How Different Iterative and Filtered Back Projection Kernels Affect Computed Tomography Numbers and Low Contrast Detectability:

Fractional flow reserve measurement using dynamic CT perfusion imaging in patients with coronary artery disease

Purposes

The objective was to evaluate the accuracy of a novel CT dynamic angiographic imaging (CT-DAI) algorithm for rapid fractional flow reserve (FFR) measurement in patients with coronary artery disease (CAD).

Materials and Methods

This retrospective study included 14 patients (age 58.5 ± 10.6 years, 11 males) with CAD who underwent stress dynamic CT myocardial perfusion scanning with a dual-source CT scanner. The included patients had analyzable proximal and distal coronary artery segments adjacent to the stenosis in the perfusion images and had corresponding invasive catheter-based FFR measurements for that stenosis. An in-house software based on the CT-DAI algorithm was used to compute FFR using the pre- and post- lesion coronary time-enhancement curves obtained from the stress myocardial perfusion images. The CT-DAI derived FFR values were then compared to the corresponding catheter-based invasive FFR values. A coronary artery stenosis was considered functionally significant for FFR value <0.8.

Results

The CT-DAI derived FFR values were in agreement with the invasive FFR values in all 15 coronary arteries in 14 patients, resulting in 100% per-vessel and per-patient diagnostic accuracy. FFR derived using CT-DAI (M = 0.768, SD = 0.156) showed an excellent linear correlation (R = 0.910, P < .001) and statistical indifference (P= .655) with that measured using invasive catheter-based method (M = 0.796, SD = 0.149). Bland-Altman analysis showed no significant proportional bias.

Conclusion

The novel CT-DAI algorithm can reliably compute FFR across a coronary artery stenosis directly from dynamic CT myocardial perfusion images, facilitating rapid on-site hemodynamic assessment of the epicardial coronary artery stenosis in patients with CAD.

Estimating follow-up CTs from geometric deformations of catheter implants in interstitial breast brachytherapy: A feasibility study using electromagnetic tracking

BACKGROUND

Electromagnetic tracking (EMT) systems have been shown to provide valuable information on the geometry of catheter implants in breast cancer patients undergoing interstitial brachytherapy (iBT). In the context of an extended patient-specific, pre-treatment verification, EMT can play a key role in determining the potential need and, if applicable, the appropriate time for treatment adaptation. To detect dosimetric shortcomings the relative position between catheters, and target volume and critical structures must be known. Since EMT cannot provide the anatomical context and standard imaging techniques such as cone-beam CT are not yet available in most brachytherapy suites, it is not possible to detect anatomic changes on a daily or fraction basis, so the need for adaptive planning cannot be identified.

PURPOSE

The aim of this feasibility study is to develop and evaluate a technique capable of estimating follow-up CTs at any time based on the initial treatment planning CT (PCT) and surrogate information about changes of the implant geometry from an EMT system.

METHODS

A deformation vector field is calculated from two different implant reconstructions acquired in treatment position through EMT, the first immediately after the PCT and the second at another time point during the course of treatment. The calculation is based on discrete displacement vectors of pairs of control and target points. These are extrapolated by means of different radial basis functions in order to cover the entire CT volume. The adequate parameters for the calculation of the deformation field were identified. By warping the PCT according to the deformation field, one obtains an estimated CT (ECT) that reflects the geometric changes. For the proof of concept, ECTs were computed for the time point of the clinical follow-up CT (FCT) that is embedded in the treatment workflow after the fourth fraction.

RESULTS

ECT and clinical FCTs of 20 patients were compared to each other quantitatively in terms of absolute Hounsfield unit differences in the planning target volume (PTV) and in a convex hull (CH) enclosing the catheters. The median differences were 31.2  and 29.5 HU for the CH and the PTV, respectively.

CONCLUSION

The proposed ECT approach was able to approximate the "anatomy of the day" and therefore, in principle, allows a dosimetric appraisal of the treatment plan quality before each fraction. In this way, it can contribute to a more detailed patient-specific quality assurance in iBT of the breast and help to identify the timing for a potential treatment adaptation.

Prediction of treatment response following ethanol sclerotherapy of renal cysts by computed tomography

OBJECTIVES

To investigate predictive factors of treatment response following ethanol sclerotherapy of large renal cysts via computed tomography (CT).

METHODS

Retrospective study reviewed 71 patients (61.0 ± 13.2 years; M:F = 32:39) who underwent pretreatment CT and were treated with sclerotherapy of a large (> 5 cm) renal cyst (mean volume: 279.8 cc) using 99% ethanol from January 2010 to February 2019. Patients were followed up at least two times, short-term (defined as < 6 months, median 2.1 months) and long-term (defined as > 1 year, median 15.5 months), via ultrasound or CT. Suboptimal response was defined as the volume of residual cyst > 20 mL in each follow-up. Predictive variables of radiologic findings and radiomics features were analyzed using logistic regression analysis.

RESULTS

Suboptimal response rates were 33.8% and 18.3% at short-term and long-term follow-ups, respectively. In radiologic findings, patients with suboptimal response in the short-term follow-up showed a more frequent estimated cyst volume ≥ 270 mL (OR 14.8, 95% CI 3.9-55.9, p < 0.001) and sinus protrusion (OR 7.0, 95% CI 1.7-28.5, p = 0.007). Cyst volume ≥ 270 mL was also associated with suboptimal response in the long-term follow-up (OR 4.6, 95% CI 1.3-16.9, p = 0.021). When radiomics features were combined, the area under the curve increased from 0.83 to 0.86 and from 0.68 to 0.82 to predict suboptimal response in short-term and long-term follow-ups, respectively.

CONCLUSION

Greater estimated volume, sinus protrusion, and radiomics features of the cysts in pretreatment CT can help predict suboptimal response of renal cyst after sclerotherapy.

ON THE POSSIBILITY TO RESOLVE GADOLINIUM- AND CERIUM-BASED CONTRAST AGENTS FROM THEIR CT NUMBERS IN DUAL-ENERGY COMPUTED TOMOGRAPHY

Cerium oxide nanoparticles with integrated gadolinium have been proved to be useful as contrast agents in magnetic resonance imaging. Of question is their performance in dual-energy computed tomography. The aims of this work are to determine (1) the relation between the computed tomography number and the concentration of the I, Gd or Ce contrast agent and (2) under what conditions it is possible to resolve the type of contrast agent. Hounsfield values of iodoacetic acid, gadolinium acetate and cerium acetate dissolved in water at molar concentrations of 10, 50 and 100 mM were measured in a water phantom using the Siemens SOMATOM Definition Force scanner; gadolinium- and cerium acetate were used as substitutes for the gadolinium-integrated cerium oxide nanoparticles. The relation between the molar concentration of the I, Gd or Ce contrast agent and the Hounsfield value was linear. Concentrations had to be sufficiently high to resolve the contrast agents.

Development and Validation of a Radiomics Model for Differentiating Bone Islands and Osteoblastic Bone Metastases at Abdominal CT

Background It is important to diagnose sclerotic bone lesions in order to determine treatment strategy. Purpose To evaluate the diagnostic performance of a CT radiomics-based machine learning model for differentiating bone islands and osteoblastic bone metastases. Materials and Methods In this retrospective study, patients who underwent contrast-enhanced abdominal CT and were diagnosed with a bone island or osteoblastic metastasis between 2015 to 2019 at either of two different institutions were included: institution 1 for the training set and institution 2 for the external test set. Radiomics features were extracted. The random forest (RF) model was built using 10 selected features, and subsequent 10-fold cross-validation was performed. In the test phase, the RF model was tested with an external test set. Three radiologists reviewed the CT images for the test set. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated for the models and each of the three radiologists. The AUCs of the radiomics model and radiologists were compared. Results A total of 177 patients (89 with a bone island and 88 with metastasis; mean age, 66 years ± 12 [standard deviation]; 111 men) were in the training set, and 64 (23 with a bone island and 41 with metastasis; mean age, 69 years ± 14; 59 men) were in the test set. Radiomics features (n = 1218) were extracted. The average AUC of the RF model from 10-fold cross-validation was 0.89 (sensitivity, 85% [75 of 88 patients]; specificity, 82% [73 of 89 patients]; and accuracy, 84% [148 of 177 patients]). In the test set, the AUC of the trained RF model was 0.96 (sensitivity, 80% [33 of 41 patients]; specificity, 96% [22 of 23 patients]; and accuracy, 86% [55 of 64 patients]). The AUCs for the three readers were 0.95 (95% CI: 0.90, 1.00), 0.96 (95% CI: 0.90, 1.00), and 0.88 (95% CI: 0.80, 0.96). The AUC of radiomics model was higher than that of only reader 3 (0.96 vs 0.88, respectively; P = .03). Conclusion A CT radiomics-based random forest model was proven useful for differentiating bone islands from osteoblastic metastases and showed better diagnostic performance compared with an inexperienced radiologist. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Vannier in this issue.

Comparison of virtual to true unenhanced abdominal computed tomography images acquired using rapid kV-switching dual energy imaging

Objective

To compare “virtual” unenhanced (VUE) computed tomography (CT) images, reconstructed from rapid kVp-switching dual-energy computed tomography (DECT), to “true” unenhanced CT images (TUE), in clinical abdominal imaging. The ability to replace TUE with VUE images would have many clinical and operational advantages.

Methods

VUE and TUE images of 60 DECT datasets acquired for standard-of-care CT of pancreatic cancer were retrospectively reviewed and compared, both quantitatively and qualitatively. Comparisons included quantitative evaluation of CT numbers (Hounsfield Units, HU) measured in 8 different tissues, and 6 qualitative image characteristics relevant to abdominal imaging, rated by 3 experienced radiologists. The observed quantitative and qualitative VUE and TUE differences were compared against boundaries of clinically relevant equivalent thresholds to assess their equivalency, using modified paired t-tests and Bayesian hierarchical modeling.

Results

Quantitatively, in tissues containing high concentrations of calcium or iodine, CT numbers measured in VUE images were significantly different from those in TUE images. CT numbers in VUE images were significantly lower than TUE images when calcium was present (e.g. in the spine, 73.1 HU lower, p < 0.0001); and significantly higher when iodine was present (e.g. in renal cortex, 12.9 HU higher, p < 0.0001). Qualitatively, VUE image ratings showed significantly inferior depiction of liver parenchyma compared to TUE images, and significantly more cortico-medullary differentiation in the kidney.

Conclusions

Significant differences in VUE images compared to TUE images may limit their application and ability to replace TUE images in diagnostic abdominal CT imaging.

Does image normalization and intensity resolution impact texture classification?

Image texture is a very important component in many types of images, including medical images. Medical images are often corrupted by noise and affected by artifacts. Some of the texture-based features that should describe the structure of the tissue under examination may also reflect, for example, the uneven sensitivity of the scanner within the tissue region. This in turn may lead to an inappropriate description of the tissue or incorrect classification. To limit these phenomena, the analyzed regions of interest are normalized. In texture analysis methods, image intensity normalization is usually followed by a reduction in the number of levels coding the intensity. The aim of this work was to analyze the impact of different image normalization methods and the number of intensity levels on texture classification, taking into account noise and artifacts related to uneven background brightness distribution. Analyses were performed on four sets of images: modified Brodatz textures, kidney images obtained by means of dynamic contrast-enhanced magnetic resonance imaging, shoulder images acquired as T2-weighted magnetic resonance images and CT heart and thorax images. The results will be of use for choosing a particular method of image normalization, based on the types of noise and distortion present in the images.

Renal cell carcinoma attenuation values on unenhanced CT: importance of multiple, small region-of-interest measurements

OBJECTIVE

Since it has been suggested that benign renal cysts can be diagnosed at unenhanced CT on the basis of homogeneity and attenuations of 20 HU or less, we determined the prevalence of renal cell carcinomas (RCCs) with these characteristics using two different methods of measuring attenuation.

MATERIALS AND METHODS

After IRB approval, two radiologists obtained unenhanced attenuation values of 104 RCCs (mean size 5.6 cm) using a single, large region of interest (ROI), two-thirds the size of the mass. They were then determined if the masses appeared heterogeneous. Of RCCs measuring 20 HU or less, those which appeared homogeneous were re-measured with multiple (6 or more), small (0.6 cm² or smaller) ROIs dispersed throughout the lesion. Masses with attenuations 20 HU or less were compared to those with masses with HU greater than 20 for any differences in demographic data.

RESULTS

Of 104 RCCS, 24 RCC had HU less than 20 using a large ROI. Of these, 21 appeared heterogeneous and 3 appeared homogeneous. Using multiple small ROIs, these three RCCs revealed maximum attenuation values above 20 HU (Range: 26-32 HU). A greater portion of RCCs measuring 20 HU or less using a large ROI were clear cell sub-type. There were no other differences.

CONCLUSIONS

Renal cell carcinoma can measure 20 HU or less at unenhanced CT when a single large ROI is used. While most appear heterogeneous, some may appear homogeneous, but will likely reveal attenuations greater than 20 HU when multiple, small ROIs are used. This knowledge may prevent some RCCs from being misdiagnosed as cysts on unenhanced CT.