Journal Article

ReCOGnAIze app to detect vascular cognitive impairment and mild cognitive impairment

Vipin, Ashwati, Tan, Farid, Mohammed, Adnan Azam, Leow, Yi Jin, Aravindhan, Kiirtaara, Agarwal, Hitesh, Kandiah, Nagaendran, Sandhu, Gurveen Kaur, Tanoto, Pricilia, Setiabudi, Eliana, Liew, Shan Yao, Qiu, Bocheng, Liau, Kai Xin, Wang, Jia Dong James

2026

Overview

INTRODUCTION Vascular cognitive impairment (VCI), a major cause of cognitive impairment, remains underdiagnosed due to varying non‐amnestic manifestations. It is important to detect VCI at the mild cognitive impairment (MCI) stage or earlier. We aimed to develop and validate ReCOGnAIze, a tablet‐based, gamified, and interpretable app to detect VCI and MCI. METHODS A multi‐phase, cross sectional study in an Asian community cohort with development phase (n = 200) and validation with 235 independent participants having comprehensive neuroimaging and neuropsychological data. RESULTS In differentiating VCI, ReCOGnAIze achieved strong performance (n = 154, AUC = 0.85), identifying digital features: processing speed and response time variability, consistent with known VCI impairments of executive functioning. Additionally, a generalizable ReCOGnAIze composite score distinguished MCI from cognitively healthy (CH) (n = 235, AUC = 0.90), outperforming the Montreal Cognitive Assessment (MoCA) (AUC = 0.70). DISCUSSION ReCOGnAIze is a scalable, explainable artificial intelligence (AI) tool that accurately detects VCI and MCI, with gamified, tablet‐based, interpretable tasks. Highlights Non‐significant differences on Montreal Cognitive Assessment (MoCA) for vascular cognitive impairment (VCI). ReCOGnAIze artificial intelligence (AI) models identify VCI with area under the curve (AUC) of 0.85. ReCOGnAIze games detect mild cognitive impairment (MCI) with AUC of 0.90, outperforming MoCA (AUC = 0.7). Processing speed and response time variability are key VCI markers.

• Publisher: John Wiley & Sons, Inc,Wiley
• Subject: Aged / Alzheimer's disease / Artificial Intelligence / cerebrovascular disease / Cognition / Cognitive ability / Cognitive Dysfunction - diagnosis / Cognitive impairment / cognitive screening / Cross-Sectional Studies / Dementia / Diabetes / digital biomarkers / digital cognitive assessment / early detection / executive dysfunction / Executive function / Female / Gamification / Humans / machine learning / Magnetic resonance imaging / Male / Medical imaging / Memory / Middle Aged / Mild cognitive impairment / Mobile Applications / Neuroimaging / neuropsychological testing / Neuropsychological Tests - statistics & numerical data / Neuropsychology / Reaction time / ReCOGnAIze / Semantics / Variability / vascular cognitive impairment