JMIR Publications

ABSTRACT

Background:

Alzheimer's dementia (AD) is characterized by progressive cognitive decline with olfactory dysfunction emerging as one of its earliest symptoms. Current olfactory assessment methods rely on subjective self-reports and require patient cooperation, limiting their clinical utility. Non-invasive olfactory bulb (OB) recordings may provide objective measurements for early detection of cognitive decline.

Objective:

This study aimed to investigate the relationship between olfactory dysfunction and cognitive decline across the AD spectrum, assessing whether non-invasive OB recordings could serve as a digital biomarker for early AD detection and progression monitoring, thereby contributing to the development of mHealth and uHealth monitoring systems.

Methods:

We recruited 71 participants: cognitively normal individuals (n=18), mild cognitive impairment (MCI) patients subdivided into early MCI (n=30) and late MCI (n=9), and AD patients (n=14). Non-invasive OB recordings were collected during olfactory stimulation using a six-channel wearable system. Support vector machine regression was employed for classification. Diffusion tensor imaging and magnetic resonance imaging assessed structural brain changes, while comprehensive neuropsychological assessments evaluated cognitive function. Gamma band power (representing local OB activity) and beta-band power (reflecting top-down cortical projections) were analyzed to evaluate neural communication patterns.

Results:

Beta-band power progressively declined across the cognitive spectrum from normal to AD, while gamma band power decreased and response latencies increased beginning in late MCI. Machine learning classification achieved 83.09% accuracy, 88.64% precision, and 100% recall in distinguishing patient groups. Reduced beta-band power correlated with cingulum hippocampus diffusivity changes (R²=0.2831, p<0.0613), and decreased OB volume corresponded with diminished gamma band power in late MCI (R²=0.2108, p<0.000153). Notably, electrophysiological changes preceded volumetric reductions. The regression model demonstrated a strong relationship between olfactory measurements and cognitive performance (R²=0.9169, p=5.308×10⁻³⁹).

Conclusions:

Bidirectional interaction between the OB and higher-order brain regions is disrupted early in the AD continuum, with beta-band alterations occurring in early stages and gamma-band changes emerging later. Digital OB recordings provide insights into cognitive decline before significant structural changes occur, positioning them as promising digital biomarkers for early AD detection, progression monitoring, and potentially evaluating therapeutic efficacy in future interventional studies. Clinical Trial: The study protocol received approval from the Institutional Review Board of Hanyang University Medical Center (2023-06-019) and was conducted in accordance with the Declaration of Helsinki.