Expert System

Tender description

We are keen to explore the use of artificial intelligence to help us in targeting inspection resources. We want to develop tools that help inspectors come to decisions about whether a particular aspect of care is of sufficient concern to trigger an inspection. We are particularly interested in an expert system approach, which would help our inspectors make an assessment of the existing data, then obtain and add data in proportion to any lack of certainty about quality or changes in quality. We need to develop exemplar systems, which give us the ability to work through the health and social care sectors. We believe the most appropriate approach is to focus on specific aspects of the regulatory model to conduct a proof of concept and see if this is possible for CQC. This may be as narrow as a single prompt under a key question, or may be broader. The approach would involve interviewing our CQC experts (inspection managers, inspectors or analysts) and organizing the information into a collection of rules. The aim is that the expert system would draw deductions from the rules and may in turn query the user. An expert system could continue to 'learn' from the answers it is given.