RMHI logo
rmhiherbal.org
[RMHI Home] [Articles]
[Catalog] [Courses/Certification] [FAQ]
[HerbalThink-TCM software]
[Subscribe/Download]
[About us] [Contact]

— updated 2017-02-19

HerbalThink-TCM — Beginner Tutorial #6:
CaseQuery and AutoSage-TCM

by Roger W. Wicke, Ph.D.
This tutorial series will help you get started using the HerbalThink-TCM software, ensure it is installed correctly, and demonstrate how to use important features and modules.

Copyright ©2016-2017 by RMH-Publications Trust; all rights reserved. Published by the Rocky Mountain Herbal Institute; c/o PO Box 579; Hot Springs, Montana [59845] USA. Education and software for health professionals:   www.rmhiherbal.org


Jump to another tutorial in this sequence:
#1: Download #2: Setup and Documentation #3: Self-Study Reference #4: TCM Herbal Tutor #5: TCM Pulse Simulator #6: CaseQuery and AutoSage-TCM #7: Completing member/admission requirements

Subtopics on this page…

 

Tutorial #6:   CaseQuery and AutoSage-TCM

CaseQuery and AutoSage-TCM together comprise expert-system software for automated pattern recognition of the clinical syndromes constituting traditional Chinese medical (TCM) pathophysiology theory, including quantitative assessment of complex cases characterized by multiple simultaneous syndrome-patterns.

CaseQuery allows users to record detailed case histories of symptom, tongue, and pulse data. Case record files are created in a standardized format that can be read and understood by the AutoSage-TCM analysis software. The user then emails this data file to RMHI, where it is processed using the AutoSage-TCM program, and then a detailed, cross-referenced report is emailed back to the user.

Both CaseQuery and AutoSage-TCM are described in much greater detail in TCHS Vol.6: User's Guide to AutoSage-TCM. Chapter 7 of that text includes instructions for CaseQuery users and a series of screen shots. It may be helpful but not necessary to read any of that material before completing this tutorial.





 

Questionnaire #6

  • Copy and paste the following into a text document using any simple text editor (Mac OS X: TextEdit;   Windows: WordPad, Notepad). You will be filling in your answers (between the [ ] marks) during this tutorial and then emailing your completed questionnaire to RMHI.
  • Most questions will require simple, short answers. These questions are designed primarily to ensure that you know how to access the appropriate material within each module and to interpret it correctly.
  • Now you are ready to actually begin the tutorial. Follow the instructions in the remainder of this tutorial, and then fill in your answers to the questionnaire when asked to do so.
============ HERBALTHINK-TCM BEGINNER TUTORIAL #6 QUESTIONNAIRE ==========
(updated 2017-02-10)

--------------- AutoSage-TCM and CaseQuery:
B-00: What is the gender and age of the individual in case #C102b_20161123 ?
  [  ]

B-01: What is his body-mass index (BMI) value? (See the first encoded 
symptom.)
  [  ]

B-02: What does the 2nd encoded symptom say about the quality of fever? 
State your short answer in plain English.
  [  ]

B-03: What is his pulse strength at position L1? (Remember to interpret the 
'shorthand' notation correctly, i.e.: mild Moderate SEVERE EXTREME!!.)
  [  ]

B-04: What is unusual about the pulse at position L2? (Why are there two 
sets of data lines specified here instead of the more typical single line?)
  [  ]

B-05: One of the tongue tissue colors is specified to be 
'TONGaTISCOLOR_1_red_d1s'. Find the link to the corresponding color sample. 
In what way does this color differ from normal — hue, saturation, and/or 
lightness?
  [  ]

B-06: What portion of the tongue is specified to be the preceding color? 
(Which sectors/rows?)
  [  ]

B-10: Referring to the section "Syndrome-pattern analysis summaries: 
ByProductOf_MAGN_RANK_PROB__overThresholdOnly", list the top-scoring 
dominant (1st-tier) syndrome followed by its importance score, 'product 
of MAGN*RANK*PROB'.
  [  ]

B-11: Examine the reasoning behind the conclusion for the top-scoring 
dominant syndrome.
   How many of its active symptom variables have scores for RANK_raw 
≥ 1.0? (List them here by copying and pasting.)  [  ]
   What is this syndrome's probability score and why is it relatively 
high? (Hint: examine the competitor syndromes.)  [  ]

B-12: Likewise, list the only other dominant syndrome followed by its 
importance score, 'product of MAGN*RANK*PROB'.
  [  ]

B-13: Likewise, list the three 2nd-tier syndromes followed by their 
respective importance scores.
  [  ]

B-14: Referring to the section "Detailed analysis of all syndromes, 
by functional group", find the syndrome 'ORGU_5_DeficiencyOfHeartYin' 
and examine its detailed analysis.
   What is its importance score (MAGN*RANK*PROB)?  [  ]
   Why is its score so low in spite of a number of its symptom variables 
being significant? (Hint: examine the competitor syndromes and their 
effect on the probability, PROB.)  [  ]

A-0: Attach the CaseQuery output file (#A001) that you created in the 
first part of this tutorial and email it with this questionnaire.

=============================== END ======================================




 

CaseQuery

To run CaseQuery, double-click on one of the following shellscript files within your installed  herbalthinktcm  application folder, appropriate for your operating system (the instructions are otherwise identical to running HerbalThink-TCM, see Help/Documentation):

  • CASEQUERY-linux.sh
  • CASEQUERY-macosx-cmd.cmd
  • CASEQUERY-windows-bat.bat

Simply follow the step-by-step instructions, answer the case-history questionnaire as completely as possible, and then when you have finished, save the final results in a case-history data file, which you will send to RMHI as an email attachment.

For this tutorial, you will be entering in a very simple case study that illustrates the mechanics of how symptoms, symptom modifier terms, tongue, and pulse data are entered. You will be emailing the case file as an attachment to Questionnaire #6, and we will check that you have entered in the data correctly and that it accurately reflects the wording of the original case report. It is important, however, to understand that this case is far from complete and does not come close to the level of detail that you should strive for when you are entering data for a real person.

Following is the test case that you now should transcribe into a data file using the CaseQuery program. Just carefully follow the instructions that appear within CaseQuery on each tab, and complete all the tabs in their natural sequence.
This case includes a small set of symptoms, but these symptoms are examples that will require you to think carefully about how you will code them using the symptom modifier tab ('SSmod').

  • CaseQuery test case #A001. This case is very sketchy and only includes a small number of symptoms. It is for practice only and is assigned to test your ability to accurately transcribe this case record into the CaseQuery format. Typically, a real case report should be MUCH more thorough than this and will include numerous symptoms.
    Important:
        · In transcribing the symptoms, choose the questionnaire entries and modifiers that most closely represent the meaning of the items below.
        · Try using the 'Find' command to quickly find specific symptoms within the symptoms tabs.
        · You will need to learn how to use the symptom-modifiers tab ('12-SSmod') and the various 'Edit' menu options. Be sure to read 'Instructions for @MODIFIERS' within the '12-SSmod' section.
        · For symptoms #1 and #3 below, see especially: 'By-single-line' mode; 'Duplicate SS LineItem' command in the 'Edit' menu.
  • F, age 40
  • Height: 5'8"
  • Weight: 170 lbs.
  • Symptoms. Has been experiencing chronic:
    1. Tendency to easily feel too hot generally, but especially severe in late summer
    2. In late summer is easily dehydrated, with dry mouth and frequent thirst
    3. Aching pain in joints and muscles, especially of legs and hips, which is worse with pressure and better with exertion (exercise and movement)
    4. Poor appetite
    5. Poor memory
  • Tongue:
    • Tongue tissue is dark red (in color chart: red_d1t ) in all sectors except at the tip (row a), where it is slightly pale (red_p1t).
    • Tongue coating is normal-white all over, and slightly thicker at rows 'd' and 'e' at the root.
  • Pulses (be sure to choose the correct 'severity' value for each parameter):
    • 71 beats/min.
    • Rhythm: normal
    • Length/duration: normal
    • At L1: floating, slightly strong, and wide.
    • At L2 it is moderately sinking and changes quality abruptly between the mid and deep levels:
      • L2mid: normal strength, slightly taut, narrow, very distinct
      • L2deep: strong, very taut, wide, smooth time-profile
    • At L3: slightly weak, slightly narrow
    • At R1: slightly wide, fuzzy edged, very smooth time-profile
    • At R2: wide, fuzzy-edged, smooth time-profile
    • At R3: sinking, slightly weak, fuzzy-edged


After you are finished, be sure to save your file. You can always open it later and continue to edit, if you cannot complete it in one session.


Important note about the 'MISC' tab:  
The CaseQuery tab labelled '1-MISC' is where data like age, height, weight, and the following categories of miscellaneous health history information may be entered:

  1. Summary of chief complaints
  2. Medical diagnoses, if any provided
  3. History of major illnesses
  4. Misc: medications, environmental and occupational health exposure risks

The preceding 4 categories of miscellaneous information are not evaluated by AutoSage-TCM, but will be of major importance to any humans who will be reading and interpreting the report. This information may also be useful to the AutoSage-TCM developers and programmers whenever questions of interpretation arise. When you use CaseQuery, you should attempt to represent as much of this information as possible by checking the appropriate symptom items within the subsequent questionnaire tabs.





 

AutoSage-TCM

In this tutorial, you will be reading the AutoSage-TCM analysis report for case #C102b_20161123, which will appear in a new window. After reading the instructions ("How to use this report") contained within that report, answer the questions in the questionnaire below.
(This case is a fictional composite of a number of real cases.)

Refer to the report's table of contents to find specific sections. In the "Summary of original case record" section, you will find the complete set of encoded symptom data, listed in a standard order identical to its sequencing within the CaseQuery tabs. To start with, go to that section now, read "Encoding of symptom-sign data explained", and then begin answering the questions below. Remember to jump back to the table of contents when you need to find a different data section.


 

Questionnaire — AutoSage-TCM and CaseQuery

  • B-00: What is the gender and age of the individual in case #C102b_20161123?
  • B-01: What is his body-mass index (BMI) value? (See the first encoded symptom.)
  • B-02: What does the 2nd encoded symptom say about the quality of fever? State your short answer in plain English.
  • B-03: What is his pulse strength at position L1? (Remember to interpret the 'shorthand' notation correctly, i.e.: mild Moderate SEVERE EXTREME!!.)
  • B-04: What is unusual about the pulse at position L2? (Why are there two sets of data lines specified here instead of the more typical single line?)
  • B-05: One of the tongue tissue colors is specified to be 'TONGaTISCOLOR_1_red_d1s'. Find the link to the corresponding color sample. In what way does this color differ from normal — hue, saturation, and/or lightness?
  • B-06: What portion of the tongue is specified to be the preceding color? (Which sectors/rows?)
  • B-10: Referring to the section "Syndrome-pattern analysis summaries: ByProductOf_MAGN_RANK_PROB__overThresholdOnly", list the top-scoring dominant (1st-tier) syndrome followed by its importance score, 'product of MAGN*RANK*PROB'.
  • B-11: Examine the reasoning behind the conclusion for the top-scoring dominant syndrome.
        How many of its active symptom variables have scores for RANK_raw ≥ 1.0? (List them here by copying and pasting.)
        What is this syndrome's probability score and why is it relatively high? (Hint: examine the competitor syndromes.)
  • B-12: Likewise, list the only other dominant syndrome followed by its importance score, 'product of MAGN*RANK*PROB'.
  • B-13: Likewise, list the three 2nd-tier syndromes followed by their respective importance scores.
  • B-14: Referring to the section "Detailed analysis of all syndromes, by functional group", find the syndrome 'ORGU_5_DeficiencyOfHeartYin' and examine its detailed analysis.
        What is its importance score (MAGN*RANK*PROB)?
        Why is its score so low in spite of a number of its symptom variables being significant? (Hint: examine the competitor syndromes and their effect on the probability, PROB.)
  • A-0: Attach the CaseQuery output file (#A001) that you created in the first part of this tutorial and email it with this questionnaire.



Summary of what you learned in this tutorial

The CaseQuery software utility allows you to create a detailed report of an individual's symptoms and clinical signs, including descriptions of tongue and pulse that are so important in traditional Chinese medicine. Learn to use the '@SSmod' tab effectively — this tool will allow you to specify a symptom's severity, its time-pattern history (chronic/acute), applicable body locations, time of day/month/year, what aggravates/relieves, and other modifiers. By applying these modifiers correctly, you can usually come close to representing the intended meaning of the patient's description in his/her own words.

The AutoSage-TCM expert system can analyze a CaseQuery report file to produce a detailed analysis of the syndrome-patterns, allowing you to determine the dominant patterns and to examine the exact reasoning behind those conclusions.





Go to   Tutorial #7:   Completing member/admission requirements