Biometric measurement and classification of IUGR using neural networks

Abstract

The examination of fetal growth is an important cause of perinatal morbidity and mortality. The accurate evaluation of fetal growth during pregnancy is difficult, but recent techniques have improved this important aspect of obstetrics and Gynecology with positive implications for prenatal patients and their babies. Ultrasound measurements play a significant role in obstetrics and Gynecology as an accurate means for the estimation of fetal growth. In this work an automated method is proposed for the Biometric measurement and classification of IUGR, using OpenGL concepts for extracting the feature values and Artificial Neural Network (ANN) model is designed for diagnosis and classification. The features to figure out whether a fetus is normal or abnormal were extracted from the 2D-ultrasound images using OpenGL concepts. The features that are considered for the determination of the IUGR are gestational age (GA), bi-parietal diameter (BPD), abdominal circumference (AC), head circumference (HC), and femur length (FL). These feature values were obtained from 2D-ultrasound image. ANN model designed for the classification is able to distinguish whether fetus is normal or abnormal based on the feature values. Two ANN models, Multilayer Perceptron (MLP) using Back propagation algorithm and Radial Basis Function (RBF) models were studied and used for the diagnosis and classification.

Machine learning improves early prediction of small‐for‐gestational‐age births and reveals nuchal fold thickness as unexpected predictor

“…In terms of predicting SGA, there have been studies that used ML models. Features used for prediction in these past studies include ultrasound biometrics measurements, 15,16 umbilical Doppler blood flow, 17,18 pregnancy risk factors, sociodemographic, maternal characteristic and medical history, 19,20 pregnancy associated plasma protein A and placental growth factor. 21 The objective of this study is to evaluate the performance of ML in predicting SGA at birth, using second trimester fetal ultrasound scans.…”

Toward deep observation: A systematic survey on artificial intelligence techniques to monitor fetus via ultrasound images

“… FCNN N/A 11 - 19 weeks Segmentation ( Li et al., 2017 ) To segment the amniotic fluid and fetal tissues in fetal US images The encoder-decoder network based on VGG16 N/A 22 ND weeks Segmentation ( Ryou et al., 2016 ) To localize the fetus and extract the best fetal biometry planes for the head and abdomen from first trimester 3D fetal US images CNN Structured Random Forests 11 - 13 weeks Classification ( Toussaint et al., 2018 ) To detect and localize fetal anatomical regions in 2D US images ResNet18 Soft Proposal Layer (SP) 22 - 32 weeks Classification ( Ravishankar et al., 2016 ) To reliably estimate abdominal circumference CNN + Gradient Boosting Machine (GBM) Histogram of Oriented Gradient (HoG) 15 - 40 weeks Classification ( Wee et al., 2010 ) To detect and recognize the fetal NT based on 2D ultrasound images by using artificial neural network techniques. Artificial Neural Network (ANN) Multilayer Perceptron (MLP) Network Bidirectional Iterations Forward Propagations Method (BIFP) N/A Classification ( Liu et al., 2019 ) To detect NT region U-Net NT Segmentation PCA NT Thickness Measurement VGG16 NT Region Detection 4 - 12 weeks Segmentation Growth disease ( Bagi and Shreedhara, 2014 ) To propose the biometric measurement and classification of IUGR, using OpenGL concepts for extracting the feature values and ANN model is designed for diagnosis and classification ANN Radial Basis Function (RBF) OpenGL 12–40 Weeks Classification ( Selvathi and Chandralekha, 2021 …”

Imputing Missing Values for Improved Statistical Inference Applied to Intrauterine Growth Restriction Problem

“…The comparisons of absolute measurements of the fetuses with reference values, as well as birth weight percentiles, allow detection of deviations between expected and actual fetal growth and identification of newborns being possibly at risk for adverse health events [28]. However, the diagnosis of IUGR is based on non-consistent definitions [29].…”