Wavelet特徴量を用いた時空間周波数解析によるエコー動画からの胎動検出

抄録

In this paper, we propose a method for detecting fetal movements in ultrasound video using wavelet-based spatio-temporal frequency analysis. Our approach extracts shape features that are robust to rotation and translation by integrating wavelet coefficients across scales. By analyzing temporal variations in these features, we can detect both the magnitude and timing of fetal movements. The proposed method was applied to ultrasound video data, demonstrating its effectiveness in capturing movement patterns. Compared to conventional sensor-based approaches, this method offers a non-invasive and noise-resistant alternative for fetal movement analysis.