ES, version 2022-01-09 /
- Medical Decisions Support:
- Make knowledge better available for decisions.
Strategies for differential diagnosis and evaluation of
- AI as "Assisted Intelligence":
- Helping but not replacing healthcare professionals.
- A task oriented approach, identifying problems and seeking
- Make healthcare AI more friendly and more transparent,
with explanations of the decision process.
- Motivations in the scope of the evolution of
- Multiple factors:
- Decisions taking account of many factors presented as
graphs, with the relative weights of relationships between
concepts. This require the management of complex medical
information in a N-dimentional space, by means of graph
technologies. Therefore both medical knowledge and patient
records need to be converted in graphs.
- Human Interface:
- Development of easy and interactive interfaces between the
graphs in human minds, i.e. neurons and synapses, and the
graphs in computers.
- Better access to already existing medical knowledge. A
large amount of biomedical knowledge is in principle already
available but the question is now how to use this knowledge
in a more efficient way.
- Maintenance of a synthesis of agreed current medical
knowledge. Conversion of the current state of the arts into
- A multidisciplinary community including healthcare
professionals, data scientists, software developers and
- Scientific: dealing with methodology and knowledge base,
supported by universities and public grants.
- Operators: dealing with installations and
sustainability, supported by the users.
- Care provider: in contact with the patient and taking
the final decision.
- International Open Source and Open Data collaborations as
a not-for-profit initiative, With support from scientific
communities, universities and seeking grants of common
interest. An exiting challenge.
- Intended to support operators of services to users in
practice, but in a non exclusive way. Business activities of
operation of services to end users, as installation,
training and maintenance.
- Transparency of the sources of information by accredited
- Research in order to improve the graph knowledge base,
particularly about the quantification of the relationships
- Learning from the follow-up of the recommendations.
- Road map:
- Incremental development: start with exercises on simple
decision schema and make improvements step by step.
Prototypes as proof of concept focusing on methodology and
interdependence of any medical specialities.
- Intended users:
- Telemedicine in situations where there is limited local
qualifications and/or not enough time per patient.
- Many specialized health organizations are working on
knowledge bases in traditional and incompatible ways.
The challenges are integration and accessibility in a
format suited for decision support.
- Quality checks which may provide warnings.
- Contribution to medical education and training, in front
of interactive decision strategies.