1. PSYCHO-INFORMATICS
APPROACH IN PSYCHIATRIC
CLASSIFICATIONS
Debdulal Dutta Roy, Ph.D.
Psychology Research Unit
INDIAN STATISTICAL INSTITUTE, KOLKATA
National Workshop on “ Psycho-informatics : Model
for Measuring Randomized Psychological and
Educational Data”
11-12th March, 2013
th
Web: www.isical.ac.in/~ddroy
2. China, Japan and Latin American countries have
own Psychiatric diagnostic classification systems.
Does India need own ?
Do the Indian Psychiatrists find difficulty to make
distinct classifications ?
What complaints are used by Indian Psychiatrists
for classifications of psychiatric disorders ?
BACKGROUND
3. Previous psychiatric case history
records of the department of
psychiatry of one post graduate
hospital. Here data are random,
non-hypothesized and least
structured.
DATA WAREHOUSE
4. Individual case wise all
complaints and diagnostic
classifications are retrieved.
There were 20 complaints and
18 classifications
DATA RETRIEVING
5. All the complaints were coded
with 1,0 and Burt table was
constructed.
Burt table data were used in
Correspondence analysis.
DATA MINING
6. Correspondence analysis provides
correspondence map
Correspondence map reveals
extent of similarities and
dissimilarities among the complaints,
diagnostic classifications and
between the two.
PATTERN RECOGNITION :
7. C o r r e s p o n d e n c e m a p o f p s y c h i a t r i c c la s s i f i c a t i o n a n d c o m p la i n t s
In p u t T a b le ( R o w s x C o lu m n s ) : 5 4 x 5 4 ( B u r t T a b le )
Dimension 2; Eigenvalue: .11454 (11.45% of Inertia)
1 .5
P S Y D E P C L :1
E M W D L C M P :1 B L U A F C M EP U: 1D E P C L : 1
N
1 .0
D E P M D C G UP I:L1 T C M P : 1
M
A N X C TM E PN : S C M P : 0
0
0 .5 C O N D O C M P :1
S CH HA I Z C UL S S:M1U PS : P I C M P : 1
LL C 1
U N T H C C EM XPC : I1T C M P : 1 AOMN A DC AC SMS:I 0N 0 D0 L : P
X L
C SOP CDNT DEO C L LPS: S 0 T:
S OHPMYC TANMRGVCBCRLCNACLSHS:CLSOC:0:0SC0SM C 0SM : O0P M: 0 A T C O M : 1
H O H C L :0 :
H O S T C M P :1 P E S MY W CED OPL N D :OP C:E0MXS UPCN:AT0TLI HMSC PUC SSM :P0P I :C00 M P : 0
D C L
CM 0 C I H C Z C LC : M P :
H LU 0
0 .0 S O M A T C O M :0
G U IL T C M P : 0
N E PU DD DE CP L CS L: :1 0 S O M A T C L S :1
D E P M D C B LP U : 0 F C M P : 0
M A
- 0 .5 G R A N D C M P :1 A N XC M P :1
M A N IC C L S : 1 S T R E C L S :1 T E N S C M H PY : 1 H O C L S : 1
C
C O N V C L S :1
- 1 .0 M A N C M P :1
P H A O N BX CC LL SS :S: 11: 1
O C DCL
- 1 .5
- 2 .0
-2 .0 - 1 .5 -1 .0 -0 .5 0 .0 0 .5 1 .0 1 .5
D i m e n s i o n 1 ; E i g e n v a lu e : . 2 2 7 7 2 ( 2 2 . 7 7 % o f In e r t i a )
CORRESPONDENCE MAP
8. Psychiatrists find difficulty to
classify some symptoms distinctly.
This suggests need for creating
Indian system of psychiatric
classifications.
Psychiatric classifications should
cover symptom dynamics and the
etiology.
DISCOVERY OF KNOWLEDGE
9. Dr. Uday Sankar Mandal of IPGMR. I was
co-guide of his MD dissertation titled
“Mapping of Psychiatric Symptoms
Across Different Demographic
Groups: An Epidemiological problem”
in 2008-2011.
Acknowledgment