Fides R Schwartz1, Darin P Clark2, Yuqin Ding3, Juan Carlos Ramirez-Giraldo4, Cristian T Badea5, Daniele Marin6. 1. Duke University Health System, Department of Radiology, 2301 Erwin Road, Box 3808, Durham, NC, 27710, United States. Electronic address: fides.schwartz@duke.edu. 2. Quantitative Imaging and Analysis Lab, Department of Radiology, Duke University, Durham, NC, 27710, United States. Electronic address: darin.clark@duke.edu. 3. Duke University Health System, Department of Radiology, 2301 Erwin Road, Box 3808, Durham, NC, 27710, United States; Department of Radiology, Zhongshan Hospital, Fudan University; Shanghai Institute of Medical Imaging, Shanghai, 200032, People's Republic of China. Electronic address: yuqin.ding@duke.edu. 4. CT R&D Collaborations at Siemens Healthineers, 2424 Erwin Road - Hock Plaza, Durham, NC, 27705, United States. Electronic address: juancarlos.ramirezgiraldo@siemens-healthineers.com. 5. Quantitative Imaging and Analysis Lab, Department of Radiology, Duke University, Durham, NC, 27710, United States. Electronic address: Cristian.badea@duke.edu. 6. Duke University Health System, Department of Radiology, 2301 Erwin Road, Box 3808, Durham, NC, 27710, United States. Electronic address: daniele.marin@duke.edu.
Abstract
PURPOSE: Dual-source (DS) CT, dual-energy (DE) field of view (FoV) is limited to the size of the smaller detector array. The purpose was to establish a deep learning-based approach to DE extrapolation by estimating missing image data using data from both tubes to evaluate renal lesions. METHOD: A DE extrapolation deep-learning (DEEDL) algorithm had been trained on DECT data of 50 patients using a DSCT with DE-FoV = 33 cm (Somatom Flash). Data from 128 patients with known renal lesions falling within DE-FoV was retrospectively collected (100/140 kVp; reference dataset 1). A smaller DE-FoV = 20 cm was simulated excluding the renal lesion of interest (dataset 2) and the DEEDL was applied to this dataset. Output from the DEEDL algorithm was evaluated using ReconCT v14.1 and Syngo.via. Mean attenuation values in lesions on mixed images (HU) were compared calculating the root-mean-squared-error (RMSE) between the datasets using MATLAB R2019a. RESULTS: The DEEDL algorithm performed well reproducing the image data of the kidney lesions (Bosniak 1 and 2: 125, Bosniak 2F: 6, Bosniak 3: 1 and Bosniak 4/(partially) solid: 32) with RSME values of 10.59 HU, 15.7 HU for attenuation, virtual non-contrast, respectively. The measurements performed in dataset 1 and 2 showed strong correlation with linear regression (r2: attenuation = 0.89, VNC = 0.63, iodine = 0.75), lesions were classified as enhancing with an accuracy of 0.91. CONCLUSION: This DEEDL algorithm can be used to reconstruct a full dual-energy FoV from restricted data, enabling reliable HU value measurements in areas not covered by the smaller FoV and evaluation of renal lesions.
PURPOSE: Dual-source (DS) CT, dual-energy (DE) field of view (FoV) is limited to the size of the smaller detector array. The purpose was to establish a deep learning-based approach to DE extrapolation by estimating missing image data using data from both tubes to evaluate renal lesions. METHOD: A DE extrapolation deep-learning (DEEDL) algorithm had been trained on DECT data of 50 patients using a DSCT with DE-FoV = 33 cm (Somatom Flash). Data from 128 patients with known renal lesions falling within DE-FoV was retrospectively collected (100/140 kVp; reference dataset 1). A smaller DE-FoV = 20 cm was simulated excluding the renal lesion of interest (dataset 2) and the DEEDL was applied to this dataset. Output from the DEEDL algorithm was evaluated using ReconCT v14.1 and Syngo.via. Mean attenuation values in lesions on mixed images (HU) were compared calculating the root-mean-squared-error (RMSE) between the datasets using MATLAB R2019a. RESULTS: The DEEDL algorithm performed well reproducing the image data of the kidney lesions (Bosniak 1 and 2: 125, Bosniak 2F: 6, Bosniak 3: 1 and Bosniak 4/(partially) solid: 32) with RSME values of 10.59 HU, 15.7 HU for attenuation, virtual non-contrast, respectively. The measurements performed in dataset 1 and 2 showed strong correlation with linear regression (r2: attenuation = 0.89, VNC = 0.63, iodine = 0.75), lesions were classified as enhancing with an accuracy of 0.91. CONCLUSION: This DEEDL algorithm can be used to reconstruct a full dual-energy FoV from restricted data, enabling reliable HU value measurements in areas not covered by the smaller FoV and evaluation of renal lesions.
Authors: Sascha Kaufmann; Alexander Sauter; Daniel Spira; Sergios Gatidis; Dominik Ketelsen; Martin Heuschmid; Claus D Claussen; Christoph Thomas Journal: Acad Radiol Date: 2013-03-13 Impact factor: 3.173
Authors: Hu Chen; Yi Zhang; Mannudeep K Kalra; Feng Lin; Yang Chen; Peixi Liao; Jiliu Zhou; Ge Wang Journal: IEEE Trans Med Imaging Date: 2017-06-13 Impact factor: 10.048
Authors: Hersh Chandarana; Alec J Megibow; Benjamin A Cohen; Ramya Srinivasan; Danny Kim; Christianne Leidecker; Michael Macari Journal: AJR Am J Roentgenol Date: 2011-06 Impact factor: 3.959
Authors: Jens Altenbernd; Till A Heusner; Adrian Ringelstein; Susanne C Ladd; Michael Forsting; Gerald Antoch Journal: Eur Radiol Date: 2010-10-10 Impact factor: 5.315
Authors: Martin Petersilka; Herbert Bruder; Bernhard Krauss; Karl Stierstorfer; Thomas G Flohr Journal: Eur J Radiol Date: 2008-10-07 Impact factor: 3.528
Authors: Darin P Clark; Fides R Schwartz; Daniele Marin; Juan C Ramirez-Giraldo; Cristian T Badea Journal: Med Phys Date: 2020-07-06 Impact factor: 4.071
Authors: Gert Jan Pelgrim; Robbert W van Hamersvelt; Martin J Willemink; Bernhard T Schmidt; Thomas Flohr; Arnold Schilham; Julien Milles; Matthijs Oudkerk; Tim Leiner; Rozemarijn Vliegenthart Journal: Eur Radiol Date: 2017-02-06 Impact factor: 5.315