Presentation Information

[TuE1-3]A Non-contact Sleep Analysis System Based on Transformer and Optical Fiber Sensor

○Yanyao Chen1, Weimin Lyu1, Shijun Xiao2, Changyuan Yu1,2 (1The Hong Kong Polytechnic Univ. Jinjiang Technology and Innovation Research Institute, 2The Hong Kong Polytechnic Univ.)

Keywords:

Fiber sensing and other applications

This paper introduces a non-contact sleep monitoring system based on a fiber optic Mach-Zehnder Interferometer (MZI). The proposed Noisy-label-learning Transformer algorithm is used to classify sleep stages, achieving an accuracy of 82.65%.