2. Assessment of Fetal and Maternal Well-
Being During Pregnancy Using Passive
Wearable Inertial Sensor
• 使用被動式可穿戴慣性傳感器評估孕期胎兒和母親的健康狀況
Eranda Somathilake , Upekha Hansanie Delay , Janith Bandara Senanayaka , Samitha Lakmal
Gunarathne ,Roshan Indika Godaliyadda , Senior Member, IEEE, Mervyn Parakrama Ekanayake
, Senior Member, IEEE,Janaka Wijayakulasooriya , Member, IEEE, and Chathura Rathnayake
3. Outline
• I. INTRODUCTION
• II. BACKGROUND
• III. DEVICE IMPLEMENTED
• IV. DATA ANALYSIS
A. 初步閱讀
B. 呼吸數據分析
C. 胎動檢測
D. 干擾去除
V. RESULTS
A. 呼吸數據分析
B. 胎動檢測
C. 干擾去除
VI. CONCLUSION
13. IV. DATA ANALYSIS
B. Respiratory Data Analysis(2/3)
比較不同類型的小波在加速度信號分類上的性能,使用了幾種類型的小波(Daubechies
小波 db2、db4、db6、Symlet 小波 sym2 和 sym4),並比較了每種小波的準確性。這種
比較可以在表 II 中觀察到。
(DWT)
14. IV. DATA ANALYSIS
B. Respiratory Data Analysis(3/3)
孕婦能量消耗maternal energy expenditure (EE)
15. IV. DATA ANALYSIS
C. Fetal Movement Detection
CNN 和 GRU 來自 13 位母親的數據
集。
數據來自十位媽媽被用來訓練,三位
媽媽被用於測試。
16. IV. DATA ANALYSIS
D. Interference Removal
Fig. 13.
Signal
estimation
using Wiener
filter.
19. V. RESULTS
B. Fetal Movement Detection(2/4)
Wavelet scalograms generated
from the data.
(a) Scalogram with fetal
movement.
(b) Scalogram without fetal
movement.
(c) Scalogram during mother’s
laughter.
23. VI. CONCLUSION
This device coupled with a three-step algorithm has
the ability to estimate the EE after activity with an
approximate accuracy of 98%.
Furthermore, implementing a GRU-based algorithm
on the data collected resulted in identifying the
occurrence of fetal movements with a training
accuracy of 90% and testing accuracy of 75%.。
From these features selected and computed were:
the mean, standard deviation,variance變異數, and the skewness偏度of each band.
得到的特徵被輸入到一個簡單的標準神經模式識別算法中。 輸入設置為上一步構建的特徵,輸出為三類活動。該網絡是一個前饋網絡,訓練是利用縮放的共軛梯度反向傳播完成的。 從數據集中,70% 的數據用於訓練,15% 用於驗證,其餘 15% 用於測試。