Our first meetup of our recently founded group
A medical profile (health record) is a container for health information that describes a medical case. It can be owned by a real patient or a virtual agent of the system. It can be used in search queries for similar cases or other relevant medical information. It can represent medical knowledge about a common case of a particular disease or an aggregation of real cases. There are different types of profiles used to represent different types of knowledge in the system. However, they all share a common information architecture to make them interoperable.
A medical condition can be characterized by the symptoms it causes and the way these symptoms develop over time (clinical representation & time course).
The purpose of archetype profiles is to capture common variations of those developments – a blue print, identifying common cases of a specific diseases. Archetype profiles can be matched with real patient profiles in order to find information about possible causes for symptoms (support for differential diagnosis) or standard methods of treatment.
Clustered profiles represent average cases of similar patients extracted by machine learning algorithms. They extend archetype profiles with more variations, eventually leading to the formation of more specific subclasses. It should be expected that some of the them will overlap significantly with existing archetype profiles which can be interpreted as a sort of verification of the archetype profiles correctness. Clustered profiles can also represent combinations of diseases that commonly appear together.
The major part of medical data will be real patient data – maintained by the users themselves as their own personal health record. Patient profiles pour in the experience and expertise of patients and make it available to the community.
They contribute to the richness and diversity of the platforms medical knowledge and extend the common medical facts with what might not have been verified by empirical research. Patient profiles bring in the creativity of people and allow the community tomeasure and verify what works and what doesn’t.
The automated formation of networks of similar profiles allows every user to participate in the knowledge exchange of groups of related patients: medical peer groups).