温度缩放能有效提升脑肿瘤分类模型在噪声下的可信度。
Evaluating Temperature Scaling Calibration Effectiveness for CNNs under Varying Noise Levels in Brain Tumour Detection
- 用温度缩放校准模型置信度，不降低准确率
- 五种噪声下校准后预期校准误差显著下降
- 适合对可靠性要求高的医疗影像场景
深度学习中的精确置信度估计对医学影像等高风险领域至关重要，过度自信的误判可能带来严重后果。本文评估了后处理校准技术温度缩放（Temperature Scaling, TS）在脑肿瘤分类任务中的有效性。我们构建了一个自定义卷积神经网络（CNN），并在合并的脑部MRI数据集上训练。为模拟真实世界不确定性，引入五类图像噪声：高斯、泊松、椒盐、斑点和均匀噪声。通过精确率、召回率、F1分数、准确率、负对数似然（NLL）和预期校准误差（ECE）对比校准前后的性能。结果显示，在所有噪声条件下，温度缩放均显著降低了ECE和NLL，且未损害分类准确率。这表明温度缩放是一种高效且计算成本低的方法，可增强医疗AI系统在噪声或不确定环境下的决策可靠性。
原文摘要 · Abstract (English)
Precise confidence estimation in deep learning is vital for high-stakes fields like medical imaging, where overconfident misclassifications can have serious consequences. This work evaluates the effectiveness of Temperature Scaling (TS), a post-hoc calibration technique, in improving the reliability of convolutional neural networks (CNNs) for brain tumor classification. We develop a custom CNN and train it on a merged brain MRI dataset. To simulate real-world uncertainty, five types of image noise are introduced: Gaussian, Poisson, Salt & Pepper, Speckle, and Uniform. Model performance is evaluated using precision, recall, F1-score, accuracy, negative log-likelihood (NLL), and expected calibration error (ECE), both before and after calibration. Results demonstrate that TS significantly reduces ECE and NLL under all noise conditions without degrading classification accuracy. This underscores TS as an effective and computationally efficient approach to enhance decision confidence of medical AI systems, hence making model outputs more reliable in noisy or uncertain settings.
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。