在严格去重的测试下，发现白血病检测模型存在过拟合，需警惕数据泄露。
A Leakage-Aware Comparative Benchmark of Machine Learning, Deep Learning, and Transformer Models for Reliable Leukemia Detection
- 按患者分组划分数据，避免同一人图像同时出现在训练和测试集
- 高效网络B1表现最佳，准确率0.913，但多数模型过拟合恶性类
- 首次提供可复现、带校准评估的白血病检测基准，适合严谨研究者
从外周血涂片图像自动分类急性淋巴细胞白血病（ALL）的研究常在C-NMC 2019数据集上报告接近完美的性能。我们指出，这种高分可能源于患者级数据泄露——由于图像级随机划分，同一患者的细胞可能同时出现在训练与测试集中。为此，我们建立了一个严格的患者不相交协议下的泄漏感知基准，比较LightGBM、RBF-SVM、EfficientNet-B0、EfficientNet-B1和ViT-Tiny。模型基于73名患者的三个患者不相交折进行训练，评估使用包含1,867张图像、来自28名未见患者的外部预实验测试集，无患者重叠。除判别能力外，还采用期望校准误差（ECE）、Brier分数和温度缩放评估校准性。在真实评估下，EfficientNet-B1表现最佳，AUROC为0.913，敏感度0.87，特异度0.80，校准后ECE为0.024。冻结特征分类器和ViT-Tiny显示高敏感度但低特异度，表明倾向于过度预测恶性类别。随机划分与患者不相交划分的对比显示，即使在保守的冻结特征设置下，随机划分仍使AUROC虚增约0.04。这些结果警示：在C-NMC 2019上进行图像级评估不可靠，并为未来研究提供了可复现、校准感知的基准。
原文摘要 · Abstract (English)
Automated classification of acute lymphoblastic leukemia (ALL) from peripheral blood smear images has often reported near-perfect performance on the C-NMC 2019 dataset. We show that such results can be inflated by patient-level data leakage caused by random image-level partitioning, where cells from the same subject may appear in both training and test folds. We establish a leakage-aware benchmark under a strict subject-disjoint protocol, comparing LightGBM, RBF-SVM, EfficientNet-B0, EfficientNet-B1, and ViT-Tiny. Models are developed using three subject-disjoint folds from 73 subjects and evaluated on an external preliminary-phase test set of 1,867 images from 28 unseen subjects with zero patient overlap. Beyond discrimination, we assess calibration using expected calibration error, Brier score, and temperature scaling. Under honest evaluation, EfficientNet-B1 achieves the best performance, with AUROC 0.913, sensitivity 0.87, specificity 0.80, and calibrated ECE 0.024. Frozen-feature classifiers and ViT-Tiny show high sensitivity but poor specificity, indicating a tendency to over-predict the malignant class. A random-versus-subject-disjoint ablation shows that random splitting inflates AUROC by about 0.04 even in the conservative frozen-feature setting. These findings caution against image-level evaluation on C-NMC 2019 and provide a reproducible, calibration-aware benchmark for future work.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。