Synthetic Data Augmentation

How it works

Comparisons

Other

Highlighted claims

• Synthetic data generation occurred only inside training datasets after partitioning to avoid validation leakage. — Synthetic data-augmented machine learning for 30-day readmission prediction in patients with chronic conditions: a retrospective real-world study
• The study tested SMOTE, CTGAN, and TVAE to address outcome imbalance. — Synthetic data-augmented machine learning for 30-day readmission prediction in patients with chronic conditions: a retrospective real-world study
• SMOTE created additional minority-class readmission cases by interpolating between observed samples. — Synthetic data-augmented machine learning for 30-day readmission prediction in patients with chronic conditions: a retrospective real-world study
• TVAE used a variational autoencoder to learn relationships in EHR data and generate realistic synthetic patient records. — Synthetic data-augmented machine learning for 30-day readmission prediction in patients with chronic conditions: a retrospective real-world study
• CTGAN learned patterns from real tabular clinical data and generated synthetic records while preserving mixed variable types and distributions. — Synthetic data-augmented machine learning for 30-day readmission prediction in patients with chronic conditions: a retrospective real-world study
• TVAE was considered the strongest augmentation method because it better preserved clinical feature distributions and generated realistic synthetic patient profiles. — Synthetic data-augmented machine learning for 30-day readmission prediction in patients with chronic conditions: a retrospective real-world study