Improved Delineation of Pulmonary Embolism and Venous Thrombosis Through Frequency Selective Nonlinear Blending in Computed Tomography Malte Niklas Bongers, MD,* Georg Bier , MD,† Christopher Kloth, MD,* Christoph Schabel, MD,* Jan Fritz, MD,‡ Konstantin Nikolaou, MD,* and Marius Horger , MD* Objective: The aim of this study was to test the hypothesis that a novel frequency selective nonlinear blending (NLB) algorithm increases the delineation of pulmo- nary embolism and venous thrombosis in portal-venous phase whole-body stag- ing computed tomography (CT). Materials and Methods: A cohort of 67 patients with incidental pulmonary em- bolism and/or venous thrombosis in contrast-enhanced oncological staging CTwere retrospectively selected. Computed tomography data sets were acquired 65 to 90 seconds after intravenous iodine contrast administration using state-of-the-art multi-detector CT scanners. A novel frequency selective NLB postprocessing tech- nique was applied to reconstructed standard CT images. Two readers determined the most suitable settings to increase the delineation of pulmonary embolism and venous thrombosis. Outcome measure included region of interest and contrast-to- noise ratio (CNR) analyses, image noise, overall image quality, subjective delinea- tion, as well as number and size of emboli and thrombi. Statistical testing included quantitative comparisons of Hounsfield units of thrombus and vessel, image noise and related CNR values and subjective image analyses of image noise, image qual- ity and thrombus delineation, number and size in standard, and NLB images. Results: Using frequency selective NLB settings with a center of 100 HU, delta of 40 HU, and a slope of 5, CNR values of pulmonary embolism (Standard CNR , 10 [6, 16]; NLB CNR , 22 [15, 30]; P < 0.001) and venous thrombosis (Standard CNR , 8 [5, 15]; NLB CNR , 12 [7, 19]; P = 0.0007) increased. Mean vascular enhancement using NLB was significantly higher than in standard images for pulmonary arteries (Stan- dard, 138 [118, 191] HU; NLB, 269 [176, 329] HU; P < 0.0001) and veins (Standard, 120 [103, 162] HU; NLB, 169 [132, 217] HU; P < 0.0001), respectively. Image noise was not significantly different between standard and NLB images (P = 0.64-0.88). There was substantial to almost perfect interrater agreement as well as a significant in- crease of overall image quality (P < 0.004) and subjective delineation of the throm- botic material (P < 0.0001) in both subgroups. Nonlinear blending images revealed 8 additional segmental and 13 subsegmental emboli. Thrombus sizes were not signif- icantly different, but subjective accuracy of the measurement could be significantly in- creased using NLB (P = 0.03). Conclusions: Postprocessing of standard whole-body staging CT images with frequency selective NLB improves image quality and the delineation of pulmo- nary embolism and venous thrombosis. Key Words: pulmonary embolism, venous thrombosis, computed tomography, nonlinear blending, contrast enhancement (Invest Radiol 2017;52: 240–244) A cute pulmonary embolism can cause substantial morbidity and mortality if not diagnosed and treated in a timely fashion. 1 In the ab- sence of contraindications, computed tomography (CT) pulmonary angiog- raphy is the accepted first-line modality of choice for the diagnosis of pulmonary embolism. Advances in multidetector CT technology, including reductions in scan time, less motion artifacts, and higher spatial resolution, have contributed to a high diagnostic accuracy for the detection of pulmo- nary emboli. The pelvic and thigh veins may be included in the same set- ting and can be an efficient means to diagnose deep vein thrombosis. Optimization of CT protocols has been focused both on improv- ing vessel attenuation by adapting the volume and iodine concentration of intravenous contrast agents, using multistep injection protocols and, in particular, through the use of low tube energy techniques. 2 The latter may also result in reduced radiation dose without compromising image quality. Other dose reduction strategies include increasing the pitch, patient-individual modulation of tube current, and lowering the tube current-time product. 3,4 The use of dual-energy CT techniques can in- crease image quality through the ability to create virtual monoenergetic low-energy data sets and may also be utilized to reduce radiation dose and contrast agent volume. 5 Finally, the use of iterative reconstruction techniques afford the use of low-dose protocols without compromising image contrast and image noise. 6 However, pulmonary embolism may be found incidentally on rou- tine CT examinations, such as follow-up staging CT in oncologic pa- tients. 7 Because standard oncologic whole-body CT staging protocols are often performed in a single-pass portal-venous phase that extends from the neck to below the pelvis, the vascular contrast opacification of the pulmonary arteries is often low and the detection of pulmonary embo- lism can be difficult. Postprocessing techniques such as nonlinear blend- ing (NLB) have been used in an attempt to improve the contrast-to-noise ratio (CNR) between the contrast opacified vessel lumen and emboli/ thrombi but have not been adapted into daily routine. 8,9 We evaluated a new NLB algorithm that is based on a frequency split technique and therefore may be capable of increasing the delineation of emboli and thrombi in suboptimally opacified vessels. We tested the hypothesis that a novel frequency selective NLB al- gorithm will increase the delineation of pulmonary embolism and venous thrombosis in portal-venous phase whole-body staging CT. MATERIALS AND METHODS The ethics committee of our institution approved this retrospec- tive study with a waiver for the need for informed consent. Patient Characteristics A retrospective search of our radiology information system for adult patients with pulmonary embolism and/or venous thrombosis on oncologic staging CT that were performed between May 2015 and May 2016, derived a cohort of 67 adult subjects (Table 1). Five subjects had a diagnosis of both pulmonary embolism and venous thrombosis. CT Scanning Protocol All CT examinations were performed on state-of-the-art mul- tislice CT scanners (SOMATOM Force/Flash; Siemens Healthcare Received for publication July 15, 2016; and accepted for publication, after revision, October 2, 2016. From the Departments of *Diagnostic and Interventional Radiology, and †Diagnostic and Interventional Neuroradiology, University Hospital of Tübingen, Tübingen, Germany; and ‡Johns Hopkins University School of Medicine, Russell H. Morgan Department of Radiology and Radiological Science, Baltimore, MD. Conflicts of interest and sources of funding: There was no funding received for this study. Jan Fritz received institutional research funds and speaker's honorarium from Siemens Healthcare USA and is a scientific advisor of Siemens Healthcare USA and Alexion Pharamceuticals, Inc. The other authors have declared that no competing interests exist. Correspondence to: Malte Niklas Bongers, MD, Department of Diagnostic and Inter- ventional Radiology, University Hospital of Tübingen, Hoppe-Seyler-Straße 3, 72076 Tübingen, Germany. E-mail: malte.bongers@med.uni-tuebingen.de. Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0020-9996/17/5204–0240 DOI: 10.1097/RLI.0000000000000333 ORIGINAL ARTICLE 240 www.investigativeradiology.com Investigative Radiology • Volume 52, Number 4, April 2017 Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.