放射治疗医学图像去面部化技术评估:患者隐私与数据效用的影响
该研究评估了四种去面部化技术在放射治疗多模态图像中对再识别风险和数据效用的影响。结果显示,Quickshear隐私保护最强但数据改变大,Carina和mri_reface在隐私与效用间取得较好平衡,biometric_mask隐私保护最弱且对口腔等结构影响显著。
要点
- Quickshear隐私保护最佳(AUC 0.61-0.74),但显著改变多个结构。
- Carina保留结构体积,隐私保护次之(AUC 0.59-0.80)。
- biometric_mask隐私保护最弱(AUC 0.76-0.94),对下颌骨和口腔影响大。
核心结论
- Carina去识别器和mri_reface在隐私与数据效用间取得较好平衡,优于面部去除方法,最佳选择可能因应用优先级而异。
研究设计与样本
- 研究对88例脑部患者(MRI、CT、RTDose)和97例头颈部患者(PET、CT、RTDose)的图像应用了四种去面部化技术:biometric_mask、quickshear、mri_reface和Carina去识别器。
主要发现
- Quickshear隐私保护最高,AUC为0.61-0.74;Carina次之,AUC为0.59-0.80;mri_reface中等,AUC为0.70-0.91;biometric_mask最低,AUC为0.76-0.94。
- Carina保留结构体积,mri_reface对眼睛和晶状体产生轻微体积变化(5%和9%);biometric_mask显著影响下颌骨(39%)和口腔(69%);quickshear显著改变多个结构(10.0%-86.6%)。
- 去面部化后CT平均强度中位变化:口腔biometric_mask为-48.3%(IQR -65.9%至-33.9%),quickshear为-77.7%(IQR -89.4%至-51.7%);眼睛Carina为+24.5%(IQR 11.6%至38.2%),mri_reface为+54.9%(IQR 16.2%至82.3%)。
- 脑部数据集中,biometric_mask和quickshear使口腔Dmean分别降低0.72 Gy(IQR 0.33至0.89 Gy)和1.61 Gy(IQR 0.72至2.12 Gy);Carina和mri_reface使眼睛Dmax分别降低2.11 Gy(IQR 0.00至3.39 Gy)和1.05 Gy(IQR 0.21至1.16 Gy);头颈部数据集趋势相似但偏差更大。
临床怎么用
- 在需要平衡隐私保护与数据效用的放疗应用中,可优先考虑Carina去识别器或mri_reface;若隐私保护优先且能接受结构改变,可选用Quickshear。
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)。
原文
PURPOSE: To assess the impact of defacing-based deidentification techniques on reidentification risk and data utility across multimodal imaging in radiation therapy. METHODS AND MATERIALS: We applied 4 defacing techniques: biometric_mask, quickshear, mri_reface, and Carina's deidentifier, to imaging from 88 brain patients (magnetic resonance imaging, computed tomography [CT], and RTDose) and 97 head and neck patients (positron emission tomography, CT, and RTDose) in The Cancer Imaging Archive. Reidentification risk was assessed using ArcFace, a deep learning-based facial recognition model, by measuring cosine similarity scores and conducting receiver operating characteristic analysis to distinguish between original and defaced images. Data integrity was evaluated by statistically comparing the volume and image intensity changes between the original and defaced images across 9 critical organs and gross tumor volume. RESULTS: Quickshear provides the highest privacy protection, achieving the lowest area under the curve across imaging modalities (area under the curve, 0.61-0.74), followed by Carina (0.59-0.80). Mri_reface showed moderate protection (0.70-0.91), whereas biometric_mask offered the least (0.76-0.94). Carina preserved structure volumes, and mri_reface produced minor volumetric changes in the eyes (5%) and lens (9%). In contrast, biometric_mask substantially affected the mandible (39%) and oral cavity (69%), whereas quickshear significantly altered multiple structures (10.0%-86.6%). Median changes in mean CT intensity after defacing were -48.3% (interquartile range [IQR], -65.9% to -33.9%) with biometric_mask and -77.7% (IQR, -89.4% to -51.7%) with quickshear in the oral cavity. For eyes, Carina and mri_reface produced changes of +24.5% (IQR, 11.6% to 38.2%) and +54.9% (IQR, 16.2% to 82.3%), respectively. In the brain data set, biometric_mask and quickshear decreased oral cavity Dmean by 0.72 Gy (IQR, 0.33 Gy to 0.89 Gy) and 1.61 Gy (IQR, 0.72 Gy to 2.12 Gy), respectively. Carina and mri_reface reduced eyes Dmax by 2.11 Gy (IQR, 0.00 Gy to 3.39 Gy) and 1.05 Gy (IQR, 0.21 Gy to 1.16 Gy), respectively. A similar trend was observed in the head and neck data set with larger deviations. CONCLUSIONS: Carina's deidentifier and mri_reface showed favorable privacy-utility trade-offs relative to facial removal; the optimal choice may vary by application priorities.
引用
- Wang D, Lee SH, Wang T, et al. Assessment of Defacing Techniques on Medical Images for Radiation Therapy: Implications for Patient Privacy and Data Utility. Int J Radiat Oncol Biol Phys. 2026. PMID: 41876069. doi:10.1016/j.ijrobp.2026.03.023
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。