Dorosti Tina, Schultheiß Manuel, Schmette Philipp, Heuchert Jule, Thalhammer Johannes, Gassert Florian T, Sellerer Thorsten, Schick Rafael, Taphorn Kirsten, Mechlem Korbinian, Birnbacher Lorenz, Schaff Florian, Pfeiffer Franz, Pfeiffer Daniela
Chair of Biomedical Physics, Department of Physics, School of Natural Sciences, Technical University of Munich, Boltzmannstrasse 11, 85748 Garching, Germany.
Munich Institute of Biomedical Engineering, Technical University of Munich, Garching, Germany.
Radiol Artif Intell. 2025 Jul;7(4):e240484. doi: 10.1148/ryai.240484.
Purpose To estimate the total lung volume (TLV) from real and synthetic frontal chest radiographs on a pixel level using lung thickness maps generated by a U-Net deep learning model. Materials and Methods This retrospective study included 5959 chest CT scans from two public datasets, the Lung Nodule Analysis 2016 (Luna16) ( = 656) and the Radiological Society of North America Pulmonary Embolism Detection Challenge 2020 ( = 5303). Additionally, 72 participants were selected from the Klinikum Rechts der Isar dataset (October 2018 through December 2019), each with a corresponding chest radiograph obtained within 7 days. Synthetic radiographs and lung thickness maps were generated using forward projection of CT scans and their lung segmentations. A U-Net model was trained on synthetic radiographs to predict lung thickness maps and estimate TLV. Model performance was assessed using mean squared error (MSE), Pearson correlation coefficient, and two-sided Student distribution. Results The study included 72 participants (45 male and 27 female participants; 33 healthy participants: mean age, 62 years [range, 34-80 years]; 39 with chronic obstructive pulmonary disease: mean age, 69 years [range, 47-91 years]). TLV predictions showed low error rates (MSE, 0.16 L; MSE, 0.20 L; MSE, 0.35 L) and strong correlations with CT-derived reference standard TLV (, 1191; = 0.99; < .001) (, 72; = 0.97; < .001) (, 72; = 0.91; < .001). When evaluated on different datasets, the U-Net model achieved the highest performance for TLV estimation on the Luna16 test dataset, with the lowest MSE (0.09 L) and strongest correlation ( = 0.99; < .001) compared with CT-derived TLV. Conclusion The U-Net-generated pixel-level lung thickness maps successfully estimated TLV for both synthetic and real radiographs. Frontal Chest Radiographs, Lung Thickness Map, Pixel-Level, Total Lung Volume, U-Net © RSNA, 2025.
目的 使用由U-Net深度学习模型生成的肺厚度图，在像素水平上从真实和合成的胸部正位X线片估计总肺容量（TLV）。材料与方法 这项回顾性研究纳入了来自两个公共数据集的5959例胸部CT扫描，即2016年肺结节分析（Luna16）（n = 656）和北美放射学会2020年肺栓塞检测挑战数据集（n = 5303）。此外，从伊萨尔河右岸医院数据集（2018年10月至2019年12月）中选取了72名参与者，每人在7天内获得了相应的胸部X线片。使用CT扫描及其肺分割的前向投影生成合成X线片和肺厚度图。在合成X线片上训练U-Net模型以预测肺厚度图并估计TLV。使用均方误差（MSE）、Pearson相关系数和双侧Student t分布评估模型性能。结果 该研究纳入了72名参与者（45名男性和27名女性参与者；33名健康参与者：平均年龄62岁[范围34 - 80岁]；39名慢性阻塞性肺疾病患者：平均年龄69岁[范围47 - 91岁]）。TLV预测显示出较低的错误率（MSE，0.16 L；MSE，0.20 L；MSE，0.35 L），并且与CT衍生的参考标准TLV具有强相关性（n = 1,191；r = 0.99；P <.001）（n = 72；r = 0.97；P <.001）（n = 72；r = 0.91；P <.001）。在不同数据集上进行评估时，与CT衍生的TLV相比，U-Net模型在Luna16测试数据集上的TLV估计性能最高，MSE最低（0.09 L）且相关性最强（r = 0.99；P <.001）。结论 U-Net生成的像素级肺厚度图成功地估计了合成和真实X线片的TLV。胸部正位X线片、肺厚度图、像素级、总肺容量、U-Net © RSNA，2025年
