Interscan Variability Research Articles

Prospective evaluation of the influence of iterative reconstruction on the reproducibility of coronary calcium quantification in reduced radiation dose 320 detector row CT

BackgroundCoronary artery calcium (CAC) predicts coronary heart disease events and is important for individualized cardiac risk assessment. This report assesses the interscan variability of CT for coronary calcium quantification using image acquisition with standard and reduced radiation dose protocols and whether the use of reduced radiation dose acquisition with iterative reconstruction (IR; “reduced-dose/IR ”) allows for similar image quality and reproducibility when compared to standard radiation dose acquisition with filtered back projection (FBP; “standard-dose/FBP”) on 320-detector row computed tomography (320-CT). Methods200 consecutive patients (60 ± 9 years, 59% male) prospectively underwent two standard- and two reduced-dose acquisitions (800 total scans, 1600 reconstructions) using 320 slice CT and 120 kV tube voltage. Automated tube current modulation was used and for reduced-dose scans, prescribed tube current was lowered by 70%. Image noise and Agatston scores were determined and compared. ResultsRegarding stratification by Agatston score categories (0, 1–10, 11–100, 101–400, >400), reduced-dose/IR versus standard-dose/FBP had excellent agreement at 89% (95% CI: 86–92%) with kappa 0.86 (95% CI: 0.81–0.90). Standard-dose/FBP rescan agreement was 93% (95% CI: 89–96%) with kappa = 0.91 (95% CI: 0.86–0.95) while reduced-dose/IR rescan agreement was similar at 91% (95% CI: 87–94%) with kappa 0.88 (95% CI: 0.83–0.93). Image noise was significantly higher but clinically acceptable for reduced-dose/IR (18 Hounsfield Unit [HU] mean) compared to standard-dose/FBP (16 HU; p < 0.0001). Median radiation exposure was 74% lower for reduced- (0.37 mSv) versus standard-dose (1.4 mSv) acquisitions. ConclusionRescan agreement was excellent for reduced-dose image acquisition with iterative reconstruction and standard-dose acquisition with filtered back projection for the quantification of coronary calcium by CT. These methods make it possible to reduce radiation exposure by 74%. Clinical trial registrationURL: https://clinicaltrials.gov/ct2/show/NCT01621594. Unique identifierNCT01621594.

Intra- and interscanner variability of magnetic resonance imaging based volumetry in multiple sclerosis

Brain volumetric measurements in multiple sclerosis (MS) reflect not only disease-specific processes but also other sources of variability. The latter has to be considered especially in multicenter and longitudinal studies.Here, we compare data generated by three different 3-Tesla magnetic resonance scanners (Philips Achieva; Siemens Verio; GE Signa MR750). We scanned two patients diagnosed with relapsing remitting MS six times per scanner within three weeks (T1w and FLAIR, 3D). We assessed T2-hyperintense lesions by an automated lesion segmentation tool and determined volumes of grey matter (GM), white matter (WM) and whole brain (GM+WM) from the lesion-filled T1-weighted images using voxel-based morphometry (SPM8/VBM8) and SIENAX (FSL). We measured cortical thickness using FreeSurfer from both, lesion-filled and original T1-weighted images. We quantified brain volume changes with SIENA.In both patients, we found significant differences in total lesion volume, global brain tissue volumes and cortical thickness measures between the scanners. Morphometric measures varied remarkably between repeated scans at each scanner, independent of the brain imaging software tool used.We conclude that for cross-sectional multicenter studies, the effect of different scanners has to be taken into account. For longitudinal monocentric studies, the expected effect size should exceed the size of false positive findings observed in this study. Assuming a physiological loss of brain volume of about 0.3% per year in healthy adult subjects (Good et al., 2001), which may double in MS (De Stefano et al., 2010; De Stefano et al., 2015), with current tools reliable estimation of brain atrophy in individual patients is only possible over periods of several years.

Feasibility of Diffusion Tractography for the Reconstruction of Intra-Thalamic and Cerebello-Thalamic Targets for Functional Neurosurgery: A Multi-Vendor Pilot Study in Four Subjects.

Functional stereotactic neurosurgery by means of deep brain stimulation or ablation provides an effective treatment for movement disorders, but the outcome of surgical interventions depends on the accuracy by which the target structures are reached. The purpose of this pilot study was to evaluate the feasibility of diffusion tensor imaging (DTI) based probabilistic tractography of deep brain structures that are commonly used for pre- and perioperative targeting for functional neurosurgery. Three targets were reconstructed based on their significance as intervention sites or as a no-go area to avoid adverse side effects: the connections propagating from the thalamus to (1) primary and supplementary motor areas, (2) to somatosensory areas and the cerebello-thalamic tract (CTT). We evaluated the overlap of the reconstructed connectivity based targets with corresponding atlas based data, and tested the inter-subject and inter-scanner variability by acquiring repeated DTI from four volunteers, and on three MRI scanners with similar sequence parameters. Compared to a 3D histological atlas of the human thalamus, moderate overlaps of 35-50% were measured between connectivity- and atlas based volumes, while the minimal distance between the centerpoints of atlas and connectivity targets was 2.5 mm. The variability caused by the MRI scanner was similar to the inter-subject variability, except for connections with the postcentral gyrus where it was higher. While CTT resolved the anatomically correct trajectory of the tract individually, high volumetric variability was found across subjects and between scanners. DTI can be applied in the clinical, preoperative setting to reconstruct the CTT and to localize subdivisions within the lateral thalamus. In our pilot study, such subdivisions moderately matched the borders of the ventrolateral-posteroventral (VLpv) nucleus and the ventral-posterolateral (VPL) nucleus. Limitations of the currently used standard DTI protocols were exacerbated by large scanner-to-scanner variability of the connectivity-based targets.

Subsolid pulmonary nodule morphology and associated patient characteristics in a routine clinical population

ObjectivesTo determine the presence and morphology of subsolid pulmonary nodules (SSNs) in a non-screening setting and relate them to clinical and patient characteristics.MethodsA total of 16,890 reports of clinically obtained chest CT (06/2011 to 11/2014, single-centre) were searched describing an SSN. Subjects with a visually confirmed SSN and at least two thin-slice CTs were included. Nodule volumes were measured. Progression was defined as volume increase exceeding the software interscan variation. Nodule morphology, location, and patient characteristics were evaluated.ResultsFifteen transient and 74 persistent SSNs were included (median follow-up 19.6 [8.3–36.8] months). Subjects with an SSN were slightly older than those without (62 vs. 58 years; p = 0.01), but no gender predilection was found. SSNs were mostly located in the upper lobes. Women showed significantly more often persistent lesions than men (94 % vs. 69 %; p = 0.002). Part-solid lesions were larger (1638 vs. 383 mm3; p < 0.001) and more often progressive (68 % vs. 38 %; p = 0.02), compared to pure ground-glass nodules. Progressive SSNs were rare under the age of 50 years. Logistic regression analysis did not identify additional nodule parameters of future progression, apart from part-solid nature.ConclusionsThis study confirms previously reported characteristics of SSNs and associated factors in a European, routine clinical population.Key Points• SSNs in women are significantly more often persistent compared to men.• SSN persistence is not associated with age or prior malignancy.• The majority of (persistent) SSNs are located in the upper lung lobes.• A part-solid nature is associated with future nodule growth.• Progressive solitary SSNs are rare under the age of 50 years.

SU-F-R-30: Interscanner Variability of Radiomics Features in Computed Tomography (CT) Using a Standard ACR Phantom

Purpose: A simple approach to investigate Interscanner variability of Radiomics features in computed tomography (CT) using a standard ACR phantom. Methods: The standard ACR phantom was scanned on CT scanners from three different manufacturers. Scanning parameters of 120 KVp, 200 mA were used while slice thickness of 3.0 mm on two scanners and 3.27 mm on third scanner was used. Three spherical regions of interest (ROI) from water, medium density and high density inserts were contoured. Ninety four Radiomics features were extracted using an in-house program. These features include shape (11), intensity (22), GLCM (26), GLZSM (11), RLM (11), and NGTDM (5) and 8 fractal dimensions features. To evaluate the Interscanner variability across three scanners, a coefficient of variation (COV) is calculated for each feature group. Each group is further classified according to the COV- by calculating the percentage of features in each of the following categories: COV less than 2%, between 2 and 10% and greater than 10%. Results: For all feature groups, similar trend was observed for three different inserts. Shape features were the most robust for all scanners as expected. 70% of the shape features had COV <2%. For intensity feature group, 2% COV varied from 9 to 32% for three scanners. All features in four groups GLCM, GLZSM, RLM and NGTDM were found to have Interscanner variability ≥2%. The fractal dimensions dependence for medium and high density inserts were similar while it was different for water inserts. Conclusion: We concluded that even for similar scanning conditions, Interscanner variability across different scanners was significant. The texture features based on GLCM, GLZSM, RLM and NGTDM are highly scanner dependent. Since the inserts of the ACR Phantom are not heterogeneous in HU values suggests that matrix based 2nd order features are highly affected by variation in noise. Research partly funded by NIH/NCI R01CA190105-01

MO-DE-207B-04: Impact of Reconstruction Field of View On Radiomics Features in Computed Tomography (CT) Using a Texture Phantom

Purpose: To investigate the impact of reconstruction Field of View on Radiomics features in computed tomography (CT) using a texture phantom. Methods: A rectangular Credence Cartridge Radiomics (CCR) phantom, composed of 10 different cartridges, was scanned on four different CT scanners from two manufacturers. A pre-defined scanning protocol was adopted for consistency. The slice thickness and reconstruction interval of 1.5 mm was used on all scanners. The reconstruction FOV was varied to result a voxel size ranging from 0.38 to 0.98 mm. A spherical region of interest (ROI) was contoured on the shredded rubber cartridge from CCR phantom CT scans. Ninety three Radiomics features were extracted from ROI using an in-house program. These include 10 shape, 22 intensity, 26 GLCM, 11 GLZSM, 11 RLM, 5 NGTDM and 8 fractal dimensional features. To evaluate the Interscanner variability across three scanners, a coefficient of variation (COV) was calculated for each feature group. Each group was further classified according to the COV by calculating the percentage of features in each of the following categories: COV≤ 5%, between 5 and 10% and ≥ 10%. Results: Shape features were the most robust, as expected, because of the spherical contouring of ROI. Intensity features were the second most robust with 54.5 to 64% of features with COV < 5%. GLCM features ranged from 31 to 35% for the same category. RLM features were sensitive to specific scanner and 5% variability was 9 to 54%. Almost all GLZM and NGTDM features showed COV ≥10% among the scanners. The dependence of fractal dimensions features on FOV was not consistent across different scanners. Conclusion: We concluded that reconstruction FOV greatly influence Radiomics features. The GLZSM and NGTDM are highly sensitive to FOV. funded in part by Grant NIH/NCI R01CA190105-01

The 68Ge phantom-based FDG-PET site qualification program for clinical trials adopted by FIL (Italian Foundation on Lymphoma)

PurposeThe quantitative assessment of Positron Emission Tomography (PET) scans using standardized uptake value and derived parameters proved to be superior to traditional qualitative assessment in several retrospective or mono-centric prospective reports. Since different scanners give different quantitative readings, a program for clinical trial qualification (CTQ) is mandatory to guarantee a reliable and reproducible use of quantitative PET in prospective multi-centre clinical trials and in every-day clinical life. MethodsWe set up, under the auspices of Italian Foundation on Lymphoma (FIL), a CTQ program consisting of the PET/CT scan acquisition and analysis of 18F and 68Ge NEMA/IEC image quality phantoms for the reduction of inter-scanner variability. Variability was estimated on background activity concentration (BAC) and sphere to background ratio (SBR). ResultsThe use of a 68Ge phantom allowed reducing the inter-scanner variability among different scanners from 74.0% to 20.5% in BAC and from 63.3% to 17.4% in SBR compared to using the 18F phantom. The CTQ criteria were fulfilled at first round in 100% and 28% of PET scanners with 68Ge and 18F respectively. ConclusionsThe 68Ge phantom proved a reliable tool for PET scanner qualification, able to significantly reduce the potential sources of error while increasing the reproducibility of PET derived quantitative parameter measurement.

The effect of late-phase contrast enhancement on semi-automatic software measurements of CT attenuation and volume of part-solid nodules in lung adenocarcinomas

ObjectivesTo evaluate the differences in semi-automatic measurements of CT attenuation and volume of part-solid nodules (PSNs) between unenhanced and enhanced CT scans. Materials and methodsCT scans including unenhanced and enhanced phases (slice thickness 0.625 and 1.25mm, respectively) for 53 adenocarcinomas presenting as PSNs in 50 patients were retrospectively evaluated. For each nodule, semi-automatic segmentation provided the diameter, mean attenuation, mass, and volume of a whole nodule and its solid component. Interscan variability and statistical significance of the differences in those measures according to the adenocarcinoma category were evaluated by one reader. ResultsAll parameters except for the mean attenuation of the solid components, were significantly increased on enhanced CT (p<0.05). For the whole nodule, the mean relative differences were as follows: the longest diameter, 1.4% (limits of agreement, −6.2–9.1); volume, 2.4% (−26.7–31.4); mass, 7.0% (−11.3–25.2); mean attenuation, 2.7% (−5.6–11). For the nodule’s solid component, those differences were as follow: the longest diameter, 6.9% (−34.4–48.2); volume, 17.9% (−77.8–113.7); mass, 18.8% (−77.8–115.4). The differences of measures between the unenhanced and enhanced CT were not significantly different between two groups of adenocarcinoma in situ/minimally invasive adenocarcinomas and invasive adenocarcinomas (p>0.05). ConclusionsAs most volumetric and attenuation measurements changed significantly after contrast enhancement, care should be taken in comparing unenhanced and enhanced CT in the evaluation of PSNs.

Toward automated classification of acetabular shape in ultrasound for diagnosis of DDH: Contour alpha angle and the rounding index

Background and objectivesThe diagnosis of Developmental Dysplasia of the Hip (DDH) in infants is currently made primarily by ultrasound. However, two-dimensional ultrasound (2DUS) images capture only an incomplete portion of the acetabular shape, and the alpha and beta angles measured on 2DUS for the Graf classification technique show high inter-scan and inter-observer variability. This variability relates partly to the manual determination of the apex point separating the acetabular roof from the ilium during index measurement. This study proposes a new 2DUS image processing technique for semi-automated tracing of the bony surface followed by automatic calculation of two indices: a contour-based alpha angle (αA), and a new modality-independent quantitative rounding index (M). The new index M is independent of the apex point, and can be directly extended to 3D surface models. MethodsWe tested the proposed indices on a dataset of 114 2DUS scans of infant hips aged between 4 and 183 days scanned using a 12MHz linear transducer. We calculated the manual alpha angle (αM), coverage, contour-based alpha angle and rounding index for each of the recordings and statistically evaluated these indices based on regression analysis, area under the receiver operating characteristic curve (AUC) and analysis of variance (ANOVA). ResultsProcessing time for calculating αA and M was similar to manual alpha angle measurement, ∼30s per image. Reliability of the new indices was high, with inter-observer intraclass correlation coefficients (ICC) 0.90 for αA and 0.89 for M. For a diagnostic test classifying hips as normal or dysplastic, AUC was 93.0% for αA vs. 92.7% for αM, 91.6% for M alone, and up to 95.7% for combination of M with αM, αA or coverage. ConclusionsThe rounding index provides complimentary information to conventional indices such as alpha angle and coverage. Calculation of the contour-based alpha angle and rounding index is rapid, shows potential to improve the reliability and accuracy of DDH diagnosis from 2DUS, and could be extended to 3D ultrasound in future.

Removing inter-subject technical variability in magnetic resonance imaging studies

Magnetic resonance imaging (MRI) intensities are acquired in arbitrary units, making scans non-comparable across sites and between subjects. Intensity normalization is a first step for the improvement of comparability of the images across subjects. However, we show that unwanted inter-scan variability associated with imaging site, scanner effect, and other technical artifacts is still present after standard intensity normalization in large multi-site neuroimaging studies. We propose RAVEL (Removal of Artificial Voxel Effect by Linear regression), a tool to remove residual technical variability after intensity normalization. As proposed by SVA and RUV [Leek and Storey, 2007, 2008, Gagnon-Bartsch and Speed, 2012], two batch effect correction tools largely used in genomics, we decompose the voxel intensities of images registered to a template into a biological component and an unwanted variation component. The unwanted variation component is estimated from a control region obtained from the cerebrospinal fluid (CSF), where intensities are known to be unassociated with disease status and other clinical covariates. We perform a singular value decomposition (SVD) of the control voxels to estimate factors of unwanted variation. We then estimate the unwanted factors using linear regression for every voxel of the brain and take the residuals as the RAVEL-corrected intensities. We assess the performance of RAVEL using T1-weighted (T1-w) images from more than 900 subjects with Alzheimer's disease (AD) and mild cognitive impairment (MCI), as well as healthy controls from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. We compare RAVEL to two intensity-normalization-only methods: histogram matching and White Stripe. We show that RAVEL performs best at improving the replicability of the brain regions that are empirically found to be most associated with AD, and that these regions are significantly more present in structures impacted by AD (hippocampus, amygdala, parahippocampal gyrus, enthorinal area, and fornix stria terminals). In addition, we show that the RAVEL-corrected intensities have the best performance in distinguishing between MCI subjects and healthy subjects using the mean hippocampal intensity (AUC=67%), a marked improvement compared to results from intensity normalization alone (AUC=63% and 59% for histogram matching and White Stripe, respectively). RAVEL is promising for many other imaging modalities.

An index for diagnosing infant hip dysplasia using 3-D ultrasound: the acetabular contact angle.

Developmental dysplasia of the hip (DDH) is a common condition that is highly treatable in infancy but can lead to the lifelong morbidity of premature osteoarthritis if left untreated. Current diagnostic methods lack reliability, which may be improved by using 3-D ultrasound. Conventional 2-D US assessment of DDH has limitations, including high inter-scan variability. We quantified DDH on 3-D US using the acetabular contact angle (ACA), a property of the 3-D acetabular shape. We assessed ACA reliability and diagnostic utility. We prospectively collected data from January 2013 to December 2014, including 114 hips in 85 children divided into three clinical diagnostic groups: (1) normal, (2) initially borderline but ultimately normal without treatment and (3) dysplastic requiring treatment. Using custom software, two observers each traced acetabula twice on two 3-D US scans of each hip, enabling automated generation of 3-D surface models and ACA calculation. We computed inter-observer and inter-scan variability of repeatability coefficients and generated receiver operating characteristic (ROC) curves. The 3-D US acetabular contact angle was reproduced 95% of the time within 6° in the same scan and within 9° in different scans of the same hip, vs. 9° and 14° for the 2-D US alpha angle (P < 0.001). Areas under ROC curves for diagnosis of developmental dysplasia of the hip were 0.954 for ACA and 0.927 for alpha angle. The 3-D US ACA was significantly more reliable than 2-D US alpha angle, and the 3-D US measurement predicted the presence of DDH with slightly higher accuracy. The ACA therefore shows promising initial diagnostic utility. Our findings call for further study of 3-D US in the diagnosis and longer-term follow-up of infant hip dysplasia.

A Novel GBM Saliency Detection Model Using Multi-Channel MRI.

The automatic computerized detection of regions of interest (ROI) is an important step in the process of medical image processing and analysis. The reasons are many, and include an increasing amount of available medical imaging data, existence of inter-observer and inter-scanner variability, and to improve the accuracy in automatic detection in order to assist doctors in diagnosing faster and on time. A novel algorithm, based on visual saliency, is developed here for the identification of tumor regions from MR images of the brain. The GBM saliency detection model is designed by taking cue from the concept of visual saliency in natural scenes. A visually salient region is typically rare in an image, and contains highly discriminating information, with attention getting immediately focused upon it. Although color is typically considered as the most important feature in a bottom-up saliency detection model, we circumvent this issue in the inherently gray scale MR framework. We develop a novel pseudo-coloring scheme, based on the three MRI sequences, viz. FLAIR, T2 and T1C (contrast enhanced with Gadolinium). A bottom-up strategy, based on a new pseudo-color distance and spatial distance between image patches, is defined for highlighting the salient regions in the image. This multi-channel representation of the image and saliency detection model help in automatically and quickly isolating the tumor region, for subsequent delineation, as is necessary in medical diagnosis. The effectiveness of the proposed model is evaluated on MRI of 80 subjects from the BRATS database in terms of the saliency map values. Using ground truth of the tumor regions for both high- and low- grade gliomas, the results are compared with four highly referred saliency detection models from literature. In all cases the AUC scores from the ROC analysis are found to be more than 0.999 ± 0.001 over different tumor grades, sizes and positions.

The use of routine non density calibrated clinical computed tomography data as a potentially useful screening tool for identifying patients with osteoporosis.

To evaluate whether lumbar vertebral body density CT attenuation values measured in Hounsfield Units (HUs) on routine Computed Tomography (CT) examinations can be reliably measured with limited variability, and to evaluate for a correlation between HUs and bone mineral density as measured by dual energy X-ray absorptiometry (DXA) scan. Retrospective review of a total of 249 routine MDCT examinations, performed to measure HUs at the first non-rib bearing lumbar vertebral body on axial images, cross-referenced to the lateral scout image. The overall ICC and RC for intra-reader variability on CT HU were 0.987 (95% CI 0.973 - 0.999) and 15.664 (95% CI 11.66-16.97). The overall ICC and RDC for inter-reader variability on CT HU were 0.952 (95% CI 0.892 - 0.999) and 30.20 (95% CI 23.73 - 34.48). The ICC and RC for interscanner variability were 0.98 (95% CI 0.95 - 0.99) and 16.67 (95% CI 13.13 - 22.85). The correlation between the L1 HUs and L1 BMD, L1 t-score, and overall t-score was 0.437, 0.392, and 0.400, respectively. CT attenuation values of the first lumbar vertebra can be measured on routine abdomen CTs with limited variability despite multiple readers and scanners. Correlation between HU and BMD as measured by DXA scan was only weakly positive, and by this method measuring the density of a lumbar vertebral body from a routine MDCT scan does not provide the sensitivity or specificity necessary for a screening test. However above a certain measured value (180 HU), patients have a low chance of osteoporosis and therefore may not need additional screening, potentially limiting radiation exposure and cost.

Diffusion-weighted magnetic resonance imaging in cancer: Reported apparent diffusion coefficients, in-vitro and in-vivo reproducibility.

There is considerable disparity in the published apparent diffusion coefficient (ADC) values across different anatomies. Institutions are increasingly assessing repeatability and reproducibility of the derived ADC to determine its variation, which could potentially be used as an indicator in determining tumour aggressiveness or assessing tumour response. In this manuscript, a review of selected articles published to date in healthy extra-cranial body diffusion-weighted magnetic resonance imaging is presented, detailing reported ADC values and discussing their variation across different studies. In total 115 studies were selected including 28 for liver parenchyma, 15 for kidney (renal parenchyma), 14 for spleen, 13 for pancreatic body, 6 for gallbladder, 13 for prostate, 13 for uterus (endometrium, myometrium, cervix) and 13 for fibroglandular breast tissue. Median ADC values in selected studies were found to be 1.28 × 10(-3) mm(2)/s in liver, 1.94 × 10(-3) mm(2)/s in kidney, 1.60 × 10(-3) mm(2)/s in pancreatic body, 0.85 × 10(-3) mm(2)/s in spleen, 2.73 × 10(-3) mm(2)/s in gallbladder, 1.64 × 10(-3) mm(2)/s and 1.31 × 10(-3) mm(2)/s in prostate peripheral zone and central gland respectively (combined median value of 1.54×10(-3) mm(2)/s), 1.44 × 10(-3) mm(2)/s in endometrium, 1.53 × 10(-3) mm(2)/s in myometrium, 1.71 × 10(-3) mm(2)/s in cervix and 1.92 × 10(-3) mm(2)/s in breast. In addition, six phantom studies and thirteen in vivo studies were summarized to compare repeatability and reproducibility of the measured ADC. All selected phantom studies demonstrated lower intra-scanner and inter-scanner variation compared to in vivo studies. Based on the findings of this manuscript, it is recommended that protocols need to be optimised for the body part studied and that system-induced variability must be established using a standardized phantom in any clinical study. Reproducibility of the measured ADC must also be assessed in a volunteer population, as variations are far more significant in vivo compared with phantom studies.

Atlas based brain volumetry: How to distinguish regional volume changes due to biological or physiological effects from inherent noise of the methodology

Fully-automated regional brain volumetry based on structural magnetic resonance imaging (MRI) plays an important role in quantitative neuroimaging. In clinical trials as well as in clinical routine multiple MRIs of individual patients at different time points need to be assessed longitudinally. Measures of inter- and intrascanner variability are crucial to understand the intrinsic variability of the method and to distinguish volume changes due to biological or physiological effects from inherent noise of the methodology.To measure regional brain volumes an atlas based volumetry (ABV) approach was deployed using a highly elastic registration framework and an anatomical atlas in a well-defined template space. We assessed inter- and intrascanner variability of the method in 51 cognitively normal subjects and 27 Alzheimer dementia (AD) patients from the Alzheimer's Disease Neuroimaging Initiative by studying volumetric results of repeated scans for 17 compartments and brain regions.Median percentage volume differences of scan–rescans from the same scanner ranged from 0.24% (whole brain parenchyma in healthy subjects) to 1.73% (occipital lobe white matter in AD), with generally higher differences in AD patients as compared to normal subjects (e.g., 1.01% vs. 0.78% for the hippocampus). Minimum percentage volume differences detectable with an error probability of 5% were in the one-digit percentage range for almost all structures investigated, with most of them being below 5%. Intrascanner variability was independent of magnetic field strength. The median interscanner variability was up to ten times higher than the intrascanner variability.

Detection and size measurements of pulmonary nodules in ultra-low-dose CT with iterative reconstruction compared to low dose CT

BackgroundIn this study, the accuracy of ultra-low-dose computed tomography (CT) with iterative reconstruction (IR) for detection and measurement of pulmonary nodules was evaluated. MethodsEighty-four individuals referred for lung cancer screening (mean age: 54.5±10.8 years) underwent low-dose computed tomography (LDCT) and ultra-low-dose CT. CT examinations were performed with attenuation-based tube current modulation. Reference tube voltage and current were set to 120kV/25mÅs for LDCT and 80kV/4mÅs for ultra-low-dose CT. CT images were reconstructed with filtered back projection (FBP) for LDCT, and with FBP and IR for ultra-low-dose CT datasets. A reference standard was established by a consensus panel of 2 different radiologists on LDCT. Volume and diameter of the solid nodules were measured on LDCT with FBP and ultra-low dose CT with FBP and IR. Interobserver and interscan variability were analyzed and compared by the Bland–Altman method. ResultsA total of 127 nodules were identified, including 105 solid nodules, 15 part solid nodules, 7 ground glass nodules. On ultra-low-dose CT scans, the effective radiation dose was 0.13±0.11mSv. A total of 113 (88.9%) and 110 (86.6%) true-positive nodules with FBP versus 117 (92.1%) and 118(92.9%) with IR were detected by two observers, respectively. The volume and size of the 105 solid nodules were measured, with mean volume/diameter of 46.5±46.6 mm3/5.1±1.6mm. There was no significant difference in nodule volume or diameter measurements between ultra-low-dose CT and LDCT protocols for solid nodules. ConclusionsUltra-low-dose CT with iterative reconstruction has high sensitivity for lung nodule detection without significant difference in nodule size and volume measurement compared to LDCT.

A framework for optimization of diffusion-weighted MRI protocols for large field-of-view abdominal-pelvic imaging in multicenter studies.

To develop methods for optimization of diffusion-weighted MRI (DW-MRI) in the abdomen and pelvis on 1.5 T MR scanners from three manufacturers and assess repeatability of apparent diffusion coefficient (ADC) estimates in a temperature-controlled phantom and abdominal and pelvic organs in healthy volunteers. Geometric distortion, ghosting, fat suppression, and repeatability and homogeneity of ADC estimates were assessed using phantoms and volunteers. Healthy volunteers (ten per scanner) were each scanned twice on the same scanner. One volunteer traveled to all three institutions in order to provide images for qualitative comparison. The common volunteer was excluded from quantitative analysis of the data from scanners 2 and 3 in order to ensure statistical independence, giving n = 10 on scanner 1 and n = 9 on scanners 2 and 3 for quantitative analysis. Repeatability and interscanner variation of ADC estimates in kidneys, liver, spleen, and uterus were assessed using within-patient coefficient of variation (wCV) and Kruskal-Wallis tests, respectively. The coefficient of variation of ADC estimates in the temperature-controlled phantom was 1%-4% for all scanners. Images of healthy volunteers from all scanners showed homogeneous fat suppression and no marked ghosting or geometric distortion. The wCV of ADC estimates was 2%-4% for kidneys, 3%-7% for liver, 6%-9% for spleen, and 7%-10% for uterus. ADC estimates in kidneys, spleen, and uterus showed no significant difference between scanners but a significant difference was observed in liver (p < 0.05). DW-MRI protocols can be optimized using simple phantom measurements to produce good quality images in the abdomen and pelvis at 1.5 T with repeatable quantitative measurements in a multicenter study.

Volumetric measurement of artificial pure ground-glass nodules at low-dose CT: Comparisons between hybrid iterative reconstruction and filtered back projection

PurposeTo compare hybrid iterative reconstruction (HIR) with filtered back projection (FBP) in the volumetry of artificial pure ground-glass nodules (GGNs) with low-dose computed tomography (CT). Materials and methodsArtificial GGNs (10mm-diameter, 523.6mm3, −660HU) in an anthropomorphic chest phantom were scanned by a 256-row multi-slice CT with three dose levels (10, 30, 100mAs). Each scan was repeated six times. Each set was reconstructed by FBP and HIR at 0.625-mm thickness. The volumes of artificial GGNs placed at the lung apex and middle lung field of the chest phantom were measured by two observers. Semi-automated measurements were performed by clicking the cursor in the center of GGNs, and manual measurements were performed by tracing GGNs on axial section. Modification of the trace was added on a sagittal or coronal section if necessary. Measurement errors were calculated for both the FBP and HIR at each dose level. We used the Wilcoxon signed rank test to identify any significant difference between the measurement errors of the FBP and HIR. Inter-observer, intra-observer, and inter-scan variabilities were evaluated by Bland Altman analysis with limits of agreements given by 95% confidence intervals. ResultsThere were significant differences in measurement errors only at the lung apex between FBP and HIR with 10mAs in both the semi-automated (observer 1, −37% vs. 7.2%; observer 2, −39% vs. 1.9%) and manual methods (observer 1, −29% vs. 7.5%; observer 2, −30% vs. 1.1%), respectively (P<0.05). HIR provided each variability equal to or less than one half of that of FBP at 10mAs in both methods. In the semi-automated method, the inter-observer and intra-observer variabilities obtained by HIR at 10mAs were −11% to 17% and −6.7% to 6.7%, whereas those for FBP at 10mAs were −29% to 30% and −38% to 20%, respectively. The inter-scan variability for FBP at 100mAs vs. HIR at 10mAs was −9.5% to 11%, and that for FBP at 100mAs vs. FBP at 10mAs was −73% to 32%. In the manual method, the inter-observer and intra-observer variabilities for HIR at 10mAs were −14% to 22% and −9.8% to 22%, and those for FBP at 10mAs were −45% to 36% and −31% to 28%, respectively. The inter-scan variability for FBP at 100mAs vs. HIR at 10mAs was −7.4% to 23%, and that for FBP at 100mAs vs. FBP at 10mAs was −52% to 26%. ConclusionHIR is superior to FBP in the volumetry of artificial pure GGNs at lung apex with low-dose CT.

A technique for semiautomatic segmentation of echogenic structures in 3D ultrasound, applied to infant hip dysplasia.

Automatic segmentation of anatomical structures and lesions from medical ultrasound images is a formidable challenge in medical imaging due to image noise, blur and artifacts. In this paper we present a segmentation technique with features highly suited to use in noisy 3D ultrasound volumes and demonstrate its use in modeling bone, specifically the acetabulum in infant hips. Quantification of the acetabular shape is crucial in diagnosing developmental dysplasia of the hip (DDH), a common condition associated with hip dislocation and premature osteoarthritis if not treated. The well-established Graf technique for DDH diagnosis has been criticized for high inter-observer and inter-scan variability. In our earlier work we have introduced a more reliable instability metric based on 3D ultrasound data. Visualizing and interpreting the acetabular shape from noisy 3D ultrasound volumes has been one of the major roadblocks in using 3D ultrasound as diagnostic tool for DDH. For this study we developed a semiautomated segmentation technique to rapidly generate 3D acetabular surface models and classified the acetabulum based on acetabular contact angle (ACA) derived from the models. We tested the feasibility and reliability of the technique compared with manual segmentation. The proposed segmentation algorithm is based on graph search. We formulate segmentation of the acetabulum as an optimal path finding problem on an undirected weighted graph. Slice contours are defined as the optimal path passing through a set of user-defined seed points in the graph, and it can be found using dynamic programming techniques (in this case Dijkstra's algorithm). Slice contours are then interpolated over the 3D volume to generate the surface model. A three-dimensional ACA was calculated using normal vectors of the surface model. The algorithm was tested over an extensive dataset of 51 infant ultrasound hip volumes obtained from 42 subjects with normal to dysplastic hips. The contours generated by the segmentation algorithm closely matched with those obtained from manual segmentation. The average RMS errors between the semiautomated and manual segmentation for the 51 volumes were 0.28 mm/1.1 voxel (with 2 node points) and 0.24 mm/0.9 voxel (with 3 node points). The semiautomatic algorithm gave visually acceptable results on images with moderate levels of noise and was able to trace the boundary of the acetabulum even in the presence of significant shadowing. Semiautomatic contouring was also faster than manual segmentation at 37 versus 56 s per scan. It also improved the repeatability of the ACA calculation with inter-observer and intra-observer variability of 1.4 ± 0.9 degree and 1.4 ± 1.0 degree. The semiautomatic segmentation technique proposed in this work offers a fast and reliable method to delineate the contours of the acetabulum from 3D ultrasound volumes of the hip. Since the technique does not rely upon contour evolution, it is less susceptible than other methods to the frequent missing or incomplete boundaries and noise artifacts common in ultrasound images. ACA derived from the segmented 3D surface was able to accurately classify the acetabulum under the categories normal, borderline and dysplastic. The semiautomatic technique makes it easier to segment the volume and reduces the inter-observer and intra-observer variation in ACA calculation compared with manual segmentation. The method can be applied to any structure with an echogenic boundary on ultrasound (such as a ventricle, blood vessel, organ or tumor), or even to structures with a bright border on computed tomography or magnetic resonance imaging.

SU-D-BRA-05: Toward Understanding the Robustness of Radiomics Features in CT

Purpose: To gauge the impact of inter-scanner variability on radiomics features in computed tomography (CT). Methods: We compared the radiomics features calculated for 17 scans of the specially designed Credence Cartridge Radiomics (CCR) phantom with those calculated for 20 scans of non–small cell lung cancer (NSCLC) tumors. The scans were acquired at four medical centers using General Electric, Philips, Siemens, and Toshiba CT scanners. Each center used its own routine thoracic imaging protocol. To produce a large dynamic range of radiomics feature values, the CCR phantom has 10 cartridges comprising different materials. The features studied were derived from the neighborhood gray-tone difference matrix or image intensity histogram. To quantify the significance of the inter-scanner variability, we introduced the metric “feature noise”, which compares the ratio of inter-scanner variability and inter-patient variability in decibels, positive values indicating substantial noise. We performed hierarchical clustering based to look for dependence of the features on the scan acquisition parameters. Results: For 5 of the 10 features studied, the inter-scanner variability was larger than the inter-patient variability. Of the 10 materials in the phantom, shredded rubber seemed to produce feature values most similar to those of the NSCLC tumors. The feature busyness had the greatest feature noise (14.3 dB), whereas texture strength had the least (−14.6 dB). Hierarchical clustering indicated that the features depended in part on the scanner manufacturer, image slice thickness, and pixel size. Conclusion: The variability in the values of radiomics features calculated for CT images of a radiomics phantom can be substantial relative to the variability in the values of these features calculated for CT images of NSCLC tumors. These inter-scanner differences and their effects should be carefully considered in future radiomics studies.