Predicting cognitive test scores from narrated stories using AI
Prediction of Cognitive Test Performance Using AI-Based Analysis of Narrative Speech
NA · Chang Gung Memorial Hospital · NCT07511270
We will test whether an AI program can estimate standard cognitive test scores from short recorded storytelling tasks in people aged 50 and older with memory or thinking concerns.
Quick facts
| Phase | NA |
|---|---|
| Study type | Interventional |
| Enrollment | 150 (estimated) |
| Ages | 50 Years and up |
| Sex | All |
| Sponsor | Chang Gung Memorial Hospital (other) |
| Locations | 1 site (Taoyuan) |
| Trial ID | NCT07511270 on ClinicalTrials.gov |
What this trial studies
Participants aged 50 and older will complete a brief storytelling task based on an image while their voice is recorded using a computer or mobile device. The recorded speech will be processed by an AI system that extracts speech and language features to generate estimates corresponding to common cognitive tests. Trained researchers will also administer standard cognitive assessments to each participant, and the AI-generated scores will be compared with these reference test results. The comparison will determine how closely the speech-based estimates match established cognitive measures and whether the method could be practical for screening or monitoring.
Who should consider this trial
Good fit: Ideal candidates are adults aged 50 or older who can hear and speak clearly, can follow simple instructions, can provide informed consent, and have subjective cognitive complaints or mild cognitive impairment without major neurological or psychiatric illness.
Not a fit: People with severe speech or hearing impairments, major neurodegenerative diseases (like advanced Parkinson's), recent stroke or traumatic brain injury, major psychiatric disorders, or current medications that substantially affect cognition are unlikely to benefit from this method.
Why it matters
Potential benefit: If successful, this approach could offer a convenient, noninvasive way to screen for or monitor cognitive changes using brief voice recordings.
How similar studies have performed: Previous research has shown promising early results linking speech and language features to cognitive impairment, but voice-based AI tests are not yet fully validated as standalone diagnostic tools.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: 1. Age 50 years or older 2. Able to communicate verbally and understand study instructions 3. Able to hear and speak clearly 4. Willing and able to provide written informed consent Exclusion Criteria: 1. Severe speech or hearing impairment 2. History of major psychiatric disorders 3. History of stroke, traumatic brain injury, Parkinson's disease, epilepsy, or other neurodegenerative diseases 4. Current use of medications that may affect cognitive performance 5. Unable to complete the study procedures due to severe attention deficits or agitation
Where this trial is running
Taoyuan
- Taoyuan — Taoyuan, Taiwan (RECRUITING)
Study contacts
- Study coordinator: Ching-Yi WU, ScD
- Email: cywu@mail.cgu.edu.tw
- Phone: +886-3-2118800
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.
Conditions: Cognitive Function Assessment, Mild Cognitive Impairment, Subjective Cognitive Decline, Artificial Intelligence, Speech Biomarkers, Storytelling Task, Cognitive Screening, Voice Analysis