发表机构
Technological University Dublin; Research Ireland Centre for Research Training in Machine Learning; University College Dublin; Dublin City University（都柏林理工大学; 爱尔兰研究机器学习研究培训中心; 都柏林大学学院; 都柏林城市大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究探讨数据增强对黑色素瘤分类模型置信度校准的影响，在SIIM-ISIC 2020数据集上对比有无增强的训练，发现增强方法能改善不确定性校准。
AI 中文摘要
准确量化预测不确定性或改进模型校准在医学图像分类中起着重要作用，尤其是在黑色素瘤诊断中，准确的不确定性量化可能对患者护理产生重大影响。改进校准的方法之一是数据增强。此外，数据增强作为一种合成增加数据集大小的方法，已被证明能够提高在不平衡数据集上训练的模型的性能。然而，数据增强作为对原始数据一部分的变换，对在不平衡数据集（特别是黑色素瘤分类）上训练的模型校准的影响尚未得到充分探索。我们在SIIM-ISIC 2020黑色素瘤分类数据集上，在两种条件下训练神经网络：有数据增强和无数据增强，并比较两种场景下AUC和期望校准误差（ECE）的差异。我们的结果表明，使用不同的增强方法可以改善不确定性校准。
英文摘要
Accurately quantifying the predictive uncertainty or improving model calibration plays an important role in medical image classification, in particular in melanoma diagnosis, where accurate uncertainty quantification can have significant implications for patient care. One of the methods for calibration improvement is data augmentation. In addition, data augmentation as a method for synthetically increasing the size of the dataset has been proven to improve the performance of models trained on imbalanced datasets. However, the impact of data augmentation, as a transformation of a part of the original data, on calibration of models trained on imbalanced datasets, in particular in melanoma classification is under-explored. We train neural networks on SIIM-ISIC 2020 melanoma classification dataset under two conditions: with and without data augmentation, and compare the differences in AUC and expected calibration error (ECE) in both scenarios. Our results shows improvements in uncertainty calibration using different augmentation methods.
CommentsPresented at the 30th Conference on Medical Image Understanding and Analysis (MIUA 2026)
