CDSS has been developed to assist physicians
diagnose difficult-to-diagnose cases where the
patient has been admitted to a tertiary care
hospital and the physician has not been able to
diagnose the case.
CDSS has at its core a database consisting of
disease-features, feature-disease and
feature-feature links. The features are quantified
using denotational semantics in the back-ground of a
disease. Using Artificial Intelligence tools like
logic formalisms, Bayesian probabilistic belief
network and others, CDSS calculates the
probabilities for a group of diseases (differential
diagnosis) for the given set of features. The system
has been designed to handle 50 features for a given
patient.
CDSS database consists of over 20,000
features and over 400,000 feature to
feature links. About 600 known diseases in internal
medicine have been covered. Besides being an
excellent decision supporting tool it can be used as
a medical reference tool for imparting knowledge.
The project is being supported by TIFAC, Department
of Science & Technology, Government of India to
commercialize it.
