Audit five hypothetical excerpts from medical imaging AI papers against a reproducibility checklist and mark each 'Rigorous' or a 'Red Flag'.
companion · Chapter 7
Preserve units, recording provenance, forward assumptions, and test evidence.
To increase our sample size from 50 to 5000, we extracted 100 patches per patient. We split the patches randomly into 80% train and 20% test sets.
We monitored validation loss and halted training when it stopped improving for 10 epochs. All subsequent metrics (AUC, Dice) were reported on the locked test set.
To handle class imbalance (90% normal), we applied random oversampling to the test set so it contained 50% abnormal cases. We report AUC on this balanced test set.
We applied random horizontal flips to all training images. The model is used to detect left vs. right kidney tumors.
Our preprocessing pipeline resampled all scans to 1x1x1mm, clipped intensities to [-100, 200] HU, and normalized to zero mean. We provide this code on GitHub.
