TY - GEN
T1 - Early Detection of Alzheimer's Disease Through Eye Movement Analysis
T2 - 2024 IEEE International Workshop on Electromagnetics: Applications and Student Innovation Competition, iWEM 2024
AU - Lin, Yu Chun
AU - Huang, Li Kai
AU - Wu, Jeng Chian
AU - Chang, Tai Ying
AU - Hu, Hsiang Wei
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Given the increasing worldwide incidence of dementia, primarily Alzheimer's disease dementia, there is an urgent requirement for non-invasive, early diagnostic techniques. This study introduces a novel approach utilizing eye movement analysis to detect early signs of cognitive decline indicative of dementia disorders. By analyzing the eye movement patternsof 95 participants, including 31 individuals with dementia and 64 cognitively normal controlsubjects, during tasks involving picture description and reading activities, distinct eye movement characteristics were identified. Using the Tobii Eye Tracker 5 for precise tracking and ASUS tablets for display, the study leveraged advanced machine learning algorithms such as XGBoost, logistic regression, and deep neural networks (DNNs) to analyze data. Significanteye movement features differentiated people with dementia from controls, indicating potential as early biomarkers. This approach demonstrated that digital eye tracking technologies combined with machine learning could offer a rapid, cost-effective solution for early all causedementia diagnosis, promising substantial improvements in patient care and management strategies. The outcomes suggest the feasibility of this method in clinical settings, highlightingthe importance of further research with larger sample sizes to refine the diagnostic models.
AB - Given the increasing worldwide incidence of dementia, primarily Alzheimer's disease dementia, there is an urgent requirement for non-invasive, early diagnostic techniques. This study introduces a novel approach utilizing eye movement analysis to detect early signs of cognitive decline indicative of dementia disorders. By analyzing the eye movement patternsof 95 participants, including 31 individuals with dementia and 64 cognitively normal controlsubjects, during tasks involving picture description and reading activities, distinct eye movement characteristics were identified. Using the Tobii Eye Tracker 5 for precise tracking and ASUS tablets for display, the study leveraged advanced machine learning algorithms such as XGBoost, logistic regression, and deep neural networks (DNNs) to analyze data. Significanteye movement features differentiated people with dementia from controls, indicating potential as early biomarkers. This approach demonstrated that digital eye tracking technologies combined with machine learning could offer a rapid, cost-effective solution for early all causedementia diagnosis, promising substantial improvements in patient care and management strategies. The outcomes suggest the feasibility of this method in clinical settings, highlightingthe importance of further research with larger sample sizes to refine the diagnostic models.
KW - All cause dementia
KW - Alzheimer's Disease
KW - Deep Neural Network
KW - Digital Diagnostics
KW - Early Diagnosis
KW - Eye Movement Analysis
KW - Machine Learning Models
UR - http://www.scopus.com/inward/record.url?scp=85203823511&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85203823511&partnerID=8YFLogxK
U2 - 10.1109/iWEM59914.2024.10649039
DO - 10.1109/iWEM59914.2024.10649039
M3 - Conference contribution
AN - SCOPUS:85203823511
T3 - IEEE International Workshop on Electromagnetics: Applications and Student Innovation Competition, iWEM 2024
BT - IEEE International Workshop on Electromagnetics
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 10 July 2024 through 12 July 2024
ER -