Artificial intelligence in prenatal diagnosis: Down syndrome risk assessment with the power of gradient boosting-based machine learning algorithms/Prenatal tanida yapay zeka: Gradient boosting tabanli makine ogrenmesi algortimalarinin gucu ile Down sendromu risk degerlendirmesi

Objective: One of the most common chromosomal abnormalities seen during pregnancy is Down syndrome (Trisomy 21). To determine the risk of Down syndrome, first-trimester combined screening tests are essential. Using data from the first-trimester screening test, this study compares machine learning an...

Full description

Saved in:
Bibliographic Details
Published in:Turkish journal of obstetrics and gynecology Vol. 22; no. 2; p. 121
Main Authors: Yalcin, Emre, Koc, Tarik Kaan, Aslan, Serpil, Demir, Suleyman Cansun, Evruke, Ismail Cuneyt, Sucu, Mete, Avan, Mesut, Uzay, Fatma Islek
Format: Journal Article
Language:English
Published: Galenos Yayinevi Tic. Ltd 01.06.2025
Subjects:
Artificial intelligence
Diagnosis
Down syndrome
Educational evaluation
Machine learning
Medical research
Medicine, Experimental
Methods
Obstetrics
Physiological aspects
Pregnancy
Pregnant women
Prenatal diagnosis
Risk assessment
Risk factors
Technology application
Turkey
Northwest Territories
ISSN:2149-9322
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objective: One of the most common chromosomal abnormalities seen during pregnancy is Down syndrome (Trisomy 21). To determine the risk of Down syndrome, first-trimester combined screening tests are essential. Using data from the first-trimester screening test, this study compares machine learning and deep learning models to forecast the risk of Down syndrome. Materials and Methods: Within the scope of the study, biochemical and biophysical data of 959 pregnant women who underwent first-trimester screening tests at Cukurova University Obstetrics and Gynecology Clinic between 2020-2024 were analyzed. After cleaning missing and erroneous data, various preprocessing and normalization techniques were applied to the final dataset consisting of 853 observations. Down syndrome risk prediction was performed using different machine learning models, and model performances were compared based on accuracy rates and other evaluation metrics. Results: Experimental results show that the CatBoost model provides the highest success rate, with an accuracy rate of 95.31%. In addition, the XGBoost and LightGBM models exhibited high performance, with accuracy rates of 95.19% and 94.84%, respectively. The study also examines the effects of the class imbalance problem on model performance in detail and evaluates various strategies to reduce this imbalance. Conclusion: The findings show that gradient boosting-based machine learning models have significant potential in Down syndrome risk prediction. This approach is expected to contribute to the reduction of unnecessary invasive tests and improve clinical decision-making processes by increasing the accuracy rate in prenatal screening processes. Future studies should aim to increase the generalization capacity of the model on larger data sets and to provide integration with different machine learning algorithms. Keywords: Down syndrome, first-trimester screening test, gradient boosting, machine learning, artificial intelligence, classification algorithms Amac: Down sendromu (Trizomi 21), prenatal donemde en sik rastlanan kromozomal anomalilerden biridir. Gebeligin birinci trimesterinde uygulanan kombine tarama testleri, Down sendromu riskinin belirlenmesi icin onemli bir arac olarak kullanilmaktadir. Bu calisma, birinci trimester tarama testi verileri kullanilarak Down sendromu riskini tahmin etmek amaciyla farkli makine ogrenmesi ve derin ogrenme modellerini karsilastirmali olarak degerlendirmeyi amaclamaktadir. Gerec ve Yontemler: Calisma kapsaminda, 2020-2024 yillari arasinda Cukurova Universitesi Kadin Dogum Klinigi'nde birinci trimester tarama testine tabi tutulan 959 gebeye ait biyokimyasal ve biyofiziksel verileri analiz edilmistir. Eksik ve hatali veriler temizlendikten sonra, 853 gozlemden olusan nihai veri seti uzerinde cesitli on isleme ve normalizasyon teknikleri uygulanmistir. Farkli makine ogrenmesi modelleri kullanilarak Down sendromu risk tahmini gerceklestirilmis, model performanslari dogruluk oranlari ve diger degerlendirme metrikleri uzerinden karsilastirilmistir. Bulgular: Deneysel sonuclar, CatBoost modelinin %95,31 dogruluk orani ile en yuksek basariyi sagladigini gostermistir. Bunun yani sira, XGBoost ve LightGBM modelleri sirasiyla %95,19 ve %94,84 dogruluk oranlari ile yuksek performans sergilemistir. Calismada ayrica sinif dengesizligi probleminin model performansi uzerindeki etkileri detayli olarak incelenmis ve bu dengesizligi azaltmaya yonelik cesitli stratejiler degerlendirilmistir. Sonuc: Elde edilen bulgular, gradient boosting tabanli makine ogrenmesi modellerinin Down sendromu risk tahmininde onemli bir potansiyele sahip oldugunu gostermektedir. Bu yaklasimin, prenatal tarama sureclerindeki dogruluk oranini artirarak, gereksiz invaziv testlerin azaltilmasina ve klinik karar alma sureclerinin iyilestirilmesine katki saglamasi beklenmektedir. Gelecekteki calismalar, daha genis veri setleri uzerinde modelin genellestirme kapasitesini artirmayi ve farkli makine ogrenmesi algoritmalariyla entegrasyon saglamayi hedeflemelidir. Anahtar Kelimeler: Down sendromu, ilk trimester tarama testi, gradyan guclendirme, makine ogrenmesi, yapay zeka, siniflandirma algoritmalari
ISSN:2149-9322
DOI:10.4274/tjod.galenos.2025.83278