通过类级数据集混合实现乳腺MRI肿瘤分类中的跨数据集泛化
Cross-Dataset Generalization in Breast MRI Tumor Classification via Class-Wise Dataset Mixing
AI总结：
研究乳腺MRI肿瘤分类中深度学习模型因域转移和数据集来源偏差而跨机构失败的问题，通过构建类级混合训练集控制偏差，显著提升了WaveViT - Small和EfficientNet - B3在MAMA - MIA上的分类准确率和F1值。
AI中文摘要：
乳腺MRI对检测乳腺肿瘤高度敏感，但检查包含许多切片且需要大量阅片时间。深度学习模型在内部划分上通常表现良好，但由于域转移和数据集来源偏差，在不同机构间可能失败。我们研究二元乳腺MRI肿瘤分类中的这种失败模式。使用杜克乳腺癌MRI和fastMRI训练EfficientNet - B3和WaveViT - Small，并仅在独立的多中心MAMA - MIA队列上评估。在标签与数据集来源完美相关的故意混淆设置中，尽管召回率很高，但外部准确率接近随机水平。然后构建一个混合训练集，其中每个类都包含来自杜克和fastMRI的样本，同时保留患者级别的划分、增强和泄漏控制。在MAMA - MIA上，数据集混合将WaveViT - Small的准确率/F1提高到0.8463/0.8625，将EfficientNet - B3提高到0.8884/0.8994。这些结果表明控制数据集来源偏差对可靠的乳腺MRI分类很重要。
英文摘要：
Breast MRI is highly sensitive for detecting breast tumors, but exams contain many slices and require substantial reading time. Deep learning models often perform well on internal splits but can fail across institutions because of domain shift and dataset-origin bias. We study this failure mode for binary breast MRI tumor classification. EfficientNet-B3 and WaveViT-Small are trained using Duke Breast Cancer MRI and fastMRI, and evaluated only on the independent multi-center MAMA-MIA cohort. In a deliberately confounded setup, where label is perfectly correlated with dataset origin, external accuracy is near chance (0.5048--0.5265), despite very high recall. We then construct a mixed training set in which each class contains samples from both Duke and fastMRI, while preserving patient-level splitting, augmentation, and leakage controls. On MAMA-MIA, dataset mixing improves accuracy/F1 to 0.8463/0.8625 for WaveViT-Small and 0.8884/0.8994 for EfficientNet-B3. These results show that controlling dataset-origin bias is important for reliable breast MRI classification.
