Open Access

A Convolutional Neural Network-based Automatic Identification and Intervention Model for Health Surveillance Data during Postpartum Recovery Periods

Yanli Wang

Yanli Wang

Nursing College, Wuwei Occupational CollegeWuwei, China
Search for this author on
Sciendo|Google Scholar
Wang, Yanli
  
Mar 21, 2025
A Convolutional Neural Network-based Automatic Identification and Intervention Model for Health Surveillance Data during Postpartum Recovery Periods's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Mar 21, 2025
Received: Oct 15, 2024
Accepted: Feb 01, 2025
DOI: https://doi.org/10.2478/amns-2025-0580
Keywords
© 2025 Yanli Wang, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

The production of PPG
The production of PPG

Figure 2.

The PPG signal conventional algorithm crest spot recognition
The PPG signal conventional algorithm crest spot recognition

Figure 3.

PPG signal dynamic threshold point recognition
PPG signal dynamic threshold point recognition

Figure 4.

PPG signal wave valley recognition
PPG signal wave valley recognition

Figure 5.

PPG signal and APG signal wave peak recognition effect
PPG signal and APG signal wave peak recognition effect

Figure 6.

The heart rate and the Pearson correlation coefficient of the heart rate
The heart rate and the Pearson correlation coefficient of the heart rate

Figure 7.

Reference blood oxygen and forecast blood oxygen contrast
Reference blood oxygen and forecast blood oxygen contrast

Figure 8.

Monitoring system hardware flow chart
Monitoring system hardware flow chart

Figure 9.

System software architecture diagram
System software architecture diagram

The correlation between feature parameters and SBP and DBP

Eigenvalue Expression SBP correlation DBP correlation
Period of contraction t1/t -0.2547 -0.2185
Diastolic time ratio t2/t -0.6122 -0.5746
The relative height of the hypotension h1/h -0.3261 -0.1435
The relative height of the double bobo h2/h -0.4512 -0.4428
K1 SABCU/SAFHU -0.0754 -0.0561
K2 SUCDE/SUHIE -0.0268 -0.0032

Psychological indicators test results

Factor Experimental Group (N= 20) Control Group (N= 20)
Preexperiment After The Experiment Preexperiment After The Experiment
Somatization 1.26±0.13 1.09±0.58* 1.38±0.14 1.3±0.21
Obsessive-Compulsive Disorder 1.52±0.36 1.28±0.51** 1.52±0.22 1.51±0.36
Interpersonal Sensitivity 1.45±0.3 1.32±0.21 1.58±0.42 1.54±0.6
Depression 1.61±0.37 1.06±0.67** 1.65±0.64 1.62±0.44△△
Anxiety 1.24±0.23 1.12±0.13* 1.47±0.52 1.42±0.22
Antagonism 1.66±0.35 1.43±0.63** 1.56±0.6 1.5±0.37△
Horror 1.18±0.65 1.07±0.51 1.27±0.25 1.26±0.62
Paranoia 1.33±0.62 1.28±0.15** 1.35±0.67 1.34±0.38*
Insanity 1.26±0.51 1.14±0.67 1.18±0.52 1.15±0.14
Other 1.42±0.14 1.18±0.48 1.36±0.51 1.34±0.2

The different algorithms are on the 12 data

Algorithm TROKIA JOSS SPECTPAP COMB Particle OSC-ANFc Ours
1 2.01 1.71 1.02 2.41 1.43 1.45 1.36
2 2.84 1.75 2.54 3.46 1.44 1.56 1.32
3 2.48 1.34 0.38 0.02 1.57 1.9 0.49
4 2.87 1.59 1.84 1.63 1.22 1.19 1.25
5 2.12 0.59 0.92 0.47 0.11 0.6 0.71
6 2.78 1.13 1.25 1.71 1.18 1.32 0.01
7 1.88 0.27 1.81 2.52 0.09 0.27 0.7
8 1.31 0.73 0.45 0.85 1.61 1.66 0.46
9 1.09 0.2 0.73 0.54 0.83 0.23 0.32
10 4.13 3.86 3.06 4.2 2.54 2.88 3.95
11 1.95 0.25 0.92 1.52 1.79 1.53 0.2
12 2.87 1.87 1.35 1.13 1.18 1.28 1.17
Mean(SD) 2.361(2.75) 1.274(2.65) 1.356(1.89) 1.705(NA) 1.249(1.75) 1.323(NA) 0.995(1.32)

AAE on the four sets of test data

Test data 1 2 3 4
AAE(%) 0.85 1.13 0.89 0.96

AAE on the seven sets of tests

Test data SBP AAE(SD)(mmHg) DBP AAE(SD)(mmHg)
1 3.69(7.33) 2.36(5.36)
2 2.61(5.1) 1.69(2.74)
3 3.59(6.78) 2.24(5.32)
4 3.12(6.42) 1.85(4.36)
5 3.95(7.69) 2.66(5.88)
6 3.85(7.42) 2.48(5.19)
7 3.55(6.94) 2.58(5.68)
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other
Journal RSS Feed