Liu Yi-Kuan, Cisneros Jorge, Nair Girish, Stevens Craig, Castillo Richard, Vinogradskiy Yevgeniy, Castillo Edward
Department of Biomedical Engineering, The University of Texas at Austin, Austin, TX, USA.
Division of Pulmonary and Critical Care, Beaumont Health, Royal Oak, MI, USA.
Int J Comput Assist Radiol Surg. 2025 May;20(5):959-970. doi: 10.1007/s11548-025-03323-2. Epub 2025 Jan 20.
Pulmonary perfusion imaging is a key lung health indicator with clinical utility as a diagnostic and treatment planning tool. However, current nuclear medicine modalities face challenges like low spatial resolution and long acquisition times which limit clinical utility to non-emergency settings and often placing extra financial burden on the patient. This study introduces a novel deep learning approach to predict perfusion imaging from non-contrast inhale and exhale computed tomography scans (IE-CT).
We developed a U-Net Transformer architecture modified for Siamese IE-CT inputs, integrating insights from physical models and utilizing a self-supervised learning strategy tailored for lung function prediction. We aggregated 523 IE-CT images from nine different 4DCT imaging datasets for self-supervised training, aiming to learn a low-dimensional IE-CT feature space by reconstructing image volumes from random data augmentations. Supervised training for perfusion prediction used this feature space and transfer learning on a cohort of 44 patients who had both IE-CT and single-photon emission CT (SPECT/CT) perfusion scans.
Testing with random bootstrapping, we estimated the mean and standard deviation of the spatial Spearman correlation between our predictions and the ground truth (SPECT perfusion) to be 0.742 ± 0.037, with a mean median correlation of 0.792 ± 0.036. These results represent a new state-of-the-art accuracy for predicting perfusion imaging from non-contrast CT.
Our approach combines low-dimensional feature representations of both inhale and exhale images into a deep learning model, aligning with previous physical modeling methods for characterizing perfusion from IE-CT. This likely contributes to the high spatial correlation with ground truth. With further development, our method could provide faster and more accurate lung function imaging, potentially expanding its clinical applications beyond what is currently possible with nuclear medicine.
肺灌注成像作为一种诊断和治疗规划工具，是一项关键的肺部健康指标。然而，当前的核医学模式面临着诸如空间分辨率低和采集时间长等挑战，这限制了其在非紧急情况下的临床应用，并且常常给患者带来额外的经济负担。本研究引入了一种新颖的深度学习方法，用于从非增强吸气和呼气计算机断层扫描（IE-CT）预测灌注成像。
我们开发了一种针对连体IE-CT输入进行修改的U-Net Transformer架构，整合了物理模型的见解，并采用了为肺功能预测量身定制的自监督学习策略。我们从九个不同的4DCT成像数据集中汇总了523张IE-CT图像用于自监督训练，旨在通过从随机数据增强中重建图像体积来学习低维IE-CT特征空间。用于灌注预测的监督训练使用了这个特征空间，并在一组44名同时进行了IE-CT和单光子发射CT（SPECT/CT）灌注扫描的患者中进行了迁移学习。
通过随机重抽样测试，我们估计我们的预测与真实值（SPECT灌注）之间的空间斯皮尔曼相关性的均值和标准差为0.742±0.037，平均中位数相关性为0.792±0.036。这些结果代表了从非增强CT预测灌注成像的新的最先进精度。
我们的方法将吸气和呼气图像的低维特征表示结合到一个深度学习模型中，与先前用于从IE-CT表征灌注的物理建模方法相一致。这可能有助于与真实值的高空间相关性。随着进一步发展，我们的方法可以提供更快、更准确的肺功能成像，有可能将其临床应用扩展到核医学目前无法实现的范围之外。
