- inconsistent.
ory, method, prescription, and medicine, i.e., chief com- 4) It is accepted if the prescription name (such as gui-
plaint, symptom, diagnosis, therapeutic method, prescription pi decoction) is consistent. It is basically accepted if
name, and TCM name and dosage) were collected from the TCM compositions are consistent but prescription
the platform as test data between December 2015 and name is inconsistent. It is unaccepted if the TCM com-
January 2018. There were 9,300 medical cases: 2,252 cases positions and prescription name are inconsistent.
TABLE 10. Collection condition for the clinical data of second-class hospitals.
According to a comprehensive assessment, it is accepted unaccepted if there is any inconsistency in the above four
if the above four assessments are completely consistent. items.
It is basically accepted if there is no inconsistent item and With the test on the five disease entities data, we achieved
any is basically consistent in the above four items. It is the coincidence rate and basic coincidence rate as shown
in Fig. 4. Figure 4(a) shows the coincidence rate and use existing algorithms and models, and their data, usually
basic coincidence rate of the system in differentiating the derived from the references and the veteran expert’s knowl-
disease location. Figure 4(b) shows the coincidence rate edge, are basically devoid of or rarely include clinical data.
and basic coincidence rate of the system in differentiat- Therefore, regardless of the rules and training, the data alone
ing the cause, characteristics, and conditions of the disease. fail to fully reflect the pathogenesis of the disease, thus it is
Figure 4(c) shows the coincidence rate and basic coincidence difficult to determine some syndrome types in practical appli-
rate recommended by the name of the system syndrome type. cation as a great number of achievements are confined to the
Figure 4(e) shows the coincidence rate and basic coincidence laboratory stage. Therefore, no comparison was conducted
rate of the comprehensive assessment on the system SDAT. with other algorithms in this paper.
The coincidence rate and basic coincidence rate of the whole
system can reach 80.47% and 96.19%, respectively. V. DIMENSION REDUCTION AND DEGREE ELEVATION
Currently, there are expert systems established as per the FOR ALGORITHM OPTIMIZATION
single and veteran expert knowledge and the SDAT system A. DIMENSION REDUCTION OF DIFFERENTIATION
established by neural networks and various data statisti- ELEMENTS
cal methods in the TCM SDAT system, with accuracy Three hundred and four dominant diseases in TCM were
reaching 80–90%. However, the methods more often than not identified by Memo No. 103, State Administration of Chinese
FIGURE 4. 2,252 cases for stomach ache, 2,096 cases for dysmenorrhea, 2,630 cases for diabetes, 2,651 cases for pneumonia, and 520 cases for
eczema were taken as the disease entities for test and analysis in Fig. 5, and have the coincidence rate and basic coincidence rate calculated
according to the rules for differential test and assessment.
Medicine and Pharmacology, Bureau of Medical Adminis- model [39] for a dimension reduction of disease entities to
tration [2012] and Memo No. 119, State Administration of reduce the complexity of the algorithm and the effect of
Chinese Medicine and Pharmacology, Bureau of Medical the uncorrelated differentiation elements of the syndrome
Administration [2012] via issuance; recognized by Memo type.
No. 130, State Administration of Chinese Medicine and
Pharmacology, Bureau of Medical Administration [2012] B. DEGREE ELEVATION FOR THE DIRECTED GRAPH
via issuance, the data platform possessed 205 dominant ANALYSIS ON THE DIFFERENTIATION ELEMENTS
disease data in TCM. The basic algorithm has universal- The degree elevation herein refers to the differentiation ele-
ity and the differentiation element is considered compre- ments constituting a syndrome type under the guidance of
hensively, with a coincidence rate of 80.47%. If further TCM theory. We attempted to use the directed graph to
adaptation to clinical data is required, then the basic algo- describe the relationship among ‘‘symptom - disease cause -
rithm shall be optimized. Therefore, we attempted to analyze disease characteristics’’ in order to form a knowledge graph
the large sample data of the data platform taking the to assist in combination of differentiation elements and to
disease entities as units and applied the differentiation promote the coincidence rate for the degree elevation of the
elements of the CLCC that have a potential variable basic algorithm.
71810 VOLUME 6, 2018
J. Chen et al.: SDAT Algorithm Model in TCM Based on Disease CLCC
TABLE 11. Comparison table before and after dimension reduction of differentiation elements.
Taking epigastric pain as an example, based on the disease, this paper constructed a quantitative model of TCM
2,252 data points achieved from the data platform between SDAT regarding the cause, location, characteristics, and con-
December 2015 and January 2018, the syndrome type ditions of the disease, and collected symptom information
elements of epigastric pain were extracted and 19-bit new on the diagnosed subject and enter it into the SDAT assis-
differentiation elements were achieved by using the CLCC tant algorithm for calculation and analysis to determine the
differentiation elements of dimension reduction in the poten- cause, location, characteristics, and conditions of the disease.
tial variable model shown in Table 10. One hundred forty-two Based on the therapy recommended by the differentiation
clinical data points on epigastric pain were used to carry out results in the knowledge base and the prescription and tradi-
a repeatability demonstration test on the basic algorithm and tional Chinese medicines recommended by the therapy, any
the basic algorithm after ‘‘dimension reduction and degree stage of all diseases could be determined by syndrome type
elevation.’’ After dimension reduction, the coincidence rate by differentiating the disease cause, disease location, dis-
of the basic algorithm increased by 7.04%, to some extent ease characteristics, and disease conditions (i.e., cause, loca-
demonstrating the feasibility of the algorithm optimization, tion, characteristics, and conditions of the disease). Unlike
which provided an idea for further optimization of other other differentiation methods concentrating on certain med-
disease entities. The specific optimization procedure will be ical records, this system, as a universal model, deals with
introduced in a subsequent paper after further verification. various diseases seen in internal medicine, surgery, gyne-
cology, pediatrics, exogenous and internal injuries. In this
VI. CONCLUSIONS model, a computable SDAT was achieved to specify and assist
By studying traditional and modern SDAT systems in TCM, in the differentiation, diagnosis, and treatment processes of
starting from the syndrome differentiation of four aspects TCM, to improve the service quality of TCM diagnosis and
(the cause, location, characteristics, and conditions) of the treatment, and to shorten the time for a TCM doctors to
gather clinical experience. The trial was conducted with real [17] Q.-Y. Jiang, X.-J. Yang, and X.-S. Sun, ‘‘An aided diagnosis model of
medical records collected by the system. The data demon- sub-health based on rough set and fuzzy mathematics: A case of TCM,’’
J. Intell. Fuzzy Syst., vol. 32, no. 6, pp. 4135–4143, 2017.
strated that the algorithm could reach a coincidence rate [18] H.-Y. Wang, ‘‘A novel method for quantitative diagnosis based on decision
of 80.47%, with a basic coincidence rate of 96.49%, which tree in traditional Chinese medicine,’’ in Proc. 8th Int. Conf. Mach. Learn.,
could benefit the diagnosis and treatment at TCM medi- vol. 1, 2009, pp. 344–349.
[19] Z.-X. Xu et al., ‘‘Recognition of TCM syndrome types of cardiovascular
cal institutions. At the same time, algorithm optimization diseases based on support vector machine and artificial neural networks,’’
was discussed for preliminary verification, which provides J. Beijing Univ. Traditional Chin. Med., vol. 34, no. 8, pp. 539–543, 2011.
[20] X. M. Yang et al., ‘‘Data mining-based detection of acupuncture treatment
a further work program in promoting the coincidence rate
on juvenile myopia,’’ J. Traditional Chin. Med., vol. 32, no. 3, pp. 372–376,
of the algorithm. More importantly, we collected the clinical 2012.
medical records and further collection would be conducted [21] C. Zou et al., ‘‘Establishment of clinical research model of traditional
Chinese medicine based on big data of the real world,’’ China J. Traditional
to provide valuable first-hand data for further study on the Chin. Med. Pharmacy, no. 10, pp. 4113–4116, 2016.
algorithm. [22] L. U. Peng et al., ‘‘Development and application of traditional Chinese
medicine inheritance support system,’’ Chin. J. Experim. Traditional Med.
REFERENCES Formulae, vol. 18, no. 9, pp. 1–4, 2012.
[23] Z. Yan et al., ‘‘Analysis on composition principles of prescriptions for
[1] J. Zhu and H. X. Zhu, ‘‘Discussion of TCM ‘syndrome’ and ‘treatment TCM therapy of GERD by using TCM inheritance support platform sys-
based on syndrome differentiation,’’’ China J. Traditional Chin. Med. tem,’’ Liaoning J. Traditional Chin. Med., vol. 44, no. 3, pp. 456–459,
Pharmacy, no. 1, pp. 21–24, 2017. 2017.
[2] W. Xu, L. Zhang, Y. Huang, Q. Yang, H. Xiao, and D. Zhang, ‘‘Discrim- [24] J. He, X. Cao, and Y. Hu, ‘‘Development of the Cardiac failure expert
ination of type 2 diabetes mellitus corresponding to different traditional system for Chinese medicine diagnosis based on database,’’ in Proc. IEEE
Chinese medicine syndromes based on plasma fatty acid profiles and Int. Conf. Bioinform. Biomed., Dec. 2014, pp. 61–66.
chemometric methods,’’ J. Ethnopharmacol., vol. 143, no. 2, pp. 463–468, [25] L. I. Feng-Gang, N. I. Zhi-Wei, and L. Gao, ‘‘TCM recipe automatic
2012. composing based on CBR and multi-strategies similarity retrieval,’’ Appl.
[3] P. Wang, H. Sun, P. Wang, Y. Han, and X. Wang, ‘‘Future perspectives of Res. Comput., vol. 27, no. 2, pp. 544–547, 2010.
personalized medicine in traditional Chinese medicine: A systems biology [26] J.-J. Xia et al., ‘‘The effect of TCM characteristic therapy on rehabilita-
approach,’’ Complementary Therapies Med., vol. 20, nos. 1–2, pp. 93–99, tion in patients with chronic kidney disease based on information shar-
2012. ing system,’’ in Proc. IEEE Int. Conf. Bioinform. Biomed., Nov. 2015,
[4] T. Ma, C. Tan, H. Zhang, M. Wang, W. Ding, and S. Li, ‘‘Bridging pp. 173–178.
the gap between traditional Chinese medicine and systems biology: The [27] P. Constantinou and A. P. Dawid, ‘‘Extended conditional independence
connection of Cold Syndrome and NEI network,’’ Mol. Biosyst., vol. 6, and applications in causal inference,’’ J. Artif. Intell. Res., vol. 5, no. 1,
no. 4, pp. 613–619, 2010. pp. 301–328, 2015.
[5] M. Jiang et al., ‘‘Syndrome differentiation in modern research of traditional [28] H. Wang et al., ‘‘Development of syndrome differentiation of pathogeny in
Chinese medicine,’’ J. Ethnopharmacol., vol. 140, no. 3, pp. 634–642, Chinese medicine syndrome differentiation system,’’ J. Traditional Chin.
2012. Med., vol. 58, no. 7, pp. 541–544, 2017.
[29] Y. U. Meili et al., ‘‘Historical evolution and modern development of
[6] M. F. Mei, ‘‘A systematic analysis of the theory and practice of syn-
Chinese medicine syndrome differentiation system,’’ J. Traditional Chin.
drome differentiation,’’ Chin. J. Integr. Med., vol. 17, no. 11, pp. 803–810,
Med., vol. 57, no. 12, pp. 991–995, 2016.
2011.
[30] X. U. Weiming et al., ‘‘Systematic retrospection and research prospects
[7] X.-H. Bai et al., ‘‘Development and application of microcosmic syndrome
of contemporary Chinese medicine syndrome differentiation methods,’’
differentiation in the modern TCM syndrome differentiation system,’’
J. Traditional Chin. Med., vol. 57, no. 18, pp. 1531–1539, 2016.
China J. Traditional Chin. Med. Pharmacy, vol. 30, no. 3, pp. 649–651, [31] Z. B. Zhang, Y. Y. Wang, A. P. Lv, L. Guo, and Y. Wang, ‘‘On combined
2015. syndrome differentiation of essential elements of syndrome and syndrome
[8] Y. U. Meili et al., ‘‘Historical evolution and modern development of target point corresponding syndromes,’’ J. Traditional Chin. Med., vol. 47,
Chinese medicine syndrome differentiation system,’’ J. Traditional Chin. no. 7, pp. 483–485, 2006.
Med., vol. 57, no. 12, pp. 991–995, 2016. [32] X. Sun et al., ‘‘Research on basic path and inherent law in the process of
[9] E. Yan, J. Song, C. Liu, and W. Hong, ‘‘A research on syndrome element Chinese medicine treatment based on syndrome differentiation,’’ J. Tradi-
differentiation based on phenomenology and mathematical method,’’ Chin. tional Chin. Med., vol. 57, no. 4, pp. 289–294, 2016.
Med., vol. 12, no. 1, p. 19, 2017. [33] X. U. Wen-Jie and Y. Q. Wang, ‘‘Application of modern methodology in the
[10] G. J. Yu et al., ‘‘Exploitation means and techniques of traditional Chinese research of TCM syndrome differentiation based on Bayesian network,’’
medicine expert system in pocket computer,’’ Discovery Nature, vol. 12, China J. Traditional Chin. Med. Pharmacy, vol. 27, no. 12, pp. 3168–3171,
no. 3, pp. 80–85, 1993. 2012.
[11] W. Chen and S. He, ‘‘Application of data mining technology in TCM [34] Q. Wang, ‘‘Classification and diagnosis basis of nine basic constitutions in
diagnosis and treatment,’’ in Proc. IEEE Int. Conf. Bioinformatics Biomed. Chinese medicine,’’ J. Beijing Univ. Traditional Chin. Med., vol. 28, no. 4,
(BIBM), Kansas City, MO, USA, Nov. 2017, pp. 1350–1353. pp. 1–8, 2005.
[12] X. U. Ming and L. U. Jia-Hui, ‘‘Discussion of TCM clinical data mining [35] J. Chen et al., ‘‘Development of clinical assistant diagnosis and treatment
based on the real world study,’’ China J. Traditional Chin. Med. Pharmacy, system based on traditional Chinese medicine syndrome differentiation and
vol. 32, no. 4, pp. 1640–1643, 2017. treatment,’’ Modernization Traditional Chin. Med. Materia Medica-World
[13] J. Y. Ren, X. U. Ming, and L. U. Jia-Hui, ‘‘Research progress on clinical Sci. Technol., vol. 17, no. 12, pp. 2436–2442, 2015.
medication regularity and syndrome type based on data mining,’’ China J. [36] B. Zhang and M. Wu, Internal Medicine of Traditional Chinese Medicine.
Traditional Chin. Med. Pharmacy, vol. 32, no. 10, pp. 4579–4582, 2017. Beijing, China: China Press of Traditional Chinese Medicine, 2016,
pp. 39–389.
[14] X.-Y. Xu, G.-Q. Han, and H.-Q. Min, ‘‘Construct concise and accurate
[37] Y. Tan, Gynecology of Traditional Chinese Medicine. Beijing, China:
classifier by atomic association rules,’’ in Proc. IEEE Int. Conf. Mach.
China Press of Traditional Chinese Medicine, 2016, pp. 56–282.
Learn., vol. 3, Aug. 2004, pp. 1604–1609.
[38] R. Ma, Pediatrics of Traditional Chinese Medicine. Beijing, China: China
[15] L. Y. Zhao et al., ‘‘Establishment of the associated model between turbid Press of Traditional Chinese Medicine, 2016, pp. 55–249.
phlegm syndrome and clinical indicators in the patients of diabetes type 2 [39] H. Chen, Surgery of Traditional Chinese Medicine. China Press of Tradi-
based on decision tree method,’’ World J. Integr. Traditional Western Med., tional Chinese Medicine, 2016, pp. 50–354.
vol. 9, no. 3, pp. 297–299, 2014. [40] Classification and Coding of Diseases in Traditional Chinese Medicine,
[16] S. Yu et al., ‘‘Research on dietetic therapy of TCM by computer- document GBT 15657-1995, 1995.
aided design,’’ in Proc. IEEE Int. Symp. IT Med. Educ., Aug. 2009, [41] Clinical Terminology of Traditional Chinese Medical Diagnosis and Treat-
pp. 1085–1090. ment, document GB/T 16751.1-1997, 1997.
JU CHEN received the M.E. degree from Sichuan CHUANBIAO WEN is currently pursuing the
Normal University in 2012. She is currently the master’s degree in software engineering from the
Executive Deputy Director of the Digital Med- University of Electronic Science and Technology
ical Research Institute, Chengdu University of of China. He is currently the Director of the Digital
Traditional Chinese Medicine (TCM), a Research Medical Research Institute, Chengdu University of
Assistant and a member of the Cloud Health Traditional Chinese Medicine (TCM). He is also
Branch, China Information Association for Tra- an Academic Leader of discipline focused on TCM
ditional Chinese Medicine and Pharmacy. She is informatics in the state administration of TCM and
principally engaged in the study of artificial intel- the President of the TCM and Pharmacy, Cloud
ligence of the TCM. She received the Third Prize Health Branch, China Information Association. He
in Sichuan Provincial Excellent Teaching Achievements in 2014, the First is principally engaged in the studies such as digital TCM science, big data
Prize of the Chengdu City Science and Technology Progress in 2015, the First analytics, cloud health platform, and artificial intelligence of the TCM. He
Prize of Science and Technology Progress from the China Association for received the First Prize of the effects of education reform regarding the
Acupuncture and Moxibustion in 2016, and the First Prize of the Science stimulation of practical training system by the thinking of the TCM in
and Technology Progress Award from the China Association for Research Sichuan, in 2011, the First and the Second Prize of the Chengdu City Science
and Advancement of Chinese Traditional Medicine in 2017. and Technology Progress Award in the platform based on the syndrome
differentiation and treatment in the TCM and the digitization system based on
the TCM from 2013 to 2015, the First Prize of the Sichuan Provincial Science
and Technology Progress Award in clinical decision supporting platform
on big data evidence-based acupuncture and moxibustion in 2016, and the
DIANXING YANG has cultivated over 20 master’s First Prize of the Science and Technology Progress Award from the China
and doctoral students. He is currently a Professor Association for Research and Advancement of the TCM in the big data
and the Ph.D. Supervisor of the Chengdu Uni- platform of syndrome differentiation and treatment in 2017.
versity of Traditional Chinese Medicine (TCM),
China, the Vice President of the China Asso-
ciation of Chinese Medicine, and the President
of the Sichuan Association of Chinese Medicine. XIWEI HUANG (M’15) received the B.Eng.
He has published more than 10 monographs as degree in information engineering from the
lead author, including Entering The New Era of Beijing Institute of Technology, China, in 2009,
the TCM-Rules of Syndrome Differentiation and and the Ph.D. degree in circuits and sys-
Treatment in the TCM. He has been engaged in teaching, scientific research, tems from Nanyang Technological University,
and the clinical practices of the TCM, oriented in clinical application, and Singapore, in 2015. From 2011 to 2014, he was
the rules of syndrome differentiation, treatment, and research on the TCM a joint Ph.D. Student with the Institute of Micro-
digitization. He received many prizes, such as the Second Prize from the electronics, Agency for Science, Technology and
National Excellent Teaching Achievements and the First and Second Prizes Research, Singapore. In 2015, he joined the School
in Sichuan Provincial Scientific and Technological Progress. of Electronics and Information, Hangzhou Dianzi
University, China, where he has been an Associate Professor since 2016. His
research interest includes CMOS multimodal sensors and systems design for
biomedical diagnostics.