Editorial Evidence-Based Patient Classification for Traditional ...downloads.hindawi.com/journals/ecam/2015/168343.pdfEditorial Evidence-Based Patient Classification for Traditional

EditorialEvidence-Based Patient Classification forTraditional Chinese Medicine

Guo-Zheng Li,1 Ka-Fai Chung,2 and Josiah Poon3

1Data Center of Traditional Chinese Medicine, China Academy of Chinese Medical Science, Beijing 100700, China2Department of Psychiatry, The University of Hong Kong, Hong Kong3School of Information Technologies, The University of Sydney, Sydney, NSW 2006, Australia

Correspondence should be addressed to Ka-Fai Chung; [email protected]

Received 19 April 2015; Accepted 19 April 2015

Copyright © 2015 Guo-Zheng Li et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Some form of categorization or subdivision of disease statesis inevitable in all branches of medicine. According to thetraditional Chinese medicine (TCM) theory, the patterns ofbodily disharmony are described in terms of eight majorparameters: yin and yang, external and internal, hot andcold, and excess and deficiency. Additional systems such asqi, blood, and body-fluid differentiation and zang fu (organ)differentiation are also used. To enhance wider acceptance,these TCM categories have to have sufficient reliability andvalidity. There are interrater and test-retest reliability andfourmajor types of validity: concurrent, predictive, construct,and content. As the TCM theory originated from ancientmedical texts and the classification is based on symptomsand signs, the key shortcomings are the highly subjectivediagnostic process and the lack of scientific support for theTCM classification. In this special issue, the evidence base ofthe TCM classification was investigated.

In the paper “Advances in Patient Classification for Tradi-tional Chinese Medicine: A Machine Learning Perspective,”the authors introduced the machine learning algorithmsfor sign classification, syndrome differentiation, and diseaseclassification. Clinical features, as derived from inspection,auscultation, olfaction, and palpation, and patients’ datasetfrom medical records could be analyzed by k nearest neigh-bor, support vector machine, linear discriminant analysis,naive bayes, decision tree, artificial neural network, graphicalmodels,multilabel learning, deep learning, or clustering anal-ysis. The use of machine learning algorithms is a step towardenhancing data reliability, but the accuracy, as comparedto experienced TCM practitioners, is still suboptimal. We

believe that the machine learning algorithms are best used todeal with large and complex dataset, while the application insign classification merits further investigation.

With the help of computer technology, the paper “Signif-icant Geometry Features in Tongue Image Analysis” catego-rized five different tongue shapes: rectangular, acute triangle,obtuse triangle, circle, and square. The authors found thattongue shape can be used to distinguish healthy and diseasestates, with an average accuracy of 76%. Further testing of therole of tongue shape in TCM classification is needed.

Two papers introduced novel statistical approaches toassist TCM classification. In the paper “Mining Symptom-Herb Patterns from Patient Records Using Tripartite Graph,”the authors designed a data mining approach to examinethe relationship between symptom, syndrome, and herb.Thistripartite information network derives more accurate infor-mation than linking symptom and herb alone. In the paper “ANovel Classification Method for Syndrome Differentiation ofPatients with AIDS,” the authors found that a novel machinelearning algorithm, minimum reference set-based multipleinstance learning, was superior to other machine learningalgorithms for TCM classification.

A few papers examined the clinical application of TCMclassification. In the paper “Yang Deficiency Body Constitu-tion Acts as a Predictor of Diabetic Retinopathy in Patientswith Type 2 Diabetes: Taichung Diabetic Body ConstitutionStudy,” 673 patientswith diabeteswere examined using a bodyconstitution questionnaire. The authors showed that yangdeficiency was an independent predictor of a lower risk ofdiabetic retinopathy, suggesting that TCM classification may

Hindawi Publishing CorporationEvidence-Based Complementary and Alternative MedicineVolume 2015, Article ID 168343, 2 pageshttp://dx.doi.org/10.1155/2015/168343

2 Evidence-Based Complementary and Alternative Medicine

have predictive validity on disease complication. As the studydesign was cross-sectional, the causal relationship betweenTCM pattern and diabetic complication should be furtherexamined in longitudinal studies.

The paper “Cerebral Activity Changes in Different Tra-ditional Chinese Medicine Patterns of Psychogenic ErectileDysfunction Patients” showed that, compared to patientswith liver-qi stagnation and spleen deficiency, patients withkidney-yang deficiency showed an increased activity in bilat-eral brainstem, cerebellum, hippocampus, and the rightinsula, thalamus, andmiddle cingulate cortex and a decreasedactivity in bilateral putamen, medial frontal gyrus, temporalpole, and the right caudate nucleus, orbitofrontal cortex,anterior cingulate cortex, and posterior cingulate cortex. Asthemain difference in cerebral activity between the TCMpat-terns is in the brain regions that are responsible for emotionalmodulation, the authors believed that TCM classificationmight help disease subclassification, which could be used toprovide personalized treatment.

In the paper “Analysis and Recognition of TraditionalChinese Medicine Pulse Based on the Hilbert-Huang Trans-form and Random Forest in Patients with Coronary HeartDisease,” the authors introduced an objective pulse measure-ment, which could distinguish patients with coronary heartdisease and healthy controls. Further studies are needed totest the reliability and validity of the pulse characteristics.Another paper titled “Correlations between Phlegm Syn-drome of Chinese Medicine and Coronary Angiography: ASystematic Review and Meta-Analysis” showed that phlegmsyndrome can help to determine the severity of coronaryheart disease.

Lastly, in the paper “Prescription of Chinese HerbalMedicine in Pattern-Based Traditional Chinese MedicineTreatment for Depression: A Systematic Review,” the authorsfound that liver qi depression, liver depression and spleendeficiency, dual deficiency of the heart and spleen, and liverdepression and qi stagnation were the most common TCMpatterns in people with depression. The authors identifiedseveral pattern-based TCM treatments that could be furtherexamined. Xiaoyao decoction was the most frequently usedherbal formula for the treatment of liver qi depression andliver depression with spleen deficiency, while Chaihu Shugandecoction was often used for liver depression and qi stag-nation. The authors called for more high quality studiesincorporating TCM pattern differentiation and treatmentprinciple to examine the efficacy of TCM treatments and theadditional benefit of pattern differentiation. Future studiesincorporating expertise in various disciplines are needed tofurther examine the evidence base of TCM classification inpatient care.

Acknowledgment

We would like to express our thanks to all the contributorsand reviewers of this special issue.

Guo-Zheng LiKa-Fai Chung

Josiah Poon

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

Documents

Editorial Evidence-Based Patient Classification for Traditional ...downloads.hindawi.com/journals/ecam/2015/168343.pdfEditorial Evidence-Based Patient Classification for Traditional