Telemedicine Self-examination of Speech and Memory for Rapid Detection of Cognitive Impairments Using Machine Learning Methods

Telemedicine Self-examination of Speech and Memory for Rapid Detection of Cognitive Impairments Using Machine Learning Methods

Early cognitive disorders diagnosis is becoming increasingly important due to population aging. The most common causes include Alzheimer's disease and frontotemporal dementia. These diseases are also manifested by changes in speech. NLP allows us to identify and classify these changes. The project aims to develop a web application for self-assessment and automated detection of cognitive disorders from speech. The application will have a form of a dialogue system using machine learning methods. The novelty of this approach is the possibility of an efficient self-assessment of a wide spectrum of the Czech population from their homes and an automated evaluation of test results. Early detection can be followed by a more detailed diagnosis and adequate treatment.

Inclusion Criteria for the group of healthy participants:

• age over 45 years
• Czech as the native language
• must meet the criteria in a questionnaire with questions focused on a history of brain injury or psychiatric history or medication use
• must achieve normal scores on questionnaires assessing mood

Inclusion Criteria for the group of patients:

- be diagnosed with mild neurocognitive impairment (MCI) or neurocognitive disorder (dementia) according to the international DSM-5 criteria