Background and aims
Subjective cognitive decline (SCD) or mild cognitive impairment (MCI) may be associated with an increased risk of dementia later in life. Early preventive measures can potentially reduce the risk of dementia. This requires early identification of first signs of cognitive impairment.
Current assessment methods are often time-consuming and require a high level of personnel effort from the persons affected, their relatives and medical staff. Therefore, it makes sense to develop low-threshold and resource-conserving assessments that enable an early diagnosis of SCD or MCI. Language changes are a promising starting point for the identification of cognitive disorders. Therefore, the aims of this study are:
- the validation of the biomarker "language" by comparison with established biomarkers,
- the training of algorithms for the computer-assisted detection of dementia-specific changes in speech,
- testing user acceptance of computer-assisted assessment, and
- the reliability of assessment performance by artificial intelligence (AI) compared to human-assisted assessment.
Prospect-AD is a longitudinal controlled study of DZNE and the company ki elements funded by the Alzheimer's Drug Discovery Foundation. It is being conducted with the inclusion of the DELCODE- and DESCRIBE-cohorts. In order to identify early, scientifically robust language changes, speech will be recorded during telephone conversations and evaluated with respect to phonetic and grammatical parameters.All conversations will be conducted automatically by the software “Mili”. Data analysis is carried out by an AI. Thus, the acceptance and attitude of the participants concerning automated calls and AI in medicine will be investigated.
A maximum of six telephone conversations lasting approximately 15 to 20 minutes are planned over a period of 15 months. All calls will be con-ducted by an automated software similar to an automatic telephone announcement. A subsequent questionnaire or personal interview will record how the conversation with Mili was perceived. The data will be uploaded pseudonymized to a server of the DZNE and analyzed by the DZNE and the company ki elements.
Principle Investigator: Prof. Dr. Stefan Teipel
Start of the study: 2022
Status: multi centric, recruiting active