Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study investigates the impact of various data augmentation strategies on the robustness of a binary malignant-versus-non-malignant classifier for dermoscopic images, particularly focusing on out-of-domain (OOD) generalization. By employing a ConvNeXt-Large backbone and evaluating performance through ROC-AUC metrics, the researchers found that a mix policy of photometric transformations significantly enhanced OOD performance, yielding a notable increase in ROC-AUC from 0.761-0.775 to 0.806-0.829 across multiple training seeds. However, while the approach showed promise in improving sensitivity on an independent clinical dataset, the results highlighted the necessity for a source-disjoint selection protocol to ensure unbiased evaluation of augmentation effects.
A mix policy of photometric transformations can boost OOD generalization in skin cancer classification by over 5% in ROC-AUC, challenging the notion that in-domain accuracy is paramount.
Background/Objectives: Dermoscopic skin lesion classifiers often lose accuracy under domain shift across imaging devices, illumination, and capture artifacts. We study how data augmentation improves the robustness of a binary malignant-versus-non-malignant classifier, with emphasis on out-of-domain (OOD) generalization. Methods: Single augmentations, photometric combinations, and composite policies were searched on a multi-source ISIC Archive collection with Derm7pt, using a ConvNeXt-Large backbone and ROC-AUC. Splits were made at the lesion-ID level, and HAM10000 and ISIC 2019-2020 were held out as a predominantly source-disjoint OOD test. Results: The largest OOD gain came from the mix policy, and photometric transformations dominated the most useful OOD operations. On an expanded pool from the same held-out sources the gain was +0.053 (95% CI +0.045 to +0.061, p<0.001), consistent across four training seeds (per-seed ROC-AUC: baseline 0.761-0.775, mix 0.806-0.829). On a small independent clinical collection, single-checkpoint sensitivity rose from 0.591 to 0.818, but this rested on 22 malignant cases and did not persist across seeds. Conclusions: Augmentations modelling real sources of domain shift can matter more than maximizing in-domain accuracy. Because the policy was selected on the same sources used to evaluate it, a source-disjoint selection protocol is needed before this effect size can be read as unbiased.