TY  - JOUR
AU  - Sadegh-Zadeh, Seyed-Ali
AU  - Mamalo, Alireza Soleimani
AU  - Saadat, Shayan
AU  - Gargari, Sahar Sayyadi
AU  - Barati, Mohammad Amin
AU  - Mehranfar, Sahar
AU  - Naderi, Zahra
PY  - 2025
DA  - 2025/07/06
TI  - Multi-Task Machine Learning for Prenatal Risk Stratification: Integrating Biomarkers, Maternal Age, and Ultrasound Measurements to Predict the Risk of Down Syndrome, Trisomy 18, Trisomy 13, and Neural Tube Defects
JO  - Frontiers in Biomedical Signal Processing
T2  - Frontiers in Biomedical Signal Processing
JF  - Frontiers in Biomedical Signal Processing
VL  - 1
IS  - 1
SP  - 24
EP  - 36
DO  - 10.62762/FBSP.2025.954863
UR  - https://www.icck.org/article/abs/FBSP.2025.954863
KW  - down syndrome
KW  - machine learning
KW  - prenatal screening
KW  - SHAP analysis
KW  - maternal biomarkers
AB  - This study developed a machine learning model for early risk stratification of Down syndrome by integrating maternal serum biomarkers and ultrasound measurements. A retrospective multicentre dataset was used, including maternal age, AFP, HCG, INHIBIN-A, and ultrasound parameters (NT, CRL). After imputing missing data and engineering features (e.g., Age_NT_interaction), a Gradient Boosting Machine (GBM) was trained and evaluated using AUROC, precision, recall, and F1-score. The model achieved high performance (AUROC: 0.9921; precision: 1.00; F1-score: 0.91; accuracy: 0.97). SHAP analysis identified key interactions—particularly Age_NT, Age_HCG, and Age_PAPP-A—as major contributors. High maternal age combined with elevated HCG or low PAPP-A was linked to increased risk, aligning with clinical knowledge. The model offers a highly accurate and interpretable approach for Down syndrome risk prediction, supporting personalized, data-driven prenatal care. Prospective validation and clinical integration are recommended.
SN  - 3071-2912
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{SadeghZadeh2025MultiTask,
  author = {Seyed-Ali Sadegh-Zadeh and Alireza Soleimani Mamalo and Shayan Saadat and Sahar Sayyadi Gargari and Mohammad Amin Barati and Sahar Mehranfar and Zahra Naderi},
  title = {Multi-Task Machine Learning for Prenatal Risk Stratification: Integrating Biomarkers, Maternal Age, and Ultrasound Measurements to Predict the Risk of Down Syndrome, Trisomy 18, Trisomy 13, and Neural Tube Defects},
  journal = {Frontiers in Biomedical Signal Processing},
  year = {2025},
  volume = {1},
  number = {1},
  pages = {24-36},
  doi = {10.62762/FBSP.2025.954863},
  url = {https://www.icck.org/article/abs/FBSP.2025.954863},
  abstract = {This study developed a machine learning model for early risk stratification of Down syndrome by integrating maternal serum biomarkers and ultrasound measurements. A retrospective multicentre dataset was used, including maternal age, AFP, HCG, INHIBIN-A, and ultrasound parameters (NT, CRL). After imputing missing data and engineering features (e.g., Age\_NT\_interaction), a Gradient Boosting Machine (GBM) was trained and evaluated using AUROC, precision, recall, and F1-score. The model achieved high performance (AUROC: 0.9921; precision: 1.00; F1-score: 0.91; accuracy: 0.97). SHAP analysis identified key interactions—particularly Age\_NT, Age\_HCG, and Age\_PAPP-A—as major contributors. High maternal age combined with elevated HCG or low PAPP-A was linked to increased risk, aligning with clinical knowledge. The model offers a highly accurate and interpretable approach for Down syndrome risk prediction, supporting personalized, data-driven prenatal care. Prospective validation and clinical integration are recommended.},
  keywords = {down syndrome, machine learning, prenatal screening, SHAP analysis, maternal biomarkers},
  issn = {3071-2912},
  publisher = {Institute of Central Computation and Knowledge}
}