Paper reading and analysis: A Robust Motion Artifact Detection Algorithm for Accurate Detection of Heart Rates

论文阅读和分析：A Robust Motion Artifact Detection Algorithm for Accurate Detection of Heart Rates from Photoplethysmographic Signals using Time-Frequency Spectral Features

main content:

1. Extract the spectral features of the PPG signal, and then use the machine learning algorithm SVM to judge the motion artifact part of the PPG signal;
2. The method of feature extraction is relatively novel and worth learning;

flow chart:

Extracted features:

1. Use VFCDM to convert PPG signal to time spectrum;

2. Calculation features:

$P_{noise}$:
$$P_{noise}=P_{TFS}-\sum _{i=1}^3\sum _{t}A{M}_{i,t}$$
$d{f}_{FM}$:
$$d{f}_{FM}=\sum _{i=2}^3\sum _t|F{M}_{\mathrm{i},t}-i\times F{M}_{1,t}|$$
$d{f}_{HR}$:
$$d{f}_{HR}=\sum _t\left|F{M}_{1,t}-\mathrm{median}\left(\frac{1}{PP}\right)\right|$$
3. Use motion to cause motion artifacts as labels;

Experimental results:
Compared with several other methods, motion artifacts can be accurately identified:

reference:

A Robust Motion Artifact Detection Algorithm for Accurate Detection of Heart Rates from Photoplethysmographic Signals using Time-Frequency Spectral Features