用质量感知门控融合影像，提升低剂量CT新病灶检测准确性
TopoGate: Quality-Aware Topology-Stabilized Gated Fusion for Longitudinal Low-Dose CT New-Lesion Prediction
- 通过三个质量信号动态调节影像与减影图的权重
- 在152对数据上实现0.65的ROC-AUC和0.14的Brier分数
- 可识别低质数据并提升性能，适合临床长期随访筛查
纵向低剂量CT随访在噪声、重建核及配准质量上存在差异，导致减影图像不稳定，易引发假阳性新病灶报警。本文提出TopoGate，一种轻量级模型，将随访影像外观视图与减影视图结合，并通过学习得到的质量感知门控机制调控二者影响。该门控由三个病例特异性信号驱动：CT外观质量、配准一致性以及基于拓扑度量的解剖拓扑稳定性。在包含152对数据（来自122名患者）的NLST--New-Lesion--LongCT队列上，TopoGate相比单视图基线显著提升判别力与校准性，实现0.65的受试者工作特征曲线下面积（ROC-AUC），标准差为0.05，贝叶斯评分（Brier score）为0.14。剔除由质量评分识别出的劣质数据对后，ROC-AUC从0.62提升至0.68，贝叶斯评分从0.14降至0.12。门控机制对质量退化响应可预测，在噪声增加时更依赖外观视图，符合放射科医生实践。该方法简单、可解释且适用于可靠的纵向低剂量CT筛查。
原文摘要 · Abstract (English)
Longitudinal low-dose CT follow-ups vary in noise, reconstruction kernels, and registration quality. These differences destabilize subtraction images and can trigger false new lesion alarms. We present TopoGate, a lightweight model that combines the follow-up appearance view with the subtraction view and controls their influence through a learned, quality-aware gate. The gate is driven by three case-specific signals: CT appearance quality, registration consistency, and stability of anatomical topology measured with topological metrics. On the NLST--New-Lesion--LongCT cohort comprising 152 pairs from 122 patients, TopoGate improves discrimination and calibration over single-view baselines, achieving an area under the ROC curve of 0.65 with a standard deviation of 0.05 and a Brier score of 0.14. Removing corrupted or low-quality pairs, identified by the quality scores, further increases the area under the ROC curve from 0.62 to 0.68 and reduces the Brier score from 0.14 to 0.12. The gate responds predictably to degradation, placing more weight on appearance when noise grows, which mirrors radiologist practice. The approach is simple, interpretable, and practical for reliable longitudinal LDCT triage.
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。