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

Basit Öğe Kaydı
dc.contributor.authorKaplan, Ela
dc.contributor.authorEkinci, Tekin
dc.contributor.authorKaplan, Selcuk
dc.contributor.authorBarua, Prabal Datta
dc.contributor.authorDogan, Sengul
dc.contributor.authorTuncer, Turker
dc.contributor.authorAcharya, U. Rajendra
dc.date.accessioned2026-06-19T06:31:57Z
dc.date.available2026-06-19T06:31:57Z
dc.date.issued2022
dc.departmentMalatya Turgut Özal Üniversitesi
dc.description.abstractObjectives. 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. © 2022 Ela Kaplan et al.
dc.identifier.doi10.1155/2022/6034971
dc.identifier.issn1555-4309
dc.identifier.pmid35655731
dc.identifier.scopus2-s2.0-85131337658
dc.identifier.scopusqualityN/A
dc.identifier.urihttps://doi.org/10.1155/2022/6034971
dc.identifier.urihttps://hdl.handle.net/20.500.12899/4889
dc.identifier.volume2022
dc.indekslendigikaynakScopus
dc.indekslendigikaynakPubMed
dc.language.isoen
dc.publisherHindawi Limited
dc.relation.ispartofContrast Media and Molecular Imaging
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/openAccess
dc.snmzKA_Scopus_20260612
dc.subjectBayes Theorem
dc.subjectFemale
dc.subjectHumans
dc.subjectMale
dc.subjectPregnancy
dc.subjectSupport Vector Machine
dc.subjectUltrasonography
dc.subjectAdult
dc.subjectArea Under The Curve
dc.subjectArticle
dc.subjectBayesian Network
dc.subjectCesarean Section
dc.subjectClassifier
dc.subjectData Accuracy
dc.subjectData Base
dc.subjectDecision Tree
dc.subjectDecomposition
dc.subjectDeep Learning
dc.subjectDiagnostic Test Accuracy Study
dc.subjectEchography
dc.subjectFeature Extraction
dc.subjectFeature Selection
dc.subjectFemale
dc.subjectFetal Sex Classification
dc.subjectFetus Echography
dc.subjectGenital System
dc.subjectHistogram
dc.subjectHuman
dc.subjectImage Analysis
dc.subjectImage Processing
dc.subjectImage Quality
dc.subjectK Nearest Neighbor
dc.subjectLearning Algorithm
dc.subjectMajor Clinical Study
dc.subjectMale
dc.subjectMeasurement Accuracy
dc.subjectMeasurement Precision
dc.subjectNeighborhood
dc.subjectPrincipal Component Analysis
dc.subjectPyramidal Fixed Size Patch Based Feature Extraction
dc.subjectQuantitative Structure Activity Relation
dc.subjectQuantization
dc.subjectReceiver Operating Characteristic
dc.subjectRetrospective Study
dc.subjectSex
dc.subjectSupport Vector Machine
dc.subjectTextural Feature
dc.subjectTraining
dc.subjectUltrasound
dc.subjectVaginal Delivery
dc.subjectBayes Theorem
dc.subjectPregnancy
dc.subjectSupport Vector Machine
dc.titlePFP-LHCINCA: Pyramidal Fixed-Size Patch-Based Feature Extraction and Chi-Square Iterative Neighborhood Component Analysis for Automated Fetal Sex Classification on Ultrasound Images
dc.typeArticle

Dosyalar

Koleksiyon

Scopus İndeksli Yayınlar Koleksiyonu
PubMed İndeksli Yayınlar Koleksiyonu