AI governance for medical imaging is formalizing: the 2026 ACR-SIIM Practice Parameter recommends local acceptance testing and ongoing drift monitoring, and the ACR Assess-AI registry monitors AI outputs using DICOM metadata for context. We argue that a necessary, currently unmonitored layer sits beneath output metrics: whether incoming studies remain within the acquisition envelope a model was validated on. Using a LUNA16-trained MONAI RetinaNet lung-nodule detector, we test whether acquisition state behaves as a structured, measurable variable. On real paired CT differing only in reconstruction kernel (NLST B30f vs B80f), kernel alone shifted AI-measured diameter and flipped a Fleischner size category in 5.2% (8 of 155) of nodules at fixed patient and acquisition, while detection confidence was unchanged (Wilcoxon p=0.22). Under controlled LIDC-IDRI perturbations the effects dissociated by axis: the noise axis degraded detection confidence (p=5.9e-32, concentrated in nodules under 6 mm) but not measurement, while the frequency/kernel axis corrupted measurement (p=8.6e-13) but not detection. A 4-feature pixel fingerprint recovered reconstruction identity (patient-level AUC about 0.95 on real CT, 0.995 on a QIBA phantom) where the ConvolutionKernel DICOM tag was uninformative (identical labels across reconstructions). The kernel axis transported across four manufacturers (leave-one-vendor-out AUC 0.94-0.98, matching the within-vendor ceiling). Acquisition state thus maps to distinct AI failure modes, frequency content to measurement reliability and noise to detection sensitivity, and is not recoverable from metadata. Acquisition-aware, input-side validation is the missing layer for the acceptance-testing and drift-monitoring requirements now entering imaging-AI accreditation.

翻译：医学影像人工智能治理正趋于规范化：2026年ACR-SIIM实践参数建议开展本地验收测试和持续漂移监测，ACR Assess-AI注册中心通过DICOM元数据监测AI输出结果。我们主张，在输出指标之下存在一个必要但尚未被监测的层面：传入研究是否仍处于模型验证时的采集包络范围内。使用基于LUNA16训练的MONAI RetinaNet肺结节检测器，我们验证采集状态是否表现为结构化的、可测量的变量。在仅重建核不同的真实配对CT（NLST B30f vs B80f）中，固定患者及采集条件下，仅核变化便使AI测量直径发生偏移，并在5.2%（155个结节中的8个）的结节中翻转Fleischner尺寸分类，而检测置信度无显著变化（Wilcoxon检验p=0.22）。在受控LIDC-IDRI扰动下，上述效应沿不同轴解离：噪声轴降低检测置信度（p=5.9e-32，集中于6毫米以下结节）而不影响测量，而频率/核轴破坏测量（p=8.6e-13）但不影响检测。基于4个特征的像素指纹可重建重建身份（真实CT上患者水平AUC约0.95，QIBA体模上为0.995），而DICOM标签ConvolutionKernel在此场景下无信息价值（不同重建的标签完全相同）。核轴效应可跨四家制造商传递（留一供应商验证AUC 0.94-0.98，与供应商内上限持平）。因此，采集状态对应不同的AI失效模式——频率内容影响测量可靠性，噪声影响检测灵敏度——且无法通过元数据恢复。面向采集的输入侧验证，正是当前影像AI认证体系中验收测试和漂移监测要求所缺失的关键环节。