VALIDATION STUDY OF INTEGRATED COGNITIVE ASSESSMENT (ICA)
Research type
Research Study
Full title
A Validation Study of a computerized, Artificial Intelligence assisted cognitive assessment test compared to a widely used cognitive assessment test and specialist clinical diagnosis in patients with Alzheimer’s Dementia and Mild Cognitive Impairment
IRAS ID
213457
Contact name
Sina Habibi
Contact email
Sponsor organisation
Cognetivity Ltd
ISRCTN Number
ISRCTN95636074
Duration of Study in the UK
1 years, 2 months, 0 days
Research summary
Summary of Research
Neurodegenerative disorders, including Dementia and Alzheimer’s disease, continue to represent a major economic, social and healthcare burden.
The current methods of screening suffer from several drawbacks making them less effective for users and stakeholders, e.g. time consuming, costly, not sufficiently sensitivity for early diagnosis, dependent on education, language and demographic.
ICA is a computerised cognitive assessment tool based image recognition and operating on Apple iPad support. It utilises an artificial intelligence (AI) engine capable of improving its own sensitivity as the number of subjects who complete the test increases and the AI engine is updated. ICA is distinctive, as it does not rely on language or education while it is not subject to learning effects i.e. patients cannot memorise it after repeated use. It does not require specialist clinicians and importantly its duration is short (5-6 mins).
The overall aim of the study is to evaluate a 5-6-minute cognitive screening tool (Integrated Cognitive Assessment, ICA) to assist frontline physicians in detection of cognitive impairment associated with disease, such as Mild Cognitive Impairment (MCI) and Alzheimer's Dementia (AD).A validation study of against a widely used cognitive assessment test (Montreal Cognitive Assessment, MoCA) and specialist Memory Clinic diagnosis is proposed. The study will also explore the combined diagnostic power of the ICA combined with cerebrospinal fluid (CSF) biomarkers for the diagnosis of MCI sufferers. The study will be conducted in 3 NHS Diagnostic Memory Clinics. Patients suffering from MCI and Mild-AD will be assessed and diagnosed as usual but participants will take part in using the test in parallel to the diagnostic process. A control group of healthy volunteers will also be recruited. The study is expected to last fourteen (14) months.
Summary of Results
Lay summary of study results: Dementia has been recognised as a global phenomenon affecting millions of people. It is described by a gradual decline in brain functioning and early symptoms may include thinking problems, changes in personality, behaviours. Mild Cognitive Impairment (MCI), a pre-dementia condition, is poorly understood and under-diagnosed, which places additional strain on healthcare systems.Current tests used to diagnose cognitive impairment present multiple limitations, such as learning, language and education biases. They also cannot be used to detect the early changes linked to declining brain function. Finding out early can help people with dementia gain access to relevant information, medications and support systems so that both them and their families understand the disease and plan their futures accordingly.
The Integrated Cognitive Assessment CognICA™ is a 5-minute computerised screening tool developed to detect such changes. The test is quick, self-administered and independent of language and educational level. It also has the ability to “improve”, as more patients take the test,and become more accurate with each subsequent use. CognICA can be used to monitor changes in brain functioning, identify those at risk of developing MCI, and refer them to healthcare services.
Fourteen NHS Trusts participated in this clinical trial which took place in two phases.
In the first phase, 159 participants aged 55-90 years old took the test and three groups were established based on already existing information from their doctors - participants with Mild Cognitive Impairment, participants with mild Dementia and healthy volunteers. Participants in all three groups took the test so CognICA could identify the differences between healthy and cognitively impaired patients
In the second phase, 258 participants were recruited to complete both CognICA and the standard cognitive impairment tests used by the NHS. Participants’ performance was compared to determine which of the two is more sensitive, specific, easier to use and cost-effective.
We demonstrated that the ICA can reliably detect cognitive impairment in patients with MCI and mild AD, with a high test-retest reliability. As a computerised cognitive test, ICA test results are automatically calculated saving health care professionals time compared to pen-and-paper tests, as well as minimising rater effects. CognICA is also shown to be independent of language and culture. Additionally as a digital test, it has the potential for easier electronic health record or research database integration.
REC name
London - Dulwich Research Ethics Committee
REC reference
18/LO/0575
Date of REC Opinion
11 May 2018
REC opinion
Further Information Favourable Opinion