Smart Insole-Based Classification of Alzheimer's Disease Using Few-Shot Learning Facilitated by Multi-Scale Metric Learning

Alzheimer's disease is a progressive brain disorder, with mild cognitive impairment being a predisposing stage. Although various diagnostic methods have been proposed, they are difficult to use regularly due to associated pain and costs. In addition, small medical dataset problems due to cost a...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on consumer electronics Vol. 70; no. 2; pp. 4699 - 4708
Main Authors Jeon, Younghoon, Kang, Jaeyong, Kim, Byeong C., Ho Lee, Kun, Song, Jong-In, Gwak, Jeonghwan
Format Journal Article
LanguageEnglish
Published IEEE 01.05.2024
Subjects
Accelerometers
Alzheimer’s disease
Consumer electronics
deep learning
few-shot learning
gait
Gyroscopes
Legged locomotion
Measurement
metric learning
Sensors
smart insole
Task analysis
Online AccessGet full text

Cover

More Information
Summary:Alzheimer's disease is a progressive brain disorder, with mild cognitive impairment being a predisposing stage. Although various diagnostic methods have been proposed, they are difficult to use regularly due to associated pain and costs. In addition, small medical dataset problems due to cost and ethical issues are challenging for artificial intelligence. Therefore, we propose multilevel gait experiment paradigms as diagnostic tools and a few-shot learning-based diagnostic model, facilitated by metric learning, to address the issues related to small data. In this study, two types of gait datasets from 69 subjects were acquired using a smart insole for various performance evaluation experiments. Experimental results showed that our proposed model improved with the increasing difficulty of paradigms and outperformed conventional deep metric learning methods.
ISSN:0098-3063
1558-4127
DOI:10.1109/TCE.2024.3386714