Citation

Ashiri M, Francisco C, Winkler J, Lithgow B, Moussavi Z. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. 2021; 2021: 7219-7222.

Copyright

(Copyright © 2021, IEEE (Institute of Electrical and Electronics Engineers))

DOI

10.1109/EMBC46164.2021.9630746

PMID

34892765

Abstract

The vestibular system, responsible for balance, is affected by Alzheimer's disease (AD). In this paper, linear and non-linear balance features were used to assess the postural stability of 13 AD individuals at mild stages in comparison with 16 healthy controls. Utilizing two accelerometers, the anterior-posterior (AP) and medial-lateral (ML) sways were recorded from the T2 vertebrae and lateral malleolus of participants standing on a solid and soft foam surface under both eyes-open and eyes-closed conditions. From the recorded signals, four features were extracted and used for statistical analysis: Number of Position Changes (NPC), Number of Non-Zero Accelerations (NNZA), Katz, and Higuchi fractal dimensions (KFD and HFD, respectively). The results show: 1) postural stability is significantly worse for the eyes-closed compared to eyes-open condition (P<0.05 for all features except HFD) as well as whilst standing on soft foam compared to the solid surface (P<0.05 for all features) in both groups; 2) balance perturbations were larger for AP sway than ML on both solid and foam surfaces in both groups (P<0.05 for NPC and NNZA); and 3) stationary balance is significantly poorer for AD individuals compared to controls (P<0.05 for all features). These observations show that both linear and non-linear characteristics of postural stability data have the potentials to be used as objective diagnostic aids for the detection of AD.

Language: en