Efficient fetal size classification combined with artificial neural network for estimation of fetal weight

• Main Authors: Yueh-Chin Cheng (Author), Gwo-Lang Yan (Author), Yu Hsien Chiu (Author), Fong-Ming Chang (Author), Chiung-Hsin Chang (Author), Kao-Chi Chung (Author)
• Format: Book
• Published: Elsevier, 2012-12-01T00:00:00Z.
• Subjects: article
• Online Access: Connect to this object online.
• Notes: 1028-4559; 10.1016/j.tjog.2012.09.009
• Keywords: artificial neural network; estimated fetal weight; ultrasonographic parameter; Gynecology and obstetrics; RG1-991
• Source: Taiwanese Journal of Obstetrics & Gynecology, Vol 51, Iss 4, Pp 545-553 (2012)

Objectives: A novel analysis was undertaken to select a significant ultrasonographic parameter (USP) for classifying fetuses to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Methods: In total, 2127 singletons were examined by prenatal ultrasound within 3 days before delivery. First, correlation analysis was used to determine a significant USP for fetal grouping. Second, K-means algorithm was utilized for fetal size classification based on the selected USP. Finally, stepwise regression analysis was used to examine input parameters of the ANN model. Results: The estimated fetal weight (EFW) of the new model showed mean absolute percent error (MAPE) of 5.26 ± 4.14% and mean absolute error (MAE) of 157.91 ± 119.90 g. Comparison of EFW accuracy showed that the new model significantly outperformed the commonly-used EFW formulas (all p < 0.05). Conclusion: We proved the importance of choosing a specific grouping parameter for ANN to improve EFW accuracy.