COVID-19 Knowledge reference model

Instructions for running the demonstration

The Pathfinder 1 demonstrator has two main components:

  • A “data model”:  a structured and standardised model of the key personal, clinical and other data that may be important in decision-making across the patient journey
  • A “knowledge model”: a collection of executable logic models of the medical expertise that underpins best practice in clinical tasks like situation assessment, decision-making, treatment planning and workflow management in patient care.

Pathfinder 1 is a demonstration of the technical feasibility of having a standard data model for COVID-19 services that AI can help with based on such a model. It is some way from being a clinically practical tool because the medical knowledge base is incomplete and more work needs to be done to make it appropriate for clinical use. We hope you will give it a try though and let us have your comments and criticisms so that we can improve it.

Click the link “Play demo”. Pathfinder will open in a new window so you can return here at any time.

Underneath the OpenClinical banner on the home page you will see a set of tabs.

The top row allows you to review the current data model and knowledge model. The first tab shows the 42 data sources included in this first version, mostly dealing with clinical data. The next tab on the top row accesses a set of “CONTEXTS” in which subsets of data are needed (such as primary care or hospital work-up). Here these are a set of forms which appear as a third and fourth row of tabs.

Skip the third tab (ALERTS) for now and have a look at the DECISIONS tab, which allows you to review the main decisions that may need to be taken during a patient’s care (e.g. triage, diagnosis, prognosis, tests and investigations, prediction of complications) via the tabs in the bottom row of tabs.

Decisions are presented as simple reports for each type of decision; don’t spend too long on reviewing them here because it gets more interesting when we add some data about a particular patient.

Colleagues at BMJ Best Practice have kindly allowed us to include test cases which in the demonstrator. Go to the “example cases” tab (second row, second tab) and select a case. A summary of the case is displayed. Click on the “commit” button at the bottom and this will update the record for this imaginary patient.

The user interface is a bit unpredictable about what it does next but please ignore that – just click on the DECISIONS tab that you reviewed earlier; you can now go through the decision reports which were previously empty but have now been updated with pros and cons for each of the decision options.

Now click on the Care plans and pathways tab on the second row. This will display links to a set of procedures that use the Pathfinder components (data, forms, decisions etc) to enact a basic care pathway for a clinical setting, such as the contexts you saw earlier. You can add data to the forms and this will then be incorporated into the patient record and all the decision reports will be appropriately updated.

When you add data to a form remember to click on  the “submit” button at the bottom of the form.

If you want to run another example case first press the “restart” button at the top of the OC page.

Instructions for running the demonstration

  Publet information
Guideline objectives

Coronavirus disease 2019 (COVID-19) is a potentially severe acute respiratory infection caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).[1] The virus was identified as the cause of an outbreak of pneumonia of unknown cause in Wuhan City, Hubei Province, China, in December 2019.[2] The clinical presentation is that of a respiratory infection with a symptom severity ranging from a mild common cold-like illness, to a severe viral pneumonia leading to acute respiratory distress syndrome that is potentially fatal.

This publet is a demonstration of feasibility of developing a set of services for supporting decision making and workflow management in the care of COVID-19 patients from home to hospital to home. It is undergoing constinuous development and testing and is not provided or usable for clinical purposes.

Target setting Multiple settings across patient journey
Target users Principally clinicians across disciplines
  • Author: J Fox, OpenClinical CIC
  • Release date:

Last reviewed: March 2020

Last updated: April 16th 2020