用生成模型合成癫痫病变MRI，减少标注需求且提升检测效果
Impact of Synthetic Lesional MR Images in Automated Focal Cortical Dysplasia Detection in Low-Data Scenarios
- 用条件生成网络根据病变掩码生成真实感脑影像
- 合成数据使检测灵敏度提升8.14%，模型置信度显著提高
- 适合数据稀缺的医学图像分析研究者参考
背景与目的：自动化检测局灶性皮质发育不良（FCD）需要大量逐体素标注的MRI数据，但此类数据难以获取。本研究旨在生成具有FCD特征的合成MRI，评估其真实性，并检验其对自动化FCD检测的影响，尤其在减少人工标注需求方面。方法：回顾性分析来自三个中心的131例FCD患者和90名健康对照的T1加权（T1w）及T2加权液体抑制反转恢复（FLAIR）MRI。通过在二值化FCD掩码上条件化生成网络生成合成MRI。两名神经放射科医生从14张真实与14张合成扫描中随机判断真伪。训练三个nnU-Net模型检测FCD：(i) 仅使用真实数据（35例FCD / 35例对照），(ii) 真实数据加合成数据增强（35例FCD / 35例对照），(iii) 扩展的真实数据（70例FCD / 70例对照）。结果：专家对真实与合成图像的区分能力有限，T1w分类准确率为60%，FLAIR为70%（组间一致性kappa=0.86）。加入合成数据后，检测灵敏度提升8.14%（p=0.12），病灶区域模型置信度由0.83±0.11升至0.89±0.12（p=0.02）。扩展真实数据模型进一步将灵敏度提升至73.8%（p<0.001），置信度达0.90±0.14（p=0.01）。结论：条件生成网络可生成逼真的合成FCD-MRI，在降低约20%标注数据需求的同时保持等效检测敏感度；当真实数据充足时，仍优于合成数据增强。
原文摘要 · Abstract (English)
Background and Purpose: Automated detection of focal cortical dysplasia (FCD) requires large volumes of voxelwise lesion-delineated MRI data, which are difficult to acquire. This study aims to generate synthetic MRI data exhibiting FCD, assess their realism, and evaluate their impact on automated FCD detection, particularly in reducing the need for manual annotations. Methods: T1-weighted (T1w) and T2-weighted Fluid-Attenuated Inversion Recovery (FLAIR) MRI scans from 131 FCD patients and 90 healthy controls from multiple (3) sites were retrospectively studied. Synthetic MRIs were generated by conditioning a generative network on binary FCD masks. Two neuroradiologists identified real images from a random set of 14 real and 14 synthetic scans. Three nnU-Net models were trained to detect FCD using: (i) real-only (35 FCD / 35 controls), (ii) real (35 FCD / 35 controls) plus synthetic augmentation, and (iii) expanded real data (70 FCD / 70 controls). Results: Experts showed limited ability to distinguish real from synthetic images, with classification accuracy of 60% for T1w and 70% for FLAIR (inter-rater agreement kappa = 0.86). Augmenting automated FCD detection with synthetic data increased sensitivity by 8.14% (p = 0.12) and improved model confidence at true lesion sites (0.83 +/- 0.11 to 0.89 +/- 0.12; p = 0.02). The expanded real-data model further improved sensitivity to 73.8% (p < 0.001) and confidence to 0.90 +/- 0.14 (p = 0.01). Conclusion: Conditional generative networks can generate realistic synthetic FCD-MRIs, reducing labeled data needs by approximately 20% while maintaining equivalent sensitivity. Equivalent amounts of real data, when available, remain more effective than synthetic augmentation.
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。