Abstract:AbstractBackgroundAlzheimer’s disease (AD) is a progressive and debilitating neurological disorder that impacts millions of individuals globally. Prompt and accurate detection and diagnosis of Alzheimer’s disease are pivotal for providing effective care and management to patients. However, early identification of the disease remains challenging due to the complexity of diagnostic procedures and the similarity of symptoms with age‐related cognitive decline. With the world transitioning towards a more digital‐focused environment, there is an increasing potential to utilize human‐computer interaction technologies to aid in the early detection of Alzheimer’s disease.MethodWe recruited 40 healthy individuals, 40 patients with mild cognitive impairment (MCI), and 40 patients with early AD. By incorporating fusion perception‐enhancing technologies into intelligent identification devices, such as eye‐tracking and body motion detection, peripheral modal biometric information can be gathered from subjects.ResultThe study found that in human‐computer interaction experiments using eye‐tracking, the patients with early AD showed significantly longer completion times and free rates of eye motion and movement trajectory than the normal group and those with MCI. The correlation between eye motion trajectory and movement trajectory was notably lower as compared to the normal group and the MCI group. During the human‐computer interaction physical movement experiment, the AD group recorded lower target accuracy rate and screen click rate as compared to the normal group and patients with MCI. Similarly, in the human‐computer interaction clock drawing experiment, the AD group demonstrated longer completion time and pen‐up thinking time, as well as lower clock drawing scores. Furthermore, the total completion time of human‐computer interaction tasks was notably shorter as compared to conventional cognitive screening tests. In conclusion, peripheral biomarkers obtained from the aforementioned human‐computer interaction task were analyzed using machine learning techniques, resulting in the identification of early‐stage Alzheimer’s disease with a classification accuracy of 91.1%.ConclusionHence, this non‐invasive and accurate technique may prove instrumental in identifying potential cognitive deficits associated with this disease while also being widely applicable for screening early AD.

Diagnosis Framework for Probable Alzheimer's Disease and Mild Cognitive Impairment Based on Multi-Dimensional Emotion Features

Digitized Biomarkers Utilizing Human‐Computer Interaction Sensing Technology for Early Auxiliary Diagnosis of Alzheimer’s Disease

A Multimodal Learning Machine Framework for Alzheimer’s Disease Diagnosis Based on Neuropsychological and Neuroimaging Data

Extended Application of Digital Clock Drawing Test in the Evaluation of Alzheimer's Disease Based on Artificial Intelligence and the Neural Basis

A Multi-Modal Classification Method for Early Diagnosis of Mild Cognitive Impairment and Alzheimer's Disease Using Three Paradigms With Various Task Difficulties

Develop a Diagnostic Tool for Dementia Using Machine Learning and Non-Imaging Features

Deep Convolution Network Based Emotion Analysis for Automatic Detection of Mild Cognitive Impairment in the Elderly

Alzheimer’s Disease Diagnosis Via Multimodal Feature Fusion

Classifying Mild Cognitive Impairment from Behavioral Responses in Emotional Arousal and Valence Evaluation Task - AI Approach for Early Dementia Biomarker in Aging Societies

Developing a machine learning model for detecting depression, anxiety, and apathy in older adults with mild cognitive impairment using speech and facial expressions: A cross-sectional observational study

A Feature-Aware Multimodal Framework with Auto-Fusion for Alzheimer’s Disease Diagnosis

A feature-aware multimodal framework with auto-fusion for Alzheimer's disease diagnosis

Multimodal Identification of Alzheimer's Disease: A Review

Toward Robust Early Detection of Alzheimer's Disease via an Integrated Multimodal Learning Approach

Patch-based Interpretable Deep Learning Framework for Alzheimer’s Disease Diagnosis Using Multimodal Data

Fully Automated Discrimination of Alzheimer's Disease Using Resting-State Electroencephalography Signals.

Automated analysis of facial emotions in subjects with cognitive impairment

An explainable machine learning based prediction model for Alzheimer's disease in China longitudinal aging study

Automatic Detection of Alzheimer's Disease with Multi-Modal Fusion of Clinical MRI Scans

Machine Learning of Cerebello-Cerebral Functional Networks for Mild Cognitive Impairment Detection

Classification of Alzheimer's Disease Based on Abnormal Hippocampal Functional Connectivity and Machine Learning