提高深度学习模型在入院头颅CT预测血肿扩大中的稳健性
本研究使用多中心890例患者队列进行交叉验证与训练,并在2个卒中中心684例连续脑出血患者中独立验证,评估对抗训练和输入修改对深度学习模型预测血肿扩大稳健性的影响。仅用干净扫描训练的模型在PGD和FGSM对抗图像测试下平均AUC从0.8分别降至0.67和0.71,而采用干净扫描与PGD对抗扫描5:3混合训练并加入Otsu多阈值分割输入可将平均AUC提升至0.77。结论认为对抗训练和纳入基于阈值的分割作为附加输入可提高模型稳健性。
要点
- 仅用干净扫描训练的模型在对抗攻击下平均AUC从0.8降至0.67(PGD)和0.71(FGSM)。
- 干净与PGD对抗扫描5:3混合训练并加入Otsu多阈值分割输入,平均AUC提升至0.77。
- FGSM对抗训练仅改善对同类攻击的稳健性,对PGD型图像的跨攻击稳健性有限。
核心结论
- 对抗训练和纳入基于阈值的分割作为附加输入可提高深度学习模型在急性脑出血入院头颅CT中预测血肿扩大的稳健性。
研究设计与样本
- 研究使用多中心队列n=890例患者进行交叉验证与训练,并使用来自2个卒中中心的n=684例连续脑出血患者队列进行独立验证。
- 采用快速梯度符号法(FGSM)和投影梯度下降(PGD)对抗攻击进行训练和测试。
- 开发并测试4个模型,分别在独立验证队列中预测≥3 mL、≥6 mL、≥9 mL和≥12 mL的血肿扩大,采用受试者工作特征曲线下面积(AUC)评估。
- 检查了对抗扫描与未扰动(干净)扫描的不同混合比例用于训练,以及加入无超参数的Otsu多阈值分割作为模型附加输入。
诊断标准与关键数值
- 仅用干净扫描训练的深度学习模型在PGD对抗图像测试时平均血肿扩大预测AUC从0.8降至0.67。
- 仅用干净扫描训练的深度学习模型在FGSM对抗图像测试时平均血肿扩大预测AUC从0.8降至0.71。
- 采用干净扫描与PGD对抗扫描5:3混合训练并加入Otsu多阈值分割至模型输入,针对PGD和FGSM对抗攻击的平均AUC均提升至0.77。
局限性
- FGSM对抗训练改善了对同类攻击的稳健性,但对PGD型图像的跨攻击稳健性有限。
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
BACKGROUND AND PURPOSE: Robustness against input data perturbations is essential for deploying deep learning models in clinical practice. Adversarial attacks involve subtle, voxel-level manipulations of scans to increase deep learning models' prediction errors. Testing deep learning model performance on examples of adversarial images provides a measure of robustness, and including adversarial images in the training set can improve the model's robustness. In this study, we examined adversarial training and input modifications to improve the robustness of deep learning models in predicting hematoma expansion (HE) from admission head CTs of patients with acute intracerebral hemorrhage (ICH).
MATERIALS AND METHODS: We used a multicenter cohort of n = 890 patients for cross-validation/training, and a cohort of n = 684 consecutive patients with ICH from 2 stroke centers for independent validation. Fast gradient sign method (FGSM) and projected gradient descent (PGD) adversarial attacks were applied for training and testing. We developed and tested 4 different models to predict ≥3 mL, ≥6 mL, ≥9 mL, and ≥12 mL HE in an independent validation cohort applying receiver operating characteristics area under the curve (AUC). We examined varying mixtures of adversarial and nonperturbed (clean) scans for training as well as including additional input from the hyperparameter-free Otsu multithreshold segmentation for model.
RESULTS: When deep learning models trained solely on clean scans were tested with PGD and FGSM adversarial images, the average HE prediction AUC decreased from 0.8 to 0.67 and 0.71, respectively. Overall, the best performing strategy to improve model robustness was training with 5:3 mix of clean and PGD adversarial scans and addition of Otsu multithreshold segmentation to model input, increasing the average AUC to 0.77 against both PGD and FGSM adversarial attacks. Adversarial training with FGSM improved robustness against similar type attack but offered limited cross-attack robustness against PGD-type images.
CONCLUSIONS: Adversarial training and inclusion of threshold-based segmentation as an additional input can improve deep learning model robustness in prediction of HE from admission head CTs in acute ICH.
引用
- Tran AT, Abou Karam G, Zeevi D, et al. Improving the Robustness of Deep Learning Models in Predicting Hematoma Expansion from Admission Head CT. AJNR Am J Neuroradiol. 2025. PMID: 39794133. doi:10.3174/ajnr.A8650
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。