An observational study in Mild Cognitive Impairment (MCI) and Alzheimer Disease (AD), sponsored by Centre for Addiction and Mental Health. Not yet recruiting at 1 site in Canada. Open to participants aged 50 Years and older, including healthy volunteers. Per ClinicalTrials.gov, last updated 2026-10-20.
Sponsored by Centre for Addiction and Mental Health · Observational
Memory and thinking problems in older adults, including mild cognitive impairment (MCI) and Alzheimer's disease (AD), can be difficult to assess promptly because evaluation often depends on in-person specialist visits. MonitorCog-AI is a computer-based platform designed to provide a more accessible way to assess and monitor cognitive changes using standard computers or tablets. Artificial intelligence (AI) will be used to analyze the digital information collected by the platform and examine whether combining several types of measures improves assessment of cognitive impairment.
The study will include 60 adults aged 50 years and older: 20 healthy adults, 20 adults with MCI, and 20 adults with AD. Participants will be assessed at baseline, 6 months, and 12 months. At each timepoint, participants will complete one clinical assessment and one digital assessment. The clinical assessment will include standard measures of memory and thinking, an assessment of daily functioning, and a brief recording of brain electrical activity. The digital assessment will take approximately 60-80 minutes and will assess five areas: cognition, eye movements measured using a webcam, language, speech, and fine-motor performance measured using a mouse. At baseline, the digital assessment will be completed in the clinic; at 6 and 12 months, it will usually be completed remotely at home.
The study has three main goals. First, it will determine whether MonitorCog-AI is feasible, acceptable, easy to use, and not overly burdensome for repeated assessment. Second, it will examine how well AI-based analysis of the digital measures distinguishes healthy adults, adults with MCI, and adults with AD, and whether combining several types of digital information performs better than any single type alone. Third, it will examine how digital measures change over 12 months and whether these changes are related to changes in clinical measures of cognition and daily functioning.
Current approaches to assessing mild cognitive impairment (MCI) and Alzheimer's disease (AD) rely largely on periodic clinic-based cognitive testing, which may provide only a limited view of change over time. There is a need for practical approaches that can repeatedly capture complementary aspects of cognitive and behavioural function outside the clinic. MonitorCog-AI combines digital measures of cognition, language, speech, eye movements, and motor function to address this need.
MonitorCog-AI is a computer-based platform that integrates five types of digital assessment: cognition, webcam-based eye movements, language, speech, and mouse-based fine-motor performance. Artificial intelligence (AI) and machine-learning methods will be used to analyze and combine the digital information collected by the platform.
The study will enroll 60 adults aged 50 years or older: 20 healthy adults, 20 adults with MCI, and 20 adults with AD. Participants will complete assessments at baseline, 6 months, and 12 months. At each timepoint, participants will complete a clinical assessment and a digital assessment. The clinical assessment will include standard measures of memory and thinking, daily functioning, and a brief electroencephalogram (EEG) recording. The digital assessment will include the five MonitorCog-AI modules. At baseline, the digital assessment will be completed in the clinic; at 6 and 12 months, it will usually be completed remotely at home.
The study objectives and corresponding hypotheses are as follows:
Objective 1: To evaluate the feasibility, acceptability, usability, and participant-perceived burden of repeated MonitorCog-AI assessments.
Hypothesis 1a: At least 80% of participants will complete the baseline assessment and at least one follow-up assessment.
Hypothesis 1b: At least 75% of participants will report an overall satisfaction rating of 4 or higher on a 5-point scale.
Objective 2: To evaluate the preliminary diagnostic performance of the digital measures for distinguishing healthy adults, adults with MCI, and adults with AD, and to determine whether combining multiple digital modalities improves discrimination compared with individual modalities.
Hypothesis 2a: The multimodal model will achieve an area under the receiver operating characteristic curve (AUC) of at least 0.75 for distinguishing healthy adults from adults with MCI and MCI from AD, and at least 0.85 for distinguishing healthy adults from adults with AD.
Hypothesis 2b: The multimodal model will improve AUC by at least 0.05 compared with the best-performing single-modality model.
Objective 3: To examine changes in digital measures over 12 months and their relationships with changes in clinical measures of cognition and daily functioning.
Hypothesis 3: Changes in digital measures over 12 months will be associated with changes in standard clinical measures of cognition and daily functioning.
3,930 studies on the registry are indexed under Cognitive Dysfunction; 1,114 are open to participants now.
This study's planned enrollment of 60 is below the median of 150 across 989 observational studies indexed under Cognitive Dysfunction.
Browse Cognitive Dysfunction studies →Centre for Addiction and Mental Health is the lead sponsor of 339 studies on the registry; 51 are open to participants now.
Counted across the registry records on this site, refreshed daily.
The study will enroll 60 adults aged 50 years or older across three groups: 20 healthy controls, 20 individuals with mild cognitive impairment (MCI), and 20 individuals with Alzheimer's disease (AD). Participants will be recruited from CAMH Geriatric Mental Health Services clinics, existing research registries, and the community through clinician referral and advertising. Diagnostic group assignment will be determined by the principal investigator or another qualified clinician using cognitive and functional assessments, clinical information, available informant information, and DSM-5 criteria.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Exclusion Criteria:
Adults aged 50 years or older without mild cognitive impairment or Alzheimer's disease who complete the study assessments at baseline, 6 months, and 12 months.
Adults aged 50 years or older with mild cognitive impairment (MCI) who complete the study assessments at baseline, 6 months, and 12 months.
Adults aged 50 years or older with Alzheimer's disease (AD) who complete the study assessments at baseline, 6 months, and 12 months.
Participant Completion Rate
Proportion of enrolled participants who complete the baseline assessment and at least one follow-up assessment at 6 or 12 months. A completion rate of at least 80% is the prespecified feasibility benchmark.
Time frame: Baseline through 12 months
System Usability Scale (SUS) Score
Usability of MonitorCog-AI will be measured using the 10-item System Usability Scale (SUS). Item responses are scored and transformed to a total score ranging from 0 to 100, with higher scores indicating better perceived system usability. Total scores will be summarized at baseline, 6 months, and 12 months.
Time frame: Baseline, 6 months, and 12 months
Participant Acceptability of MonitorCog-AI
Participant acceptability will be measured as the percentage of participants reporting an overall satisfaction rating of 4 or 5 on a study-specific 5-point satisfaction scale. Scores range from 1 to 5, with higher scores indicating greater satisfaction. At least 75% reporting a rating of 4 or higher is the prespecified acceptability benchmark.
Time frame: Baseline, 6 months, and 12 months
User Burden Scale (UBS) Score
Participant-perceived burden of completing MonitorCog-AI will be measured using the 20-item User Burden Scale (UBS), which assesses difficulty of use, physical burden, time and social burden, mental and emotional burden, privacy burden, and financial burden. Each item is scored from 0 to 4, yielding a total score ranging from 0 to 80, with higher scores indicating greater perceived burden. Total scores will be summarized at baseline, 6 months, and 12 months.
Time frame: Baseline, 6 months, and 12 months
Diagnostic Performance of MonitorCog-AI
Diagnostic performance of the AI-based multimodal model will be evaluated for healthy controls versus mild cognitive impairment (MCI), healthy controls versus Alzheimer's disease (AD), and MCI versus AD. Performance will be summarized using area under the receiver operating characteristic curve (AUC) with 95% confidence intervals, sensitivity, specificity, and calibration measures where estimable.
Time frame: Baseline
Incremental Performance of the Multimodal AI Model
The diagnostic performance of the multimodal AI model will be compared with the best-performing single-modality model for each diagnostic contrast. Differences in area under the receiver operating characteristic curve (AUC) will be estimated with 95% confidence intervals.
Time frame: Baseline
Change From Baseline in SATURN Total Score at 6 Months
Digital cognitive performance will be assessed using the Self-Administered Tasks Uncovering Risk of Neurodegeneration (SATURN). Total scores range from 0 to 30 points, with higher scores indicating better cognitive performance. Change from baseline will be calculated as the 6-month total score minus the baseline total score. The possible change ranges from -30 to +30 points.
Time frame: Baseline and 6 months
Change From Baseline in SATURN Total Score at 12 Months
Digital cognitive performance will be assessed using the Self-Administered Tasks Uncovering Risk of Neurodegeneration (SATURN). Total scores range from 0 to 30 points, with higher scores indicating better cognitive performance. Change from baseline will be calculated as the 12-month total score minus the baseline total score. The possible change ranges from -30 to +30 points.
Time frame: Baseline and 12 months
Correlation Between Changes in SATURN and Montreal Cognitive Assessment Scores at 6 Months
The correlation between change from baseline to 6 months in digital cognition, measured using the Self-Administered Tasks Uncovering Risk of Neurodegeneration (SATURN) total score, and clinical cognition, measured using the Montreal Cognitive Assessment (MoCA) total score, will be estimated using Spearman's rank correlation coefficient (rho). Both instruments have total scores ranging from 0 to 30 points, with higher scores indicating better cognitive performance. Change scores are calculated as 6-month scores minus baseline scores. Spearman's rho is unitless and ranges from -1 to +1.
Time frame: Baseline and 6 months
Correlation Between Changes in SATURN and Montreal Cognitive Assessment Scores at 12 Months
The correlation between change from baseline to 12 months in digital cognition, measured using the Self-Administered Tasks Uncovering Risk of Neurodegeneration (SATURN) total score, and clinical cognition, measured using the Montreal Cognitive Assessment (MoCA) total score, will be estimated using Spearman's rank correlation coefficient (rho). Both instruments have total scores ranging from 0 to 30 points, with higher scores indicating better cognitive performance. Change scores are calculated as 12-month scores minus baseline scores. Spearman's rho is unitless and ranges from -1 to +1.
Time frame: Baseline and 12 months
Plan to share: Undecided
No publications or documents are linked to this record.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
This study is not yet recruiting, as verified in Sep 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
Centre for Addiction and Mental Health →