Towards a General Type 2 Fuzzy Reasoning Approach to Modeling Modifiable Dementia Risk Factors

Dementia is the most critical neurodegenerative disease that gradually destroys memory and other cognitive functions. Hence early detection is utmost necessary and to build an effective detection model, it is required to understand its type, symptoms, stages and causes, as well as diagnosis methodologies. Many risk factors for dementia, identified from observational studies, are potentially modifiable; based on a large-scale review of observational studies, the Lancet Commission of Dementia Prevention, Intervention and Care estimated that approximately 40% of cases of clinically classified dementia could be attributed to twelve key modifiable dementia risk factors (MDRFs). Based on this evidence, the aim of this paper is to develop a fuzzy reasoning model that allows giving linguistic values to the data and information obtained from patients and giving an evaluation based on fuzzy rules; thus being able to interpret from these modifiable risk factors the percentage of risk that the patient has of increasing them according to their characteristics and lifestyle. The fuzzy reasoning is developed using general type-2 fuzzy systems to provide more robust interpretation to capture and manage the uncertainty presented in the data.