Introduction: Neurodegenerative disorders, such as Alzheimer’s Disease (AD), continue to present a major economic, social and healthcare burden. These diseases remain underdiagnosed or are diagnosed too late for meaningful interventions. The development of screening tests capable of detecting AD during early, preferably asymptomatic, stages has been a highly unmet need. Since such tests will be used for screening large populations of people, they should be non-invasive, inexpensive, and ideally independent of language, education, culture and practice.
Methods: Taking advantage of new artificial intelligence (AI) and machine learning techniques, we have developed a 5-minute integrated cognitive assessment tool (ICA) that meets the above-mentioned criteria, and has the flexibility to learn from new data to improve its predictive power. The ICA is computerised, can be conducted without expert supervision, and is designed based on a rapid visual categorisation task, tracking participants’ response-patterns to natural stimuli to detect small changes in their cognitive performance.
- The ICA engages brain areas affected in early stages of AD, and shows high sensitivity in detecting cognitive impairment.
- The ICA is self administered and language independent, and as such the test can be used as an aid for early diagnosis of AD, ideally even in pre-symptomatic stages, and is appropriate for large- scale screening of cognitive impairment, and micro-monitoring of cognitive performance.
- The ICA has demonstrated to be free from a learning bias (i.e. practice effect).
- As an iPad test, the ICA has potential for integration with EHR and electronic medical record or research database integration.
- The above attributes yield significant clinical benefits in the day-to-day identification of MCI and AD in specialist clinical settings, in primary care and in remote cognitive monitoring.