PFP‐LHCINCA: Pyramidal Fixed‐Size Patch‐Based Feature Extraction and Chi‐Square Iterative Neighborhood Component Analysis for Automated Fetal Sex Classification on Ultrasound Images

Abstract

Objectives. Fetal sex determination with ultrasound (US) examination is indicated in pregnancies at risk of X-linked genetic disorders or ambiguous genitalia. However, misdiagnoses often arise due to operator inexperience and technical difficulties while acquiring diagnostic images. We aimed to develop an efficient automated US-based fetal sex classification model that can facilitate efficient screening and reduce misclassification. Methods. We have developed a novel feature engineering model termed PFP-LHCINCA that employs pyramidal fixed-size patch generation with average pooling-based image decomposition, handcrafted feature extraction based on local phase quantization (LPQ), and histogram of oriented gradients (HOG) to extract directional and textural features and used Chi-square iterative neighborhood component analysis feature selection (CINCA), which iteratively selects the most informative feature vector for each image that minimizes calculated feature parameter-derived k-nearest neighbor-based misclassification rates. The model was trained and tested on a sizeable expert-labeled dataset comprising 339 males’ and 332 females’ fetal US images. One transverse fetal US image per subject zoomed to the genital area and standardized to 256 × 256 size was used for analysis. Fetal sex was annotated by experts on US images and confirmed postnatally. Results. Standard model performance metrics were compared using five shallow classifiers—k-nearest neighbor (kNN), decision tree, naïve Bayes, linear discriminant, and support vector machine (SVM)—with the hyperparameters tuned using a Bayesian optimizer. The PFP-LHCINCA model achieved a sex classification accuracy of ≥88% with all five classifiers and the best accuracy rates (>98%) with kNN and SVM classifiers. Conclusions. US-based fetal sex classification is feasible and accurate using the presented PFP-LHCINCA model. The salutary results support its clinical use for fetal US image screening for sex classification. The model architecture can be modified into deep learning models for training larger datasets.

Evolving the Era of 5D Ultrasound? A Systematic Literature Review on the Applications for Artificial Intelligence Ultrasound Imaging in Obstetrics and Gynecology

“…Further, Rueda et al aimed at investigating the fetal nutritional status using AI-assisted assessment of the adipose and fat-free tissue of the fetal arm in US images [181]. Lastly, an AI model for the automated classification of fetal sex in 2D US images of the genital area was established by Kaplan et al that helped reduce misclassification and facilitate screening [182].…”

“…sification of fetal sex in 2D US images of the genital area was established by Kaplan et al that helped reduce misclassification and facilitate screening [182].…”

Transfer learning for accurate fetal organ classification from ultrasound images: a potential tool for maternal healthcare providers

“…This may not apply to other types of ultrasound scans. Other studies have explored the use of ultrasound in the classification of maternal–fetal standard planes in specific clinical scenarios, such as the detection of fetal gender 24 and the estimation of fetal gestational age 25 , 26 . In summary, various research has demonstrated that ultrasound may be used to classify maternal fetal standard plans, with good results for both 2D and 3D approaches, as well as the application of AI algorithms.…”

Transfer Learning for Accurate Fetal Organ Classification from Ultrasound Images: A Potential Tool for Maternal Healthcare Providers

“…Some studies focus on the identi cation of one fetal standard plan this may not apply to other types of ultrasound scans. Other studies have explored the use of ultrasound in the classi cation of maternal-fetal standard planes in speci c clinical scenarios, such as the detection of fetal gender [24] and the estimation of fetal gestational age [25] [26]. In summary, various research has demonstrated that ultrasound may be used to classify maternal fetal standard plans, with good results for both 2D and 3D approaches, as well as the application of AI algorithms.…”