IntroductionEffective management of clinical information, including patient personal identification data and clinical data, becomes a process which depends more and more on information technological (IT) support in healthcare.Aim and MethodsThe implementation of an ophthalmology decision-support system requires studying the inner processes as a target for modeling, and the ability to built “what is known” (knowledge), from separate observation data which is generated during all sorts of medical procedures. In the experiments we have made with the elicitation and representation of causal knowledge required for diagnosing glaucoma, this process has been structured in interviewing sessions.ResultsThe central result of our approach is the system knowledge base, structured in the form of a causal network, representing the factual knowledge in the system as a set of variable-value pairs, together with state and context variables, and their tabular representations of the causal relations attached to action nodes. In the case study illustrating the general approach, we presented a diagnostic problem solution as a pair consisting of a system model and a set of input data, making an important distinction between contextual input data, which drive and focus the diagnosis reasoning process, and observations, which is data that has to be explained. “Knowledge” data obtained as a result of the process is further incorporated into the system and used for decision support during the long-term monitoring of affection.DiscussionIn spite of multiple barriers that complicate collecting, recording or sharing of clinical information, such as privacy regulations, decentralization, the heterogeneous nature of medical systems, or deficient inter-specialties communication, knowledge-based representations may solve some of the problems caused by a vast amount and distributed nature of medical data.ConclusionKnowledge representation based on patient's clinical data is important for diseases such as glaucoma that require the establishment of logical relations during the long-term monitoring. Computer-assisted objective monitoring of glaucoma offers more useful information for taking correct decisions, such as the information on patient's medical history, previous laboratory test results or drug administration the implication and influence of risk factors, and their role related to affection compensation.