Deep learning-based image reconstruction of 40-keV virtual monoenergetic images of dual-energy CT for the assessment of hypoenhancing hepatic metastasis

被引:30
作者
Lee, Taehee [1 ]
Lee, Jeong Min [1 ,2 ,3 ]
Yoon, Jeong Hee [1 ,2 ]
Joo, Ijin [1 ,2 ]
Bae, Jae Seok [1 ]
Yoo, Jeongin [1 ]
Kim, Jae Hyun [1 ]
Ahn, Chulkyun [4 ]
Kim, Jong Hyo [2 ,4 ,5 ]
机构
[1] Seoul Natl Univ Hosp, Dept Radiol, 101 Daehak Ro, Seoul 03080, South Korea
[2] Seoul Natl Univ, Dept Radiol, Coll Med, 103 Daehak Ro, Seoul 03080, South Korea
[3] Seoul Natl Univ, Inst Radiat Med, Med Res Ctr, 101 Daehak Ro, Seoul 03080, South Korea
[4] Seoul Natl Univ, Grad Sch Convergence Sci & Technol, Dept Transdisciplinary Studies, Seoul 08826, South Korea
[5] Adv Inst Convergence Technol, Ctr Med IT Convergence Technol Res, Suwon 16229, South Korea
关键词
Liver; Metastasis; Dual-energy scanned projection radiography; Image enhancement; Deep learning; LIVER METASTASES; OBSERVER PERFORMANCE; CHEST TOMOSYNTHESIS; PULMONARY NODULES; CONTRAST; ALGORITHM;
D O I
10.1007/s00330-022-08728-0
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Objectives To evaluate the diagnostic value of deep learning model (DLM) reconstructed dual-energy CT (DECT) low-keV virtual monoenergetic imaging (VMI) for assessing hypoenhancing hepatic metastases. Methods This retrospective study included 131 patients who underwent contrast-enhanced DECT (80-kVp and 150-kVp with a tin filter) in the portal venous phase for hepatic metastasis surveillance. Linearly blended images simulating 100-kVp images (100-kVp), standard 40-keV VMI images (40-keV VMI), and post-processed 40-keV VMI using a vendor-agnostic DLM (i.e., DLM 40-keV VMI) were reconstructed. Lesion conspicuity and diagnostic acceptability were assessed by three independent reviewers and compared using the Wilcoxon signed-rank test. The contrast-to-noise ratios (CNRs) were also measured placing ROIs in metastatic lesions and liver parenchyma. The detection performance of hepatic metastases was assessed by using a jackknife alternative free-response ROC method. The consensus by two independent radiologists was used as the reference standard. Results DLM 40-keV VMI, compared to 40-keV VMI and 100-kVp, showed a higher lesion-to-liver CNR (8.25 +/- 3.23 vs. 6.05 +/- 2.38 vs. 5.99 +/- 2.00), better lesion conspicuity (4.3 (4.0-4.7) vs. 3.7 (3.7-4.0) vs. 3.7 (3.3-4.0)), and better diagnostic acceptability (4.3 (4.0-4.3) vs. 3.0 (2.7-3.3) vs. 4.0 (4.0-4.3)) (p < 0.001 for all). For lesion detection (246 hepatic metastases in 68 patients), the figure of merit was significantly higher with DLM 40-keV VMI than with 40-keV VMI (0.852 vs. 0.822, p = 0.012), whereas no significant difference existed between DLM 40-keV VMI and 100-kVp (0.852 vs. 0.842, p = 0.31). Conclusions DLM 40-keV VMI provided better image quality and comparable diagnostic performance for detecting hypoenhancing hepatic metastases compared to linearly blended images.
引用
收藏
页码:6407 / 6417
页数:11
相关论文
共 43 条
[1]   Combined Low-dose Simulation and Deep Learning for CT Denoising: Application of Ultra-low-dose Cardiac CTA [J].
Ahn, Chul Kyun ;
Jin, Hyeongmin ;
Heo, Changyong ;
Kim, Jong Hyo .
MEDICAL IMAGING 2019: PHYSICS OF MEDICAL IMAGING, 2019, 10948
[2]   Combined Low-dose Simulation and Deep Learning for CT Denoising: Application in Ultra-low-dose Chest CT [J].
Ahn, Chulkyun ;
Heo, Changyong ;
Kim, Jong Hyo .
INTERNATIONAL FORUM ON MEDICAL IMAGING IN ASIA 2019, 2019, 11050
[3]   Deep learning reconstruction improves image quality of abdominal ultra-high-resolution CT [J].
Akagi, Motonori ;
Nakamura, Yuko ;
Higaki, Toru ;
Narita, Keigo ;
Honda, Yukiko ;
Zhou, Jian ;
Yu, Zhou ;
Akino, Naruomi ;
Awai, Kazuo .
EUROPEAN RADIOLOGY, 2019, 29 (11) :6163-6171
[4]   A software tool for increased efficiency in observer performance studies in radiology [J].
Börjesson, S ;
Håkansson, M ;
Båth, M ;
Kheddache, S ;
Svensson, S ;
Tingberg, A ;
Grahn, A ;
Ruschin, M ;
Hemdal, B ;
Mattsson, S ;
Månsson, LG .
RADIATION PROTECTION DOSIMETRY, 2005, 114 (1-3) :45-52
[5]   Analysis of location specific observer performance data: Validated extensions of the jackknife free-response (JAFROC) method [J].
Chakraborty, Dev P. .
ACADEMIC RADIOLOGY, 2006, 13 (10) :1187-1193
[6]   Observer studies involving detection and localization: Modeling, analysis, and validation [J].
Chakraborty, DP ;
Berbaum, KS .
MEDICAL PHYSICS, 2004, 31 (08) :2313-2330
[7]  
Cronbach LJ, 1951, PSYCHOMETRIKA, V16, P297
[8]   A noise-optimized virtual monoenergetic reconstruction algorithm improves the diagnostic accuracy of late hepatic arterial phase dual-energy CT for the detection of hypervascular liver lesions [J].
De Cecco, Carlo N. ;
Caruso, Damiano ;
Schoepf, U. Joseph ;
De Santis, Domenico ;
Muscogiuri, Giuseppe ;
Albrecht, Moritz H. ;
Meinel, Felix G. ;
Wichmann, Julian L. ;
Burchett, Philip F. ;
Varga-Szemes, Akos ;
Sheafor, Douglas H. ;
Hardie, Andrew D. .
EUROPEAN RADIOLOGY, 2018, 28 (08) :3393-3404
[9]   Methods for Calculating Sensitivity and Specificity of Clustered Data: A Tutorial [J].
Genders, Tessa S. S. ;
Spronk, Sandra ;
Stijnen, Theo ;
Steyerberg, Ewout W. ;
Lesaffre, Emmanuel ;
Hunink, M. G. Myriam .
RADIOLOGY, 2012, 265 (03) :910-916
[10]   Deep-learning-based direct synthesis of low-energy virtual monoenergetic images with multi-energy CT [J].
Gong, Hao ;
Marsh, Jeffrey F. ;
D'Souza, Karen N. ;
Huber, Nathan R. ;
Rajendran, Kishore ;
Fletcher, Joel G. ;
McCollough, Cynthia H. ;
Leng, Shuai .
JOURNAL OF MEDICAL IMAGING, 2021, 8 (05)