Standard Clinical Computed Tomography Research Articles

Quality control of radiomic features using 3D-printed CT phantoms.

.Purpose: The lack of standardization in quantitative radiomic measures of tumors seen on computed tomography (CT) scans is generally recognized as an unresolved issue. To develop reliable clinical applications, radiomics must be robust across different CT scan modes, protocols, software, and systems. We demonstrate how custom-designed phantoms, imprinted with human-derived patterns, can provide a straightforward approach to validating longitudinally stable radiomic signature values in a clinical setting.Approach: Described herein is a prototype process to design an anatomically informed 3D-printed radiomic phantom. We used a multimaterial, ultra-high-resolution 3D printer with voxel printing capabilities. Multiple tissue regions of interest (ROIs), from four pancreas tumors, one lung tumor, and a liver background, were extracted from digital imaging and communication in medicine (DICOM) CT exam files and were merged together to develop a multipurpose, circular radiomic phantom (18 cm diameter and 4 cm width). The phantom was scanned 30 times using standard clinical CT protocols to test repeatability. Features that have been found to be prognostic for various diseases were then investigated for their repeatability and reproducibility across different CT scan modes.Results: The structural similarity index between the segment used from the patients’ DICOM image and the phantom CT scan was 0.71. The coefficient variation for all assessed radiomic features was across 30 repeat scans of the phantom. The percent deviation (pDV) from the baseline value, which was the mean feature value determined from repeat scans, increased with the application of the lung convolution kernel, changes to the voxel size, and increases in the image noise. Gray level co-occurrence features, contrast, dissimilarity, and entropy were particularly affected by different scan modes, presenting with .Conclusions: Previously discovered prognostic and popular radiomic features are variable in practice and need to be interpreted with caution or excluded from clinical implementation. Voxel-based 3D printing can reproduce tissue morphology seen on CT exams. We believe that this is a flexible, yet practical, way to design custom phantoms to validate and compare radiomic metrics longitudinally, over time, and across systems.

Molecular Imaging of Pulmonary Tuberculosis in an Ex-Vivo Mouse Model Using Spectral Photon-Counting Computed Tomography and Micro-CT

Assessment of disease burden and drug efficacy is achieved preclinically using high resolution micro computed tomography (CT). However, micro-CT is not applicable to clinical human imaging due to operating at high dose. In addition, the technology differences between micro-CT and standard clinical CT prevent direct translation of preclinical applications. The current proof-of-concept study presents spectral photon-counting CT as a clinically translatable, molecular imaging tool by assessing contrast uptake in an ex-vivo mouse model of pulmonary tuberculosis (TB). Iodine, a common contrast used in clinical CT imaging, was introduced into a murine model of TB. The excised mouse lungs were imaged using a standard micro-CT subsystem (SuperArgus) and the contrast enhanced TB lesions quantified. The same lungs were imaged using a spectral photoncounting CT system (MARS small-bore scanner). Iodine and soft tissues (water and lipid) were materially separated, and iodine uptake quantified. The volume of the TB infection quantified by spectral CT and micro-CT was found to be 2.96 mm3 and 2.83 mm3, respectively. This proof-of-concept study showed that spectral photon-counting CT could be used as a predictive preclinical imaging tool for the purpose of facilitating drug discovery and development. Also, as this imaging modality is available for human trials, all applications are translatable to human imaging. In conclusion, spectral photon-counting CT could accelerate a deeper understanding of infectious lung diseases using targeted pharmaceuticals and intrinsic markers, and ultimately improve the efficacy of therapies by measuring drug delivery and response to treatment in animal models and later in humans.

PWD-3DNet: A Deep Learning-Based Fully-Automated Segmentation of Multiple Structures on Temporal Bone CT Scans

The temporal bone is a part of the lateral skull surface that contains organs responsible for hearing and balance. Mastering surgery of the temporal bone is challenging because of this complex and microscopic three-dimensional anatomy. Segmentation of intra-temporal anatomy based on computed tomography (CT) images is necessary for applications such as surgical training and rehearsal, amongst others. However, temporal bone segmentation is challenging due to the similar intensities and complicated anatomical relationships among critical structures, undetectable small structures on standard clinical CT, and the amount of time required for manual segmentation. This paper describes a single multi-class deep learning-based pipeline as the first fully automated algorithm for segmenting multiple temporal bone structures from CT volumes, including the sigmoid sinus, facial nerve, inner ear, malleus, incus, stapes, internal carotid artery and internal auditory canal. The proposed fully convolutional network, PWD-3DNet, is a patch-wise densely connected (PWD) three-dimensional (3D) network. The accuracy and speed of the proposed algorithm was shown to surpass current manual and semi-automated segmentation techniques. The experimental results yielded significantly high Dice similarity scores and low Hausdorff distances for all temporal bone structures with an average of 86% and 0.755 millimeter (mm), respectively. We illustrated that overlapping in the inference sub-volumes improves the segmentation performance. Moreover, we proposed augmentation layers by using samples with various transformations and image artefacts to increase the robustness of PWD-3DNet against image acquisition protocols, such as smoothing caused by soft tissue scanner settings and larger voxel sizes used for radiation reduction. The proposed algorithm was tested on low-resolution CTs acquired by another center with different scanner parameters than the ones used to create the algorithm and shows potential for application beyond the particular training data used in the study.

A Proposed Computed Tomography Contrast Agent Using Carboxybetaine Zwitterionic Tantalum Oxide Nanoparticles

The aim of this study was to produce and evaluate a proposed computed tomography (CT) contrast agent based on carboxybetaine zwitterionic (CZ)-coated soluble tantalum oxide (TaO) nanoparticles (NPs). We chose tantalum to provide superior imaging performance compared with current iodine-based clinical CT contrast agents. We developed the CZ coating to provide biological and physical performance similar to that of current iodinated contrast agents. In addition, the aim of this study was to evaluate the imaging, biological, and physicochemical performance of this proposed contrast agent compared with clinically used iodinated agents. We evaluated CT imaging performance of our CZ-TaO NPs compared with that of an iodinated agent in live rats, imaged centrally located within a tissue-equivalent plastic phantom that simulated a large patient. To evaluate vascular contrast enhancement, we scanned the rats' great vessels at high temporal resolution during and after contrast agent injection. We performed several in vivo CZ-TaO NP studies in healthy rats to evaluate tolerability. These studies included injecting the agent at the anticipated clinical dose (ACD) and at 3 times and 6 times the ACD, followed by longitudinal hematology to assess impact to blood cells and organ function (from 4 hours to 1 week). Kidney histological analysis was performed 48 hours after injection at 3 times the ACD. We measured the elimination half-life of CZ-TaO NPs from blood, and we monitored acute kidney injury biomarkers with a kidney injury assay using urine collected from 4 hours to 1 week. We measured tantalum retention in individual organs and in the whole carcass 48 hours after injection at ACD. Carboxybetaine zwitterionic TaO NPs were synthesized and analyzed in detail. We used multidimensional nuclear magnetic resonance to determine surface functionality of the NPs. We measured NP size and solution properties (osmolality and viscosity) of the agent over a range of tantalum concentrations, including the high concentrations required for standard clinical CT imaging. Computed tomography imaging studies demonstrated image contrast improvement of approximately 40% to 50% using CZ-TaO NPs compared with an iodinated agent injected at the same mass concentration. Blood and organ analyses showed no adverse effects after injection in healthy naive rats at 3 times the ACD. Retention of tantalum at 48 hours after injection was less than 2% of the injected dose in the whole carcass, which very closely matched the reported retention of existing commercial iodine-based contrast agents. Urine analysis of sensitive markers for acute kidney injury showed no responses at 1 week after injection at 3 times the ACD; however, a moderate response in the neutrophil gelatinase-associated lipocalin biomarker was measured at 24 and 48 hours. Compared with other TaO NPs reported in the literature, CZ-TaO NPs had relatively low osmolality and viscosity at concentrations greater than 200 mg Ta/mL and were similar in these physical properties to dimeric iodine-based contrast agents. We found that a CZ-TaO NP-based contrast agent is potentially viable for general-purpose clinical CT imaging. Our results suggest that such an agent can be formulated with clinically viable physicochemical properties, can be biologically safe and cleared rapidly in urine, and can provide substantially improved image contrast at CT compared with current iodinated agents.

SU‐F‐SPS‐03: Direct Measurement of Organ Doses Resulting From Head and Cervical Spine Trauma CT Protocols

Purpose:This retrospective study analyzes the exposure history of emergency department (ED) patients undergoing head and cervical spine trauma computed tomography (CT) studies. This study investigated dose levels received by trauma patients and addressed any potential concerns regarding radiation dose issues.Methods:Under proper IRB approval, a cohort of 300 trauma cases of head and cervical spine trauma CT scans received in the ED was studied. The radiological image viewing software of the hospital was used to view patient images and image data. The following parameters were extracted: the imaging history of patients, the reported dose metrics from the scanner including the volumetric CT Dose Index (CTDIvol) and Dose Length Product (DLP). A postmortem subject was scanned using the same scan techniques utilized in a standard clinical head and cervical spine trauma CT protocol with 120 kVp and 280 mAs. The CTDIvol was recorded for the subject and the organ doses were measured using optically stimulated luminescent (OSL) dosimeters. Typical organ doses to the brain, thyroid, lens, salivary glands, and skin, based on the cadaver studies, were then calculated and reported for the cohort.Results:The CTDIvol reported by the CT scanner was 25.5 mGy for the postmortem subject. The average CTDIvol from the patient cohort was 34.1 mGy. From these metrics, typical average organ doses in mGy were found to be: Brain (44.57), Thyroid (33.40), Lens (82.45), Salivary Glands (61.29), Skin (47.50). The imaging history of the cohort showed that on average trauma patients received 26.1 scans over a lifetime.Conclusion:The average number of scans received on average by trauma ED patients shows that radiation doses in trauma patients may be a concern. Available dose tracking software would be helpful to track doses in trauma ED patients, highlighting the importance of minimizing unnecessary scans and keeping doses ALARA.

Tracheal tube obstruction in mechanically ventilated patients assessed by high-resolution computed tomography.

Tracheal intubation compromises mucus clearance and secretions accumulate inside the tracheal tube (TT). The aim of this study was to evaluate with a novel methodology TT luminal obstruction in critically ill patients. This was a three-phase study: (1) the authors collected 20 TTs at extubation. High-resolution computed tomography (CT) was performed to determine cross-sectional area (CSA) and mucus distribution within the TT; (2) five TTs partially filled with silicone were used to correlate high-resolution CT results and increased airflow resistance; and (3) 20 chest CT scans of intubated patients were reviewed for detection of secretions in ventilated patients' TT. Postextubation TTs showed a maximum CSA reduction of (mean±SD) 24.9±3.9% (range 3.3 to 71.2%) after a median intubation of 4.5 (interquartile range 2.5 to 6.5) days. CSA progressively decreased from oral to lung end of used TTs. The luminal volume of air was different between used and new TTs for all internal diameters (P<0.01 for new vs. used TTs for all studied internal diameters). The relationship between pressure drop and increasing airflow rates was nonlinear and depended on minimum CSA available to ventilation. Weak correlation was found between TT occlusion and days of intubation (R²=0.352, P=0.006). With standard clinical chest CT scans, 6 of 20 TTs showed measurable secretions with a CSA reduction of 24.0±3.9%. TT luminal narrowing is a common finding and correlates with increased airflow resistance. The authors propose high-resolution CT as a novel technique to visualize and quantify secretions collected within the TT lumen.

Variation in the human ribs geometrical properties and mechanical response based on X-ray computed tomography images resolution

Computational models of the human body are commonly used for injury prediction in automobile safety research. To create these models, the geometry of the human body is typically obtained from segmentation of medical images such as computed tomography (CT) images that have a resolution between 0.2 and 1mm/pixel. While the accuracy of the geometrical and structural information obtained from these images depend greatly on their resolution, the effect of image resolution on the estimation of the ribs geometrical properties has yet to be established. To do so, each of the thirty-four sections of ribs obtained from a Post Mortem Human Surrogate (PMHS) was imaged using three different CT modalities: standard clinical CT (clinCT), high resolution clinical CT (HRclinCT), and microCT. The images were processed to estimate the rib cross-section geometry and mechanical properties, and the results were compared to those obtained from the microCT images by computing the ‘deviation factor’, a metric that quantifies the relative difference between results obtained from clinCT and HRclinCT to those obtained from microCT. Overall, clinCT images gave a deviation greater than 100%, and were therefore deemed inadequate for the purpose of this study. HRclinCT overestimated the rib cross-sectional area by 7.6%, the moments of inertia by about 50%, and the cortical shell area by 40.2%, while underestimating the trabecular area by 14.7%. Next, a parametric analysis was performed to quantify how the variations in the estimate of the geometrical properties affected the rib predicted mechanical response under antero-posterior loading. A variation of up to 45% for the predicted peak force and up to 50% for the predicted stiffness was observed. These results provide a quantitative estimate of the sensitivity of the response of the FE model to the resolution of the images used to generate it. They also suggest that a correction factor could be derived from the comparison between microCT and HRclinCT images to improve the response of the model developed based on HRclinCT images.

Aging adult skull remains through radiological density estimates: A comparison of different computed tomography systems and the use of computer simulations to judge the accuracy of results

IntroductionThe objective of this article was to explore age-at-death estimates in forensic medicine, which were methodically based on age-dependent, radiologically defined bone-density (HC) decay and which were investigated with a standard clinical computed tomography (CT) system. Such density decay was formerly discovered with a high-resolution flat-panel CT in the skulls of adult females. The development of a standard CT methodology for age estimations – with thousands of installations – would have the advantage of being applicable everywhere, whereas only few flat-panel prototype CT systems are in use worldwide. MethodsA Multi-Slice CT scanner (MSCT) was used to obtain 22,773 images from 173 European human skulls (89 male, 84 female), taken from a population of patients from the Department of Neuroradiology at the University Hospital Giessen and Marburg during 2010 and 2011. An automated image analysis was carried out to evaluate HC of all images. The age dependence of HC was studied by correlation analysis. The prediction accuracy of age-at-death estimates was calculated. Computer simulations were carried out to explore the influence of noise on the accuracy of age predictions. ResultsHuman skull HC values strongly scatter as a function of age for both sexes. Adult male skull bone-density remains constant during lifetime. Adult female HC decays during lifetime, as indicated by a correlation coefficient (CC) of −0.53. Prediction errors for age-at-death estimates for both of the used scanners are in the range of ±18 years at a 75% confidence interval (CI). Computer simulations indicate that this is the best that can be expected for such noisy data. ConclusionsOur results indicate that HC-decay is indeed present in adult females and that it can be demonstrated both by standard and by high-resolution CT methods, applied to different subject groups of an identical population. The weak correlation between HC and age found by both CT methods only enables a method to estimate age-at-death with limited practical relevance since the errors of the estimates are large. Computer simulations clearly indicate that data with less noise and CCs in the order of −0.97 or less would be necessary to enable age-at-death estimates with an accuracy of ±5 years at a 75% CI.

The effect of CT dose on glenohumeral joint congruency measurements using 3D reconstructed patient-specific bone models

The study of joint congruency at the glenohumeral joint of the shoulder using computed tomography (CT) and three-dimensional (3D) reconstructions of joint surfaces is an area of significant clinical interest. However, ionizing radiation delivered to patients during CT examinations is much higher than other types of radiological imaging. The shoulder represents a significant challenge for this modality as it is adjacent to the thyroid gland and breast tissue. The objective of this study was to determine the optimal CT scanning techniques that would minimize radiation dose while accurately quantifying joint congruency of the shoulder. The results suggest that only one-tenth of the standard applied total current (mA) and a pitch ratio of 1.375:1 was necessary to produce joint congruency values consistent with that of the higher dose scans. Using the CT scanning techniques examined in this study, the effective dose applied to the shoulder to quantify joint congruency was reduced by 88.9% compared to standard clinical CT imaging techniques.

Initiating a post-mortem computed tomography service: the radiologist’s perspective

Many radiologists have extensive clinical Computed Tomography (CT) experience that is useful for post-mortem scanning; but there are some differences in approach. Firstly clinical scan protocols optimize diagnostic information whilst keeping radiation dose low. This is not an issue for post-mortem scanning, but X-ray tube heat loading and difficulty in handling thousands of images can require similar compromises. Knowledge of normal post-mortem changes that can simulate disease is required for image reporting. Our practice is to use CT as an adjunct to the autopsy. It is therefore important to know what information is useful to the pathologist prior to the autopsy and what areas may be easily visible to imaging but not the pathologist. When reporting post-mortem scans there may be multiple abnormalities related to trauma or post-mortem changes that are readily apparent on external examination. In this case reporting all abnormalities may be unnecessary and time consuming. It is our practice to issue a short report, taking about the same amount of time as a standard clinical CT report. This is done with the option of re-discussing the images at a later date in more detail if required.

Grundlagen der Flachdetektor-CT (FD-CT)

Flat detectors (FDs) have been developed for use in radiography and fluoroscopy to replace standard X-ray film, film-screen combinations and image intensifiers (II). In comparison to X-ray film and II, FD technology offers higher dynamic range, dose reduction, fast digital readout and the possibility for dynamic acquisitions of image series, yet keeping to a compact design. It appeared logical to employ FD designs also for computed tomography (CT) imaging. FDCT has meanwhile become widely accepted for interventional and intra-operative imaging using C-arm systems. Additionally, the introduction of FD technology was a milestone for soft-tissue CT imaging in the interventional suite which was not possible with II systems in the past. This review focuses on technical and performance issues of FD technology and its wide range of applications for CT imaging. FDCT is not aimed at challenging standard clinical CT as regards to the typical diagnostic examinations, but it has already proven unique for a number of dedicated CT applications offering distinct practical advantages, above all the availability of immediate CT imaging during an intervention.

Targeted Gold Nanoparticles Enable Molecular CT Imaging of Cancer

X-ray based computed tomography (CT) is among the most convenient imaging/diagnostic tools in hospitals today in terms of availability, efficiency, and cost. However, in contrast to magnetic resonance imaging (MRI) and various nuclear medicine imaging modalities, CT is not considered a molecular imaging modality since targeted and molecularly specific contrast agents have not yet been developed. Here we describe a targeted molecular imaging platform that enables, for the first time, cancer detection at the cellular and molecular level with standard clinical CT. The method is based on gold nanoprobes that selectively and sensitively target tumor selective antigens while inducing distinct contrast in CT imaging (increased X-ray attenuation). We present an in vitro proof of principle demonstration for head and neck cancer, showing that the attenuation coefficient for the molecularly targeted cells is over 5 times higher than for identical but untargeted cancer cells or for normal cells. We expect this novel imaging tool to lead to significant improvements in cancer therapy due to earlier detection, accurate staging, and microtumor identification.

Flat-detector computed tomography (FD-CT)

Flat-panel detectors or, synonymously, flat detectors (FDs) have been developed for use in radiography and fluoroscopy with the defined goal to replace standard X-ray film, film-screen combinations and image intensifiers by an advanced sensor system. FD technology in comparison to X-ray film and image intensifiers offers higher dynamic range, dose reduction, fast digital readout and the possibility for dynamic acquisitions of image series, yet keeping to a compact design. It appeared logical to employ FD designs also for computed tomography (CT) imaging. Respective efforts date back a few years only, but FD-CT has meanwhile become widely accepted for interventional and intra-operative imaging using C-arm systems. FD-CT provides a very efficient way of combining two-dimensional (2D) radiographic or fluoroscopic and 3D CT imaging. In addition, FD technology made its way into a number of dedicated CT scanner developments, such as scanners for the maxillo-facial region or for micro-CT applications. This review focuses on technical and performance issues of FD technology and its full range of applications for CT imaging. A comparison with standard clinical CT is of primary interest. It reveals that FD-CT provides higher spatial resolution, but encompasses a number of disadvantages, such as lower dose efficiency, smaller field of view and lower temporal resolution. FD-CT is not aimed at challenging standard clinical CT as regards to the typical diagnostic examinations; but it has already proven unique for a number of dedicated CT applications, offering distinct practical advantages, above all the availability of immediate CT imaging in the interventional suite or the operating room.

A three-dimensional finite element model from computed tomography data: a semi-automated method.

Three-dimensional finite element analysis is one of the best ways to assess stress and strain distributions in complex bone structures. However, accuracy in the results may be achieved only when accurate input information is given. A semi-automated method to generate a finite element (FE) model using data retrieved from computed tomography (CT) was developed. Due to its complex and irregular shape, the glenoid part of a left embalmed scapula bone was chosen as working material. CT data were retrieved using a standard clinical CT scanner (Siemens Somatom Plus 2, Siemens AG, Germany). This was done to produce a method that could later be utilized to generate a patient-specific FE model. Different methods of converting Hounsfield unit (HU) values to apparent densities and subsequently to Young's moduli were tested. All the models obtained were loaded using three-dimensional loading conditions taken from literature, corresponding to an arm abduction of 90 degrees. Additional models with different amounts of elements were generated to verify convergence. Direct comparison between the models showed that the best method to convert HU values directly to apparent densities was to use different equations for cancellous and cortical bone. In this study, a reliable method of determining both geometrical data and bone properties from patient CT scans for the semi-automated generation of an FE model is presented.

In vivo measurement of lung volumes in mice.

We describe longitudinal measurements of functional residual capacity (FRC) in breathing mice using a clinical computed tomography (CT) scanner. Lungs of anesthetized mice from the A/J and C3H/HeJ strains were scanned over a 10-s period. Using a fixed threshold for CT density, we could accurately and reproducibly obtain the amount of air in the lungs at FRC, with a 10% coefficient of variation. Total lung volume, and the fractions in left and right lungs, were measured in the two strains from 4 to 12 wk of age. Results show that in both strains the FRC increases only up to 6 wk of age and then remains stable despite a steady increase in body weight. Over this time period, FRC was consistently about 50% greater in the C3H/HeJ strain compared with the A/J strain. The C3H/HeJ strain also has a significantly smaller fraction of the total lung volume in the left lung. We conclude that accurate measurements of FRC in breathing mice can be made using a standard clinical CT scanner. This method may be useful for repeated noninvasive assessment of both structural and functional changes in the lungs of experimental and genetically manipulated mice.