用于脑肿瘤检测的深度神经网络的可解释性
Explainability of Deep Neural Networks for Brain Tumor Detection
AI总结：
本研究应用可解释AI技术评估VGG-16、ResNet-50等CNN与ViT-Base-16在脑肿瘤检测中的性能，发现架构较浅的CNN因不易受小数据集欠拟合影响而表现更优，且能更有效地可视化肿瘤以支持医疗决策。
摘要：
医学图像分类对于支持医疗专业人员进行决策和培训至关重要。虽然卷积神经网络（CNN）传统上一直主导该领域，但基于 Transformer 的模型正受到关注。在本研究中，我们应用可解释 AI（XAI）技术来评估各种模型在真实世界医疗数据上的性能，并找出改进方向。我们将 VGG-16、ResNet-50 和 EfficientNetV2L 等 CNN 模型与 Transformer 模型 ViT-Base-16 进行了比较。结果表明，数据增强影响甚微，但超参数调优和高级建模能提升性能。CNN，尤其是 VGG-16 和 ResNet-50，优于 ViT-Base-16 和 EfficientNetV2L，这可能是由于数据有限导致的欠拟合。LIME 和 SHAP 等 XAI 方法进一步揭示，性能更好的模型能更有效地可视化肿瘤。这些发现表明，架构较浅的 CNN 更适合小数据集，并能支持医疗决策。
英文摘要：
Medical image classification is crucial for supporting healthcare professionals in decision-making and training. While Convolutional Neural Networks (CNNs) have traditionally dominated this field, Transformer-based models are gaining attention. In this study, we apply explainable AI (XAI) techniques to assess the performance of various models on real-world medical data and identify areas for improvement. We compare CNN models such as VGG-16, ResNet-50, and EfficientNetV2L with a Transformer model: ViT-Base-16. Our results show that data augmentation has little impact, but hyperparameter tuning and advanced modeling improve performance. CNNs, particularly VGG-16 and ResNet-50, outperform ViT-Base-16 and EfficientNetV2L, likely due to underfitting from limited data. XAI methods like LIME and SHAP further reveal that better-performing models visualize tumors more effectively. These findings suggest that CNNs with shallower architectures are more effective for small datasets and can support medical decision-making.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
