Back to search

University of the Aegean

Detection of cognitive impairment and enhancement of cognitive abilities using serious games and machine learning

Abstract

dc:description

Mild Cognitive Impairment (MCI) constitutes a critical transitional stage in the continuum of cognitive decline, during which timely, valid, and adaptive interventions can slow down or even reverse progression toward dementia. Despite significant advances in digital health, existing approaches to serious games for cognitive rehabilitation are characterized by theoretical fragmentation, limited multimodality, and a lack of empirically grounded design models. This dissertation addresses these gaps through the development and evaluation of CogniFrame, a novel, multidimensional, human-centered, and scientifically grounded theoretical framework for the design of serious games aimed at cognitive enhancement and assessment in older adults with MCI. The research methodology follows a multi-level approach that integrates theoretical synthesis, expert evaluation, clinical investigation, and computational analysis. CogniFrame is grounded in contemporary theories of cognitive neuroscience, learning principles, and human-centered design, structuring eight dimensions that guide the lifecycle of digital interventions. CogniFrame was empirically applied and evaluated through the development of the COGNIPLAT platform, whose effectiveness was examined via a pilot randomized controlled trial involving older adults with MCI, using standardized psychometric instruments (MoCA, RAVLT, DST, TMT, FAQ). In parallel, interaction data generated during gameplay were analyzed using machine learning algorithms (e.g., RF, SVM) to investigate their validity and reliability as digital biomarkers of cognitive function and cognitive impairment. The results demonstrate significant improvements in memory, attention, and executive functions among participants who used the COGNIPLAT platform, confirming the activation of cognitive mechanisms embedded in the design of CogniFrame. Furthermore, the machine learning models developed showed that interaction metrics can classify cognitive status with high accuracy and serve as reliable indicators for early detection. The integration of theoretical, methodological, and computational findings indicates that CogniFrame constitutes a comprehensive and empirically validated framework that bridges the gap between conceptual models and actionable, meaningful digital interventions. Overall, this dissertation provides a unified theoretical foundation, a validated design methodology, empirical evidence of cognitive improvement, and new insights into the use of digital biomarkers in individuals with MCI. Particular emphasis is placed on the dimension of detecting and monitoring cognitive status through machine learning techniques, which leverage low-burden interaction data to enable timely, reliable, and scalable assessment of cognitive function. The findings indicate that serious games—when designed based on CogniFrame—can function simultaneously as tools for cognitive rehabilitation and as adaptive, data-driven systems for continuous cognitive monitoring and early detection.

Degree

thesis:*
Grantor dc:publisher
University of the Aegean
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Skikos, Georgios
  • Σκίκος, Γεώργιος

Subjects

dc:subject × 13

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/61498

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Skikos, Georgios; Σκίκος, Γεώργιος. Detection of cognitive impairment and enhancement of cognitive abilities using serious games and machine learning. University of the Aegean, 2026. http://hdl.handle.net/10442/hedi/61498