Turkish Speech Analysis Development Study

Overview

The aim of this observational study is to define speech alterations in the diagnosis of AD-related minor neurocognitive disorder (m-NBB) and major neurocognitive disorder (M-NBB). It is the development of a speech analysis method that can be used in future studies or routine clinical evaluations in these patients by using artificial intelligence and speech recognition methods that are planned to be developed based on the findings obtained. Purposes of the research 1. To define language disorders in the diagnosis of AD-related minor neurocognitive disorder (m-NBB) and major neurocognitive disorder (M-NBB). 2. Developing a speech analysis method that can be used in the early diagnosis of neurocognitive disorder patients by using artificial intelligence and speech recognition methods, which are planned to be developed based on the findings. 3. Evaluation of whether speech analysis distinguishes patients with m-NBB and M-NBB due to AD 4. Evaluation of whether speech analysis distinguishes minor NBD patients from healthy elderly patients. Standardized Mini-Mental Test, Clock Drawing Test, Montreal Cognitive Assessment Scale, Rey Auditory Verbal Learning Test, Verbal Fluency Test, Enhanced Cued Recall Test, Functional Activities Questionnaire, Trail Making Test, and Hamilton Depression Scale to evaluate the severity of depressive symptoms will be administered to the participants to assess cognitive functions. . In order to evaluate speech the Cookie Theft Picture Description Test, which is a part of the Boston Aphasia Test, will be administered to the participants.

The diagnosis of Alzheimer's Disease (AD) is based on clinical criteria using neuropsychological tests and neuroimaging, and is currently a diagnosis of exclusion and neuropathological examination/autopsy is required for a definitive diagnosis. Biomarkers, which have been tried to be developed recently, have not gained enough place in clinical use because they are considered to be high cost and/or risky.In recent years, the use of features of speech signals and spontaneous speech in the detection of AD has yielded successful results. It is thought that advances in classification, sound processing and speech recognition will facilitate disease diagnosis. It is thought that new researches should be made in Turkish based on literature data, since the Turkish language structure is agglutinative, some words such as the verb 'to be' do not have Turkish equivalents, the sentence order is in the form of subject/object/predicate, and personal pronouns are used less frequently than in English. In this respect, it is thought that the planned study will make an important contribution to the field in terms of being the first study to make a comprehensive assessment of language problems in patients with AD and Mild Cognitive Impairment and to use artificial intelligence methods. To the best of our knowledge, as there is currently no speech-based corpus for the assessment of progression from healthy aging to AD, one of the important contributions of the data to be obtained from this study is that produced by healthy elderly, amnestic type minor neurocognitive disorder and patients with major neurocognitive disorder due to AD. It may be the basis for the creation of a corpus of standard speech samples. Patients aged 60 and over followed up with the diagnosis of Minor Neurocognitive Disorder due to AD/ formerly Mild Cognitive Impairment (MCI) and major neurocognitive disorder due to mild or moderate AD according to DSM-5 and healthy elderly volunteers will be included to study. Healthy elderly volunteers will be invited to the study by advertisement, who do not have any neurological or psychiatric disorders, who do not use drugs, substances or alcohol that may affect cognitive functions, who do not have sensory defects that prevent the application of the tests, who have an SMMT score of 25 and above, who are 60 years of age and older will be included. Exclusion criteria will also be the presence of additional neurological or psychiatric disease in patient groups, and sensory deficits that prevent the application of tests. For all three groups, the criterion of having received at least 5 years of education will be sought. Patients with advanced AD diagnosis will not be included in the study. In addition, patients with a HAM-D score above 7 will be excluded from the study, since depression is thought to affect neuropsychological test performances. The sample size was calculated by evaluating the results of studies conducted with similar methods in the literature, and the effect size was accepted as 0.25 with the G\*Power3.1 program, and the sample size was calculated with a 95% confidence interval, 80% power and 5% margin of error. As a result of the analysis, the total sample size was calculated as 159; A total of 180 participants (120 patients, 60 healthy controls) were planned to be included in the study, including 60 m-NBB, 60 M-NBB due to AD, and 60 healthy controls.