合成OCTA图像虽像真图，但关键血管细节失真，影响临床诊断。
In Defense of OCTA: The Reconstruction-Utility Gap in OCT-to-OCTA Synthesis
- 用真实OCTA分割器检测合成图像的下游任务表现
- 细小毛细血管结构丢失严重（Dice下降0.163，大血管仅0.031）
- 适合关注临床实用性的医学影像研究者阅读
光学相干断层扫描血管成像（OCTA）可显示视网膜血流，提供评估糖尿病视网膜病变缺血程度的微血管灌注与黄斑无血管区等生物标志物。由于OCTA设备不如结构型OCT普及，近期研究尝试从OCT生成OCTA，报告重建质量高（3D PSNR > 31 dB，SSIM > 0.9）。我们不关注图像是否相似，而关注其是否支持原始OCTA的测量目标。使用冻结的真实OCTA分割器作为探针，对两种合成方法（XOCT、TransPro）进行测试，发现随着结构精细度提升，下游分割性能下降：大血管保持较好（Dice从0.862降至0.831），而细小毛细血管网络显著崩溃（Dice从0.798降至0.635，损失量为大血管的五倍；配对Wilcoxon检验p < 1e-3），TransPro整体更差。匹配模糊对照实验表明该细节非模糊所致，而是虚构。在私有Spectralis数据集上重新训练后，两种方法均未能再现新生血管病灶（定性分析，n=3）。重建保真度≠临床实用性，我们提出应以下游任务保真度作为评估OCT到OCTA合成的核心标准。
原文摘要 · Abstract (English)
Optical coherence tomography angiography (OCTA) images retinal blood flow, giving capillary-perfusion and foveal-avascular-zone biomarkers that grade diabetic-retinopathy ischemia. Because OCTA hardware is less common than structural OCT, recent work synthesizes it from OCT, reporting strong reconstruction (3D PSNR > 31 dB, SSIM > 0.9). We ask not whether the synthetic image looks similar, but whether it supports the measurements OCTA is acquired for. A frozen real-OCTA segmenter, applied as a probe to two synthesizers (XOCT, TransPro), shows downstream Dice falling with structural fineness: large vessels survive (0.862 -> 0.831) while the fine capillary network collapses (0.798 -> 0.635, five times the large-vessel loss; paired Wilcoxon p < 1e-3), TransPro worse throughout. A matched-blur control shows this detail is fabricated, not blurred. Retrained on a private Spectralis dataset, neither synthesizer reproduces the neovascular lesion (qualitative, n=3). Reconstruction fidelity is not clinical utility; we establish downstream-task fidelity as the evaluation OCT-to-OCTA synthesis needs.
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。