“School of Cognitive%20Sciences”

Back to Papers Home
Back to Papers of School of Cognitive%20Sciences

Paper   IPM / Cognitive%20Sciences / 16485
School of Cognitive Sciences
  Title:   Employing Artificial Intelligence in the Development of a self-administered, computerized cognitive assessment for the assessment of neurodegeneration
  Author(s): 
1.  C. Kalafatis
2.  M. Modarres
3.  H. Marefat
4.  M. Khanbagi
5.  H. Karimi
6.  Z. Vahabi
7.  S. Khaligh Razavi
  Status:   In Proceedings
  Proceeding: Alzheimer�??s & Dementia, Alzheimer�??s Association International Conference (AAIC), 2019 July 14-18
  Year:  2019
  Supported by:  IPM
  Abstract:
Background We developed the Integrated Cognitive Assessment (ICA), a novel, computerised cognitive assessment test that aims at screening for cognitive impairment in a way that can simplify and accelerate that diagnosis of Alzheimer's Dementia (AD) and Mild Cognitive Impairment (MCI) .The test is self�?�administered, takes approximately five minutes and primarily focuses on measuring speed of information processing and utilises artificial intelligence in order to improve its predictive power.The ICA is independent of language and education and is free from learning bias (i.e. practice effect).
Methods We carried a head�?�to�?�head comparison of the ICA against the Montreal Cognitive Assessment (MoCA) and the Addenbrooke's Cognitive Assessment (ACE�?�III) in 69 participants of which 19 had a diagnosis of Mild AD, 21 had a diagnosis of MCI and 29 healthy volunteers where participants' age range varied from 50 to 83 years (mean= 67.1 years; Standard Deviation= 7.7 years).
Results ICA's AI component (AI�?�engine) utilised a logistic regression model and yielded an Area�?�under�?�the�?�Receiver�?�Operating�?�Characteristic�?�Curve (AUC) score of 81.7
Conclusions ICA showed convergent validity with MoCA and ACE�?�III, with higher accuracy in detecting cognitive impairment compared to MoCA and ACE�?�III. The above attributes yield significant clinical benefits in the screening of MCI and AD, both in primary care and in specialist clinical settings. ICA has advantages over MoCA and ACE�?�III because of its efficient administration, shorter duration, automatic scoring, language independency, capacity for very regular monitoring of the trajectory of disease progression and potential for medical record or research database integration.

Download TeX format
back to top
scroll left or right