Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong SAR.
Department of Nuclear Medicine, Peking University Third Hospital, Beijing, China.
Respir Res. 2024 Oct 28;25(1):389. doi: 10.1186/s12931-024-03004-9.
Pulmonary embolism (PE) is life-threatening and requires timely and accurate diagnosis, yet current imaging methods, like computed tomography pulmonary angiography, present limitations, particularly for patients with contraindications to iodinated contrast agents. We aimed to develop a quantitative texture analysis pipeline using machine learning (ML) based on non-contrast thoracic computed tomography (CT) scans to discover intensity and textural features correlated with regional lung perfusion (Q) physiology and pathology and synthesize voxel-wise Q surrogates to assist in PE diagnosis.
We retrospectively collected Tc-labeled macroaggregated albumin Q-SPECT/CT scans from patients suspected of PE, including an internal dataset of 76 patients (64 for training, 12 for testing) and an external testing dataset of 49 patients. Quantitative CT features were extracted from segmented lung subregions and underwent a two-stage feature selection pipeline. The prior-knowledge-driven preselection stage screened for robust and non-redundant perfusion-correlated features, while the data-driven selection stage further filtered features by fitting ML models for classification. The final classification model, trained with the highest-performing PE-associated feature combination, was evaluated in the testing cohorts based on the Area Under the Curve (AUC) for subregion-level predictability. The voxel-wise Q surrogate was then synthesized using the final selected feature maps (FMs) and model score maps (MSMs) to investigate spatial distributions. The Spearman correlation coefficient (SCC) and Dice similarity coefficient (DSC) were used to assess the spatial consistency between FMs or MSMs and Q-SPECT scans.
The optimal model performance achieved an AUC of 0.863 during internal testing and 0.828 on the external testing cohort. The model identified a combination containing 14 intensity and textural features that were non-redundant, robust, and capable of distinguishing between high- and low-functional lung regions. Spatial consistency assessment in the internal testing cohort showed moderate-to-high agreement between MSMs and reference Q-SPECT scans, with median SCC of 0.66, median DSCs of 0.86 and 0.64 for high- and low-functional regions, respectively.
This study validated the feasibility of using quantitative texture analysis and a data-driven ML pipeline to generate voxel-wise lung perfusion surrogates, providing a radiation-free, widely accessible alternative to functional lung imaging in managing pulmonary vascular diseases.
Not applicable.
肺栓塞（PE）是一种危及生命的疾病，需要及时准确的诊断，然而目前的成像方法，如计算机断层肺动脉造影（CTPA），存在局限性，特别是对于对碘造影剂有禁忌的患者。我们旨在开发一种基于机器学习（ML）的定量纹理分析管道，该管道使用非对比胸部 CT（CT）扫描来发现与区域肺灌注（Q）生理学和病理学相关的强度和纹理特征，并综合体素级 Q 替代物以辅助 PE 诊断。
我们回顾性地收集了疑似 PE 患者的 Tc 标记的大聚合白蛋白 Q-SPECT/CT 扫描，包括内部数据集的 76 名患者（64 名用于训练，12 名用于测试）和外部测试数据集的 49 名患者。从分割的肺亚区提取定量 CT 特征，并经历了一个两阶段的特征选择管道。基于知识驱动的预筛选阶段筛选出稳健且非冗余的与灌注相关的特征，而基于 ML 模型的驱动的选择阶段则通过拟合 ML 模型进行分类来进一步筛选特征。使用与最高性能的 PE 相关的特征组合训练最终分类模型，然后根据 AUC 评估其在测试队列中的亚区预测能力。然后使用最终选择的特征图（FM）和模型得分图（MSM）合成体素级 Q 替代物，以研究空间分布。使用 Spearman 相关系数（SCC）和 Dice 相似系数（DSC）评估 FM 或 MSM 与 Q-SPECT 扫描之间的空间一致性。
内部测试中最优模型的 AUC 达到 0.863，外部测试中达到 0.828。该模型确定了一组包含 14 个强度和纹理特征的组合，这些特征是不冗余的、稳健的，并且能够区分高功能和低功能肺区。内部测试队列中的空间一致性评估显示，MSM 与参考 Q-SPECT 扫描之间具有中等至高的一致性，高功能区和低功能区的中位数 SCC 分别为 0.66、0.86 和 0.64。
这项研究验证了使用定量纹理分析和基于数据的 ML 管道生成体素级肺灌注替代物的可行性，为管理肺血管疾病提供了一种无辐射、广泛可及的功能性肺成像替代方法。
不适用。
