|
Inference Engine
Last edited |
10/27/2008
Written by Maurice M. Ohayon, MD, DSc, PhD
Sleep-EVAL
is a non-monotonic, level-2 expert system endowed with the ability to make
logical connections based on patient information (causal reasoning mode).
The knowledge processor, or inference engine, is the part of the expert system
that finds solutions to problems.
The inference engine uses its knowledge base to pose questions, to infer
hypotheses and to deduce diagnostic conclusions.
the Neural Networks
This is a computational structure composed of three
types of layers (subgroups of processing elements): input layers, output layers
and hidden layers.
More simply put, a neural network is useful for
solving pattern-matching problems.
In Sleep-EVAL, two neural networks are used by the
inference engine to manage any uncertainty in the subject’s answers as well as
in criteria and diagnoses:
-
The first neural network is a fixed one whose
function is to manage fuzzy sets of answers.
-
The second network is unfixed.
Its function is to calculate relative weights at the level of a criterion and
also on a series of criteria.
The cumulative weights are used to determine the
presence or absence of a criterion or a diagnosis.
In the end, each explored
object (including diagnoses) will have a degree of certainty (or weight) ranging
from 0.4 (completely present) to -0.4 (completely absent).
the Mathematical Preprocessor
It performs a number of mathematical operations,
such as converting age into months or weeks and hours into minutes or seconds,
comparing duration of symptoms or discrepancies between hours, and setting the
range of responses to be entered by numerical keypad.
How does it work?
-
First, the inference engine starts with a
general questionnaire to be asked to all participants.
-
Second, using the answers provided to this
questionnaire, the inference engine establishes the most adequate decision
trees in order to get the most probable diagnoses.
-
Once a potential diagnosis is found, the inference
engine explores all the other avenues to have another diagnosis and may ask
additional questions if it lacks the
necessary elements to confirm or deny the
diagnosis (differential diagnosis process).
References
Ohayon MM.
Sleep-EVAL, Knowledge Based
System for the Diagnosis of Sleep and Mental Disorders. Registration #437699,
Copyright Office, Canadian Intellectual Property Office. Ottawa: Industry
Canada, 1994. (English, French, German, Italian, Portuguese, Spanish, Finnish,
Swedish, Korean, Chinese versions).
Ohayon M. Knowledge Based System
Sleep-EVAL: Decisional Trees and Questionnaires. Quebec National Library, ISBN 2-921483-06-8, 1995.
Ohayon MM.
Improving decisionmaking processes
with the fuzzy logic approach in the epidemiology of sleep disorders.
J Psychosom Res 1999
Oct;47(4):297-311
|
|
Sleep-EVAL
Research
