A ‘Mini Linguistic State Examination’ to Classify Primary Progressive Aphasia

Background: There are few available methods for qualitatively evaluating patients with primary progressive aphasia (PPA). Commonly adopted approaches time-consuming, of limited accuracy, or designed to assess different patient populations. This paper introduces a new clinical test - the Mini Linguistic State Examination (MLSE) which was uniquely enable clinician and subclassify both classical mixed presentations PPA. The adoption novel assessment method (error classification) greatly amplifies information that can be derived from set standard linguistic tasks allows five-dimensional profile defined. Methods: Fifty-four 30 matched controls were recruited. Five domains language competence (motor speech, phonology, semantics, syntax, working memory) assessed using sequence 11 distinct assays. A random forest classification used diagnostic accuracy predicting PPA subtypes create decision tree as guide classification. Findings: prediction model 96% accurate overall (92% logopenic variant, 93% semantic 98% non-fluent variant PPA). produced correct 91% participants whose data not included in training set.      Interpretation: MLSE is cognitive incorporating powerful, yet straightforward, approach scoring. Rigorous its confirmed excellent matching syndromes gold diagnoses.  Adoption by clinicians will have decisive impact on consistency uniformity described clinically. It also facilitate screening cohort-based research, including future therapeutic trials, suitable describing, quantifying monitoring deficits other brain disorders. Funding Information: research funded Medical Research Council Grant award (Ref MR/N025881/1) PG, MLR JR, KP SC. Additional support provided through grants MRC (UAG051 G101400), Wellcome Trust (103838) ERC (GAP: 670428), funding awarded National Institute Health Cambridge Biomedical Centre CBU (MC_UU_00005/18). Declaration Interests: JBR reports consultancy unrelated work Biogen, UCB, Asceneuron Althira; receipt grants, this work, Janssen, AZ-Medimmune, Lilly. authors declare no conflicts interest. Ethics Approval Statement: Written informed consent all participants. study protocol reviewed approved London (Chelsea) Ethics Committee [Ref. 16/LO/1735].