iMedication – a Rule-based Expert Network for Intelligent Medication

Safe medication is a growing concern in the health sector. Doctors are no longer able to keep abreast of the steadily growing amount of drugs and their potential for unintended interactions and adverse drug events (ADE).

Starting with the prescription situation, iMedication will set up a comprehensive knowledge–based ADE reporting process which aims to

  1. identify as early as possible suspected adverse drug events (ADEs),
  2. support plausibility checks about an ADE suspicion and
  3. report verified ADEs in a sufficiently detailed and standardised way.

The  reported  ADEs  will  be  collected  in  a  knowledge  base  using  the  Arden  notation.  Arden rules will then specify drug risk patterns which can be used by doctors as additional checks for the drug prescription.  On   a   scientific-technological   level   iMedication   will   semantically   integrate   distributed knowledge about drugs and ADEs in order to support the prescription task. Expert rules will be used in order to support the decision making process for selecting the appropriate drug. A rule  engine  stores  the  medication  rules,  the  rules  for  the  drug  risk  patterns  and  possible recommendations.  iMedication  uses  Arden  syntax  for  specifying  the  rules,  which  is  an artificial-intelligence-based  formal  knowledge  representation  language  for  specifying  and processing  statements  about  medical  conditions  and  for  specifying  decision  rules,  e,g,  for recommendations.

iMedication  aims  at  increasing  patient  safety  through  the  linking  of  distributed  knowledge repositories,  through  supporting  the  early  identification  of  adverse  drug  events  and  through integrating the operative knowledge of local as well as remote, experts.