Updated
Updated · arxiv.org · Nov 4
MvBody: Multi-View-Based Hybrid Transformer Using Optical 3D Body Scan for Explainable Cesarean Section Prediction
Updated
Updated · arxiv.org · Nov 4

MvBody: Multi-View-Based Hybrid Transformer Using Optical 3D Body Scan for Explainable Cesarean Section Prediction

1 articles · Updated · arxiv.org · Nov 4

Summary

  • Researchers have developed MvBody, a multi-view hybrid Transformer model that predicts cesarean section (CS) risk using 3D optical body scans and self-reported medical data.
  • The model achieved 84.62% accuracy and an AUC-ROC of 0.724, outperforming traditional machine learning and advanced 3D analysis methods on an independent test set.
  • MvBody could support prenatal care in resource-limited settings, offering explainable predictions and highlighting key risk factors such as body shape, age, and obstetric history.

Sources

Center100%
arxiv.org14h ago
MvBody: Multi-View-Based Hybrid Transformer Using Optical 3D Body Scan for Explainable Cesarean Section Prediction

Sources

1 total
Center100%
arxiv.org14h ago
MvBody: Multi-View-Based Hybrid Transformer Using Optical 3D Body Scan for Explainable Cesarean Section Prediction