COMP201 Java Programming

1
An Evidence-Based Approach to TCM Patient Class
Definition and Differentiation
Nevin L. Zhang The Hong Kong Univ. of Sci.
Tech. http//www.cse.ust.hk/lzhang

• Joint Work with
• HKUST Yuan Shihong, Chen
  Tao, Wang Yi, Liu Tengfei, Poon Kin Man, Liu Hua
• Beijing TCM U Wang Tianfang, Zhao Yan, Xu
  Wenjie, Wang Qingguo
• Shanghai TCM U Xu Zhaoxia, Wang Yiqing
• Academy of TCM Zhou Xuezhong, Zhang Runshun,
  Gong Yanbin, He Liyun, Wang Jie, Liu Baoyan
• Beijing Dongfang Hospital Zhang Yongling, Chen
  Boxing, Fu Chen

2
TCM is Worthy of Research

• Traditional Chinese Medicine (TCM) is important
  to the Chinese people.
• Culture tradition
• Health care
• It is used by many others. WHO report
• Annual global herbal medicine market US60
  billion
• Traditional medicine treatment at least once in
  life
• 90 of Canadian, 49 of French people,
• 48 of Australians, 42 of Americans.

3
Spectrum of TCM Research
A visit to TCM Doctor
Patient Information Collection Inspection (?)) Auscultation Olfaction (?)) Inquiry (?)) Palpation (?))
Patient Classification Syndrome differentiation (??) Determine pattern of disharmony
Treatment Herbal medicine Acupuncture Tui Na, Cupping, Qigong, .., etc
4
Spectrum of TCM Research
A visit to TCM Doctor Research
Patient Information Collection Inspection (?)) Auscultation Olfaction (?)) Inquiry (?)) Palpation (?)) Instruments .
Patient Classification Syndrome differentiation (??) Determine pattern of disharmony
Treatment Herbal medicine Acupuncture Tui Na, Cupping, Qigong, .., etc
5
Spectrum of TCM Research
A visit to TCM Doctor Research
Patient Information Collection Inspection (?)) Auscultation Olfaction (?)) Inquiry (?)) Palpation (?)) Instruments .
Patient Classification Syndrome differentiation (??) Determine pattern of disharmony
Treatment Herbal medicine Acupuncture Tui Na, Cupping, Qigong, .., etc Efficacy Effective component of herbs Action mechanism of herbs Safety issue .
6
Spectrum of TCM Research
A visit to TCM Doctor Research
Patient Information Collection Inspection (?)) Auscultation Olfaction (?)) Inquiry (?)) Palpation (?)) Instruments .
Patient Classification Syndrome differentiation (??) Determine pattern of disharmony Supervised learning Labeled Data Symptoms signs, class labels assigned by expert
Treatment Herbal medicine Acupuncture Tui Na, Cupping, Qigong, .., etc Efficacy Effective component of herbs Action mechanism Safety .
7
Spectrum of TCM Research
A visit to TCM Doctor Research
Patient Information Collection Inspection (?)) Auscultation Olfaction (?)) Inquiry (?)) Palpation (?)) Instruments .
Patient Classification Syndrome differentiation (??) Determine pattern of disharmony Supervised learning Labeled Data Symptoms signs, class labels assigned by expert Our work cluster analysis Unlabeled Data symptoms signs
Treatment Herbal medicine Acupuncture Tui Na, Cupping, Qigong, .., etc Efficacy Effective component of herbs Action mechanism Safety .
8
Use of TCM to Treat Western Medicine Diseases

• Common practice in China
• Patients of a WM disease subdivided into several
  TCM classes
• Different classes are treated differently.
• Example
• WM disease Depression
• TCM Classes
• Liver-Qi Stagnation (????). Treatment principle
  ????, Prescription ?????
• Deficiency of Liver Yin and Kidney Yin
  (?????)Treatment principle ????, Prescription
  ?????????
• Vacuity of both heart and spleen (????).
  Treatment principle ????, Prescription ???
• .

9
Key Question

• How should patients of a WM disease be divided
  into subclasses from the TCM perspective?
• What TCM classes are there among patients of the
  WM disease?
• What are the characteristics of each TCM class?
• In practice, no consensus. Different researchers
  use different schemes
• Gao and Fang STAGNATION OF LIVER QI, SPIRIT
  INJURED BY WORRY, and HEART-SPLEEN DUAL VACUITY.
• You et al. LIVER DEPRESSION AND SPLEEN VACUITY,
  HEART-SPLEEN DUAL VACUITY, and DEFICIENCY OF
  LIVER-YIN AND KIDNEY-YIN.
• Guo et al. LIVER DEPRESSION AND SPLEEN VACUITY,
  LIVER BLOOD STASIS AND STAGNATION, HEART-SPLEEN
  DUAL VACUITY, and SPLEEN AND KIDNEY DUAL VACUITY.
  
• Definition of the classes also vague
• Our objective Provide evidence for the TCM
  sub-classing task through analysis clinic symptom
  data so that some standard can be established.

10
The Key Idea
Page 10

• Imagine sub-classing patients of a WM disease D
  from TCM perspective

• Also providing a basis for defining the TCM class
  Z and for differentiating class Z patients from
  other D patients

11
Outline

• Introduction
• Data Analysis Tool
• Case Study
• Another Perspective on the Results
• Conclusions

12
Cluster Analysis

• Grouping of objects into clusters so that objects
  in the same cluster are similar while objects
  from different clusters are dissimilar.

• Result of clustering is often a partition of all
  the objects.

13
How to Cluster Those?
Page 13
14
How to Cluster Those?
Page 14
Style of picture
15
How to Cluster Those?
Page 15
Type of object in picture
16
How to Cluster Those?
Page 16

• Complex data usually have multiple facets and be
  meaningfully partitioned in multiple ways.
  Multidimensional clustering / Multi-Clustering
• TCM symptom data are complex and need
  multidimensional clustering.
• No previous methods can perform multidimensional
  clustering.
• So, we developed our own method.
• It is a model-based method
• Latent tree models

17
Latent Tree Models

• Tree-structured probabilistic graphical model
• Leaf nodes represents observed variables
• Internal nodes represent latent variables
• Links represents dependence and quantified by
  probability distributions
• Generalization of latent class models

18
Latent Tree Analysis
From data on observed variables, obtain latent
tree model

• Learning latent tree models Determine
• Number of latent variables
• Number of possible states for each latent
  variable
• Model Structure
• Conditional probability distributions
• Algorithmic work http//www.cse.ust.hk/lzhang/lt
  m/index.htm

19
Latent Tree Analysis Multidimensional Clustering

• Each latent variable gives a partition
• Y1(Analytic Skill) cluster 1 (low), cluster 2
  (high)
• Y2 (Literal skill) cluster 1 (low), cluster
  2 (high)

20
Outline

• Introduction
• Data Analysis Tool
• Case Study
• Another Perspective on the Results
• Conclusions

21
Case Study Depression

• Subjects
• 604 depressive patients aged between 19 and 69
  from 9 hospitals
• Selected using the Chinese classification of
  mental disorder clinic guideline CCMD-3
• Exclusion
• Subjects we took anti-depression drugs within two
  weeks prior to the survey women in the
  gestational and suckling periods, .. etc
• Symptom variables
• From the TCM literature on depression between
  1994 and 2004.
• Searched with the phrase ?? and ? on the
  CNKI (China National Knowledge Infrastructure)
  data
• Kept only those on studies where patients were
  selected using the ICD-9, ICD-10, CCMD-2, or
  CCMD-3 guidelines.
• 143 symptoms reported in those studies altogether.

22
The Depression Data

• Data as a table
• 604 rows, each for a patient
• 143 columns, each for a symptom
• Table cells 0 symptom not present, 1 symptom
  present
• Removed Symptoms occurring lt10 times
• 86 symptoms variables entered latent tree
  analysis.
• Structure of the latent tree model obtained on
  the next two slides.

23
Model Obtained for a Depression Data (Top)
24
Model obtained for a Depression Data (Bottom)
25
Question

• Each latent variable gives a partition of the
  patients.
• Do the partitions provide evidence for the
  following questions
• What TCM classes are there among depressive
  patients?
• What are the characteristics of each of the
  classes?

26
The Empirical Partitions

• The first cluster (Y29 s0) consists of 54 of
  the patients and while the cluster (Y29 s1)
  consists of 46 of the patients.
• The two symptoms fear of cold and cold limbs
  do not occur often in the first cluster
• While they both tend to occur with high
  probabilities (0.8 and 0.85) in the second
  cluster.

27
Probabilistic Symptom co-occurrence pattern

• Probabilistic symptom co-occurrence pattern
• The table indicates that the two symptoms fear
  of cold and cold limbs tend to co-occur in the
  cluster Y29 s1
• Pattern meaningful from the TCM perspective.
• TCM asserts that YANG DEFICIENCY (??) can lead
  to, among other symptoms, fear of cold and
  cold limbs
• So, the co-occurrence pattern suggests the TCM
  symdrome type (??) YANG DEFICIENCY (??).

• The partition Y29 suggests that
• Among depressive patients, there is a subclass of
  patient with YANG DEFICIENCY.
• In this subclass, fear of cold and cold
  limbs
• co-occur with high probabilities (0.8 and
  0.85)

28
Probabilistic Symptom co-occurrence pattern

• Y28 s1 captures the probabilistic co-occurrence
  of aching lumbus, lumbar pain like pressure
  and lumbar pain like warmth.
• This pattern is present in 27 of the patients.
• It suggests that
• Among depressive patients, there is a subclass
  that correspond to the TCM concept of KINDNEY
  DEPRIVED OF NOURISHMENT (????)
• Characteristics of the subclass given by
  distributions for Y28 s1

29
Probabilistic Symptom co-occurrence pattern

• Y27 s1 captures the probabilistic co-occurrence
  of weak lumbus and knees and cumbersome
  limbs.
• This pattern is present in 44 of the patients
• It suggests that,
• Among depressive patients, there is a subclass
  that correspond to the TCM concept of KIDNEY
  DEFICIENCY (??)
• Characteristics of the subclass given by
  distributions for Y27 s1
• Y27, Y28, Y29 together provide evidence for
  defining KIDNEY YANG DEFICIENCY

30
Probabilistic Symptom co-occurrence pattern

• Pattern Y23 s1 provides evidence for defining
  LIVER QI STAGNATION ( ????)
• Pattern Y22 s1 provides evidence for defining
  LIVER QI STAGNATION

31
Probabilistic Symptom co-occurrence pattern

• Pattern Y21 s1 evidence for defining STAGNANT
  QI TURNING INTO FIRE (????)
• Y19 s1 evidence for defining QI STAGNATION IN
  HEAD
• Y17 s1 evidence for defining HEART QI
  DEFICIENCY
• Y16 s1 evidence for defining QI STAGNATION
• Y15 s1 evidence for defining QI DEFICIENCY

32
Probabilistic Symptom co-occurrence pattern

• Y11 s1 evidence for defining DEFICIENCY OF
  STOMACH/SPLEEN YIN (????)
• Y10 s1 evidence for definingYIN DEFICIENCY (??)
• Y9 s1 evidence for defining DEFICIENCY OF BOTH
  QI AND YIN (????)

33
Symptom Mutual-Exclusion Patterns

• Some empirical partitions reveal symptom
  exclusion patterns
• Y1 reveals the mutual exclusion of white
  tongue coating, yellow tongue coating and
  yellow-white tongue coating
• Y2 reveals the mutual exclusion of thin tongue
  coating, thick tongue coating and little
  tongue coating.

34
Summary

• By analyzing 604 cases of depressive patient data
  using latent tree models we have discovered a
  host of probabilistic symptom co-occurrence
  patterns and symptom mutual-exclusion patterns.
• Most of the co-occurrence patterns have clear TCM
  syndrome connotations, while the mutual-exclusion
  patterns are also reasonable and meaningful.
• The patterns can be used as evidence for the task
  of defining TCM classes in the context of
  depressive patients and for differentiating
  between those classes.

35
Outline

• Introduction
• Data Analysis Tool
• Case Study
• Another Perspective on the Results
• Conclusions

36
Statistical Validation of TCM Postulates
37
Value of Work in View of Others

• D. Haughton and J. Haughton. Living Standards
  Analytics Development through the Lens of
  Household Survey Data. Springer. 2012
• Zhang et al. provide a very interesting
  application of latent class models to diagnoses
  in traditional Chinese medicine (TCM).
• The results tend to confirm known theories in
  Chinese traditional medicine.
• This is a significant advance, since the
  scientific bases for these theories are not
  known.
• The model proposed by the authors provides at
  least a statistical justification for them.

38
Concluding Remarks

• Latent tree analysis is tool for
• Systematically identifying co-occurrence
  patterns of symptoms
• Introduce latent structure to explain the
  patterns
• Provide evidence in support of TCM postulates
  about symptom occurrence
• Tool for multidimensional clustering
• Each latent variable represents a partition of
  data
• Provide evidence for TCM patient class definition
  and differentiation

39

• Thank You!