Haase Robert, Pinetz Thomas, Kobler Erich, Bendella Zeynep, Paech Daniel, Clauberg Ralf, Foltyn-Dumitru Martha, Wagner Verena, Schlamp Kai, Heussel Gudula, Heussel Claus Peter, Vahlensieck Martin, Luetkens Julian A, Schlemmer Heinz-Peter, Specht-Riemenschneider Louisa, Radbruch Alexander, Effland Alexander, Deike Katerina
From the Department of Diagnostic and Interventional Neuroradiology, University Hospital Bonn, Venusberg-Campus 1, Bonn, Germany (R.H., E.K., Z.B., D.P., R.C., A.R., K.D.); Institute of Applied Mathematics, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany (T.P., A.E.); Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA (D.P.); Division of Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany (D.P., H.-P.S.); Department of Diagnostic and Interventional Radiology With Nuclear Medicine, Thoraxklinik at University Hospital Heidelberg, Heidelberg, Germany (M.F.-D., K.S., G.H., C.P.H.); Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany (M.F.-D.); Department of Diagnostic and Interventional Radiology, University Hospital Heidelberg, Heidelberg, Germany (V.W., C.P.H.); Translational Lung Research Center Heidelberg, Member of the German Center of Lung Research (DZL), Heidelberg, Germany (C.P.H.); Praxisnetz, Radiology and Nuclear Medicine, Bonn, Germany (M.V.); Department of Diagnostic and Interventional Radiology, University Hospital Bonn, Venusberg-Campus 1, Bonn, Germany (J.A.L.); Chair of Civil Law, Data Protection Law, Law of Data Economy, Digitalization and AI Law, Faculty of Law, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany (L.S.-R.); and German Center for Neurodegenerative Diseases (DZNE), Helmholtz Association of German Research Centers, Venusberg-Campus 1, Bonn, Germany (A.R., K.D.).
Invest Radiol. 2025 May 1;60(5):340-348. doi: 10.1097/RLI.0000000000001137. Epub 2024 Nov 19.
Small lesions are the limiting factor for reducing gadolinium-based contrast agents in brain magnetic resonance imaging (MRI). The purpose of this study was to compare the sensitivity and precision in metastasis detection on true contrast-enhanced T1-weighted (T1w) images and artificial images synthesized by a deep learning method using low-dose images.
In this prospective, multicenter study (5 centers, 12 scanners), 917 participants underwent brain MRI between October 2021 and March 2023 including T1w low-dose (0.033 mmol/kg) and full-dose (0.1 mmol/kg) images. Forty participants with metastases or unremarkable brain findings were evaluated in a reading (mean age ± SD, 54.3 ± 15.1 years; 24 men). True and artificial T1w images were assessed for metastases in random order with 4 weeks between readings by 2 neuroradiologists. A reference reader reviewed all data to confirm metastases. Performances were compared using mid- P McNemar tests for sensitivity and Wilcoxon signed rank tests for false-positive findings.
The reference reader identified 97 metastases. The sensitivity of reader 1 did not differ significantly between sequences (sensitivity [precision]: true, 66.0% [98.5%]; artificial, 61.9% [98.4%]; P = 0.38). With a lower precision than reader 1, reader 2 found significantly more metastases using true images (sensitivity [precision]: true, 78.4% [87.4%]; artificial, 60.8% [80.8%]; P < 0.001). There was no significant difference in sensitivity for metastases ≥5 mm. The number of false-positive findings did not differ significantly between sequences.
One reader showed a significantly higher overall sensitivity using true images. The similar detection performance for metastases ≥5 mm is promising for applying low-dose imaging in less challenging diagnostic tasks than metastasis detection.
小病灶是脑磁共振成像（MRI）中减少钆基造影剂使用的限制因素。本研究的目的是比较在真实对比增强T1加权（T1w）图像和通过深度学习方法利用低剂量图像合成的人工图像上转移瘤检测的敏感性和准确性。
在这项前瞻性、多中心研究（5个中心，12台扫描仪）中，917名参与者在2021年10月至2023年3月期间接受了脑部MRI检查，包括T1w低剂量（0.033 mmol/kg）和全剂量（0.1 mmol/kg）图像。40名有转移瘤或脑部检查无异常的参与者接受了阅片评估（平均年龄±标准差，54.3±15.1岁；24名男性）。由2名神经放射科医生对真实和人工T1w图像进行随机顺序的转移瘤评估，两次阅片间隔4周。一名参考阅片者复查所有数据以确认转移瘤。使用中P McNemar检验比较敏感性，使用Wilcoxon符号秩检验比较假阳性结果。
参考阅片者共识别出97个转移瘤。阅片者1在不同序列间的敏感性无显著差异（敏感性[准确性]：真实图像，66.0%[98.5%]；人工图像，61.9%[98.4%]；P = 0.38）。阅片者2的准确性低于阅片者1，但使用真实图像时发现的转移瘤显著更多（敏感性[准确性]：真实图像，78.4%[87.4%]；人工图像，60.8%[80.8%]；P < 0.001）。对于≥5 mm的转移瘤，敏感性无显著差异。不同序列间假阳性结果的数量无显著差异。
一名阅片者使用真实图像时总体敏感性显著更高。对于≥5 mm的转移瘤，相似的检测性能有望将低剂量成像应用于比转移瘤检测更具挑战性小的诊断任务中。
