Enhanced interpretation of the Mini-Mental State Examination

摘要: The goal of the research reported in this thesis is to contribute early and accurate detection dementia. Early dementia essential maximising effectiveness treatment against memory loss. This pursued by interpreting Mini-Mental State Examination (MMSE) novel ways. MMSE most widely used screening tool for dementia, it a questionnaire 30 items. objectives are as follows: • reduce dimensions relevant ones order inform predictive model using computational methods on data set results, construct predicting diagnosis informed features extracted from previous step applying, comparing combining traditional modelling methods, propose semantic analysis sentence writing question utilise information recorded MMS examinations which has not been considered previously. Traditional inadequate such due assumptions normally distributed data. Alternative discrete investigated method computing theoretic measures proposed. demonstrate that an automated improves accuracy differentiating between types Finally, models proposed integrate annotations with derive rules difficult distinguish