When Does Data Augmentation Help with Few Chest-Radiograph Training Patients?
- 1Institute for Optimization and Data Analysis (IODA), School of Engineering and Computer Science, Bern University of Applied Sciences, 2502 Biel/Bienne, Switzerland
Abstract
Scarce medical images often prompt data augmentation. We ask when it earns its opportunity cost under a complete information budget. Across four collections, classical and GAN candidates are compared with real-only training, full-pool exposure, score adjustment and direct label use. Transformations reduce within-patient variation; source mismatch and between-patient uncertainty remain. Fixed 50-model ResNet50 ensembles with ranked classical candidates and probability averaging outperform equally adjusted real-only and full-pool ensembles at five sizes. We evaluate procedures in 50 paired draws at 50, 100, 150, 200 and 250 image-training patients. The design counts 200 regularization patients plus disjoint 100-patient selection and adjustment sets. Evaluation includes 38,743 CheXpert patients. At 250 patients, guidance reduces maximum group cross-entropy by 0.0442 versus real-only training (simultaneous 95% interval, 0.0048–0.0836). Adding the 100 selection patients to each original training cohort outperforms using their labels only to guide augmentation at 50–200 patients. Thus, use independent labels first when cohorts are smallest; consider selected augmentation once patient evidence, strong representation and aggregation are available. The 250-patient result marks the upper tested setting, not a universal threshold. AUROC remains stable while sensitivity and specificity shift, requiring operating-point review.
Related
- CT-Referenced Diagnostic Performance and Complementary Error Patterns of Artificial Intelligence Versus a Single Emergency Physician in Chest Radiograph Interpretation: A Retrospective Paired Diagnostic Accuracy Study
- Phantom Fairness: A Reproducible Audit of How Automatically Extracted Labels Can Conceal Demographic Disparities in Chest Radiograph Classifiers
- Calibration, Uncertainty, and Data Efficiency of Deep Learning Chest Radiograph Classifiers Under Cross-Dataset Distribution Shift: A Multi-Dataset Benchmark
- AI-Powered Preoperative Chest Radiograph-Derived Cardiovascular Border Indices and Risk of Postoperative Major Adverse Cardiovascular Events: A Large-Scale Retrospective Cohort Study.
- Calibrated per-pixel uncertainty for low-dose paediatric chest-radiograph denoising at no fidelity cost
BibTeX
@article{sariyar2026when,
title = {{When Does Data Augmentation Help with Few Chest-Radiograph Training Patients?}},
author = {Murat Sariyar},
journal = {Mathematics},
year = {2026},
month = oct,
doi = {10.3390/math14203732},
url = {https://www.mdpi.com/2227-7390/14/20/3732},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - When Does Data Augmentation Help with Few Chest-Radiograph Training Patients?
AU - Murat Sariyar
PY - 2026
DA - 2026/10/14
JO - Mathematics
DO - 10.3390/math14203732
UR - https://www.mdpi.com/2227-7390/14/20/3732
AB - Scarce medical images often prompt data augmentation. We ask when it earns its opportunity cost under a complete information budget. Across four collections, classical and GAN candidates are compared with real-only training, full-pool exposure, score adjustment and direct label use. Transformations reduce within-patient variation; source mismatch and between-patient uncertainty remain. Fixed 50-model ResNet50 ensembles with ranked classical candidates and probability averaging outperform equally adjusted real-only and full-pool ensembles at five sizes. We evaluate procedures in 50 paired draws at 50, 100, 150, 200 and 250 image-training patients. The design counts 200 regularization patients plus disjoint 100-patient selection and adjustment sets. Evaluation includes 38,743 CheXpert patients. At 250 patients, guidance reduces maximum group cross-entropy by 0.0442 versus real-only training (simultaneous 95% interval, 0.0048–0.0836). Adding the 100 selection patients to each original training cohort outperforms using their labels only to guide augmentation at 50–200 patients. Thus, use independent labels first when cohorts are smallest; consider selected augmentation once patient evidence, strong representation and aggregation are available. The 250-patient result marks the upper tested setting, not a universal threshold. AUROC remains stable while sensitivity and specificity shift, requiring operating-point review.
SN - 2227-7390
ER -