Clinical decision support

It is an important goal, that the practical usage of the examinations provided by Abiomics and of their results be supported, i.e. that they could be use in clinical decision support.

In this area, the following main tasks can be supported.

  • Diagnosis based on current knowledge.
    Domain models developed during data analysis allow for the examination of concrete cases, and hence to infer about the other parts of the models from the results of examinations, i.e. the disease itself or its causes.
  • “Personalized” medicine (choice of treatment, medicals).
    If the relations between the underlying mechanisms of a disease and its genetics are known, then knowing the genetics of the patient can predict which possible treatment can provide the most effective aid in the given case.
  • Planning of further examinations based on cost and utility functions.
    Domain models can provide help in planning further examinations based on already-known results in a concrete case, since it can be determined which examinations will provide the most expected further information, which can help increase the efficiency of the medical protocol.
  • Development of decision models and protocols.
    Through the exploitation of the above results, general-purpose methodologies and protocols can be provided for given diagnoses. Such a protocol consists of the following steps.
    • Organizing electronic data collecting.
      The collection of clinical (and/or genetic) data which serves as the basis of analyses is a time-consuming and costly task, the success (the confidence and cleanness of data) of which is crucial regarding the following phases. Hence a methodology supported by efficient IT solutions can increase both efficiency and confidence.
    • Labs.
    • Biobanking.
      Physical storage of measurement samples is fundamental regarding the documentation of the entire process, and can be important for a posterior analysis or control.
    • Connecting to central data repositories.
    • Statistical data analysis.
      The main goal of statistical data analysis –beyond basic checks (e.g. the representativeness of data, the examination of marginal distributions)– is to discover the basic relations within the domain. The whole scale of applicable methods is very wide, ranging from the widely known standard classical statistical tests to the newly developed Bayesian statistical methods.
    • Validation.
    • Deployment.
      That the protocols developed through the above steps are carried out precisely, can be supported particularly by a suitable software tool. The deployment of it in a given institute is an inevitable part of the practical utilization of the results.