Detectability Of Low-contrast Lesions Research Articles

Low-contrast lesion detection in neck CT: a multireader study comparing deep learning, iterative, and filtered back projection reconstructions using realistic phantoms.

Computed tomography (CT) reconstruction algorithms can improve image quality, especially deep learning reconstruction (DLR). We compared DLR, iterative reconstruction (IR), and filtered back projection (FBP) for lesion detection in neck CT. Nine patient-mimicking neck phantoms were examined with a 320-slice scanner at six doses: 0.5, 1, 1.6, 2.1, 3.1, and 5.2 mGy. Each of eight phantoms contained one circular lesion (diameter 1 cm; contrast -30 HU to the background) in the parapharyngeal space; one phantom had no lesions. Reconstruction was made using FBP, IR, and DLR. Thirteen readers were tasked with identifying and localizing lesions in 32 images with a lesion and 20 without lesions for each dose and reconstruction algorithm. Receiver operating characteristic (ROC) and localization ROC (LROC) analysis were performed. DLR improved lesion detection with ROC area under the curve (AUC) 0.724 ± 0.023 (mean ± standard error of the mean) using DLR versus 0.696 ± 0.021 using IR (p = 0.037) and 0.671 ± 0.023 using FBP (p < 0.001). Likewise, DLR improved lesion localization, with LROC AUC 0.407 ± 0.039 versus 0.338 ± 0.041 using IR (p = 0.002) and 0.313 ± 0.044 using FBP (p < 0.001). Dose reduction to 0.5 mGy compromised lesion detection in FBP-reconstructed images compared to doses ≥ 2.1 mGy (p ≤ 0.024), while no effect was observed with DLR or IR (p ≥ 0.058). DLR improved the detectability of lesions in neck CT imaging. Dose reduction to 0.5 mGy maintained lesion detectability when denoising reconstruction was used. Deep learning enhances lesion detection in neck CT imaging compared to iterative reconstruction and filtered back projection, offering improved diagnostic performance and potential for x-ray dose reduction. Low-contrast lesion detectability was assessed in anatomically realistic neck CT phantoms. Deep learning reconstruction (DLR) outperformed filtered back projection and iterative reconstruction. Dose has little impact on lesion detectability against anatomical background structures.

Lesion detection in digital breast tomosynthesis: human reader experiments indicate no benefit from the integration of information from multiple planes.

In digital breast tomosynthesis (DBT), radiologists need to review a stack of 20 to 80 tomosynthesis images, depending upon breast size. This causes a significant increase in reading time. However, it is currently unknown whether there is a perceptual benefit to viewing a mass in the 3D tomosynthesis volume. To answer this question, this study investigated whether adjacent lesion-containing planes provide additional information that aids lesion detection for DBT-like and breast CT-like (bCT) images. Human reader detection performance was determined for low-contrast targets shown in a single tomosynthesis image at the center of the target (2D) or shown in the entire tomosynthesis image stack (3D). Using simulations, targets embedded in simulated breast backgrounds, and images were generated using a DBT-like (50deg angular range) and a bCT-like (180deg angular range) imaging geometry. Experiments were conducted with spherical and capsule-shaped targets. Eleven readers reviewed 1600 images in two-alternative forced-choice experiments. The area under the receiver operating characteristic curve (AUC) and reading time were computed for the 2D and 3D reading modes for the DBT and bCT imaging geometries and for both target shapes. Spherical lesion detection was higher in 2D mode than in 3D, for both DBT- and bCT-like images (DBT: , , ; bCT: , , ), but equivalent for capsule-shaped signals (DBT: , , ; bCT: , , ). Average reading time was up to 134% higher for 3D viewing (). For the detection of low-contrast lesions, there is no inherent visual perception benefit to reviewing the entire DBT or bCT stack. The findings of this study could have implications for the development of 2D synthetic mammograms: a single synthesized 2D image designed to include all lesions present in the volume might allow readers to maintain detection performance at a significantly reduced reading time.

Task-Driven CT Image Quality Optimization for Low-Contrast Lesion Detectability with Tunable Neural Networks.

Low-contrast lesions are difficult to detect in noisy low-dose CT images. Improving CT image quality for this detection task has the potential to improve diagnostic accuracy and patient outcomes. In this work, we use tunable neural networks for CT image restoration with a hyperparameter to control the variance/bias tradeoff. We use clinical images from a super-high-resolution normal-dose CT scan to synthesize low-contrast low-dose CT images for supervised training of deep learning CT reconstruction models. Those models are trained using with multiple noise realizations so that variance and bias can be penalized separately. We use a training loss function with one hyperparameter called the denoising level, which controls the variance/bias tradeoff. Finally, we evaluate the CT image quality to find the optimal denoising level for low-contrast lesion detectability. We evaluate performance using a shallow neural network model classification model to represent a suboptimal image observer. Our results indicate that the optimal networks for low-contrast lesion detectability are those that prioritize bias reduction rather than mean-squared error, which demonstrates the potential clinical benefit of our proposed tunable neural networks.

Decorrelated compounding methods in synthetic transmit aperture ultrasound imaging and its application

Speckles exist in most ultrasound images. They are generated due to the interference of the ultrasound waves scattered from multiple scattering particles in one resolution cell. Although they are very useful in many ultrasound applications, they limit the detection of low-contrast targets in a scattering medium. In conventional compounding, multiple correlated sub-images from various imaging apertures and frequency bandwidths are generated and then averaged incoherently to reduce the speckles. In this paper, we present our study on a method called Decorrelated Compounding to reduce speckles. In decorrelated compounding, a decorrelation procedure was applied to the correlated sub-images to further reduce speckle variance in synthetic transmit aperture (STA) ultrasound imaging. Decorrelated compounding was shown to improve the detectability of low-contrast lesions in terms of lesion signal-to-noise ratio (lSNR), visual detection, and statistical tests of the performance in detecting low-contrast lesions. The application of the proposed method to monitoring ultrasound thermal therapy and to measuring the attenuation coefficient will also be discussed.

Performance of a deep learning‐based CT image denoising method: Generalizability over dose, reconstruction kernel, and slice thickness

Deep learning (DL) is rapidly finding applications in low-dose CT image denoising. While having the potential to improve the image quality (IQ) over the filtered back projection method (FBP) and produce images quickly, performance generalizability of the data-driven DL methods is not fully understood yet. The main purpose of this work is to investigate the performance generalizability of a low-dose CT image denoising neural network in data acquired under different scan conditions, particularly relating to these three parameters: reconstruction kernel, slice thickness, and dose (noise) level. A secondary goal is to identify any underlying data property associated with the CT scan settings that might help predict the generalizability of the denoising network. We select the residual encoder-decoder convolutional neural network (REDCNN) as an example of a low-dose CT image denoising technique in this work. To study how the network generalizes on the three imaging parameters, we grouped the CT volumes in the Low-Dose Grand Challenge (LDGC) data into three pairs of training datasets according to their imaging parameters, changing only one parameter in each pair. We trained REDCNN with them to obtain six denoising models. We test each denoising model on datasets of matching and mismatching parameters with respect to its training sets regarding dose, reconstruction kernel, and slice thickness, respectively, to evaluate the denoising performance changes. Denoising performances are evaluated on patient scans, simulated phantom scans, and physical phantom scans using IQ metrics including mean-squared error (MSE), contrast-dependent modulation transfer function (MTF), pixel-level noise power spectrum (pNPS), and low-contrast lesion detectability (LCD). REDCNN had larger MSE when the testing data were different from the training data in reconstruction kernel, but no significant MSE difference when varying slice thickness in the testing data. REDCNN trained with quarter-dose data had slightly worse MSE in denoising higher-dose images than that trained with mixed-dose data (17%-80%). The MTF tests showed that REDCNN trained with the two reconstruction kernels and slice thicknesses yielded images of similar image resolution. However, REDCNN trained with mixed-dose data preserved the low-contrast resolution better compared to REDCNN trained with quarter-dose data. In the pNPS test, it was found that REDCNN trained with smooth-kernel data could not remove high-frequency noise in the test data of sharp kernel, possibly because the lack of high-frequency noise in the smooth-kernel data limited the ability of the trained model in removing high-frequency noise. Finally, in the LCD test, REDCNN improved the lesion detectability over the original FBP images regardless of whether the training and testing data had matching reconstruction kernels. REDCNN is observed to be poorly generalizable between reconstruction kernels, more robust in denoising data of arbitrary dose levels when trained with mixed-dose data, and not highly sensitive to slice thickness. It is known that reconstruction kernel affects the in-plane pNPS shape of a CT image, whereas slice thickness and dose level do not, so it is possible that the generalizability performance of this CT image denoising network highly correlates to the pNPS similarity between the testing and training data.

Assessment of task-based image quality for abdominal CT protocols linked with national diagnostic reference levels

ObjectivesTo assess task-based image quality for two abdominal protocols on various CT scanners. To establish a relationship between diagnostic reference levels (DRLs) and task-based image quality.MethodsA protocol for the detection of focal liver lesions was used to scan an anthropomorphic abdominal phantom containing 8- and 5-mm low-contrast (20 HU) spheres at five CTDIvol levels (4, 8, 12, 16, and 20 mGy) on 12 CTs. Another phantom with high-contrast calcium targets (200 HU) was scanned at 2, 4, 6, 10, and 15 mGy using a renal stones protocol on the same CTs. To assess the detectability, a channelized Hotelling observer was used for low-contrast targets and a non-prewhitening observer with an eye filter was used for high contrast targets. The area under the ROC curve and signal to noise ratio were used as figures of merit.ResultsFor the detection of 8-mm spheres, the image quality reached a high level (mean AUC over all CTs higher than 0.95) at 11 mGy. For the detection of 5-mm spheres, the AUC never reached a high level of image quality. Variability between CTs was found, especially at low dose levels. For the search of renal stones, the AUC was nearly maximal even for the lowest dose level.ConclusionsComparable task-based image quality cannot be reached at the same dose level on all CT scanners. This variability implies the need for scanner-specific dose optimization.Key Points• There is an image quality variability for subtle low-contrast lesion detection in the clinically used dose range.• Diagnostic reference levels were linked with task-based image quality metrics.• There is a need for specific dose optimization for each CT scanner and clinical protocol.

Image texture, low contrast liver lesion detectability and impact on dose: Deep learning algorithm compared to partial model-based iterative reconstruction

ObjectivesTo compare deep learning (True Fidelity, TF) and partial model based Iterative Reconstruction (ASiR-V) algorithm for image texture, low contrast lesion detectability and potential dose reduction. MethodsAnthropomorphic phantoms (mimicking non-overweight and overweight patient), containing lesions of 6 mm in diameter with 20HU contrast, were scanned at five different dose levels (2,6,10,15,20 mGy) on a CT system, using clinical routine protocols for liver lesion detection. Images were reconstructed using ASiR-V 0% (surrogate for FBP), 60 % and TF at low, medium and high strength. Noise texture was characterized by computing a normalized Noise Power Spectrum filtered by an eye filter. The similarity against FBP texture was evaluated using peak frequency difference (PFD) and root mean square deviation (RMSD). Low contrast detectability was assessed using a channelized Hotelling observer and the area under the ROC curve (AUC) was used as figure of merit. Potential dose reduction was calculated to obtain the same AUC for TF and ASiR-V. ResultsFBP-like noise texture was more preserved with TF (PFD from -0.043mm-1 to -0.09mm-1, RMSD from 0.12mm-1 to 0.21mm-1) than with ASiR-V (PFD equal to 0.12 mm-1, RMSD equal to 0.53mm-1), resulting in a sharper image. AUC was always higher with TF than ASIR-V. In average, TF compared to ASiR-V, enabled a radiation dose reduction potential of 7%, 25 % and 33 % for low, medium and high strength respectively. ConclusionCompared to ASIR-V, TF at high strength does not impact noise texture and maintains low contrast liver lesions detectability at significant lower dose.

A deep learning- and partial least square regression-based model observer for a low-contrast lesion detection task in CT.

This work aims to develop a new framework of image quality assessment using deep learning-based model observer (DL-MO) and to validate it in a low-contrast lesion detection task that involves CT images with patient anatomical background. The DL-MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS-DA) model and an internal noise component. The CNN was previously trained to achieve the state-of-the-art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL-MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL-MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11mm) and three contrast levels (15, 20, and 25HU). The lesions were inserted into patient liver images using a projection-based method. A validated noise insertion tool was used to synthesize CT exams with 50% and 25% of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two-alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL-MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL-MO and HO. A statistically significant positive correlation was observed between DL-MO and HO for the 2AFC low-contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95% confidence interval (0.950, 0.996)]. Bland-Altman agreement analysis did not indicate statistically significant differences. The proposed DL-MO is highly correlated with HO in a low-contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL-MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.

Statistical weights for model-based reconstruction in cone-beam CT with electronic noise and dual-gain detector readout

Cone-beam CT (CBCT) systems commonly incorporate a flat-panel detector (FPD) with multiple-gain readout capability to reduce electronic noise and extend dynamic range. In this work, we report a penalized weighted least-squares (PWLS) method for CBCT image reconstruction with a system model that includes the electronic noise characteristics of FPDs, including systems with dynamic-gain or dual-gain (DG) readout in which the electronic noise is spatially varying. Statistical weights in PWLS were modified to account for the contribution of the electronic noise (algorithm denoted ), and the method was combined with a certainty-based approach that improves the homogeneity of spatial resolution (algorithm denoted ). The methods were tested in phantom studies designed to stress DG readout characteristics and translated to a clinical study for CBCT of patients with head traumas. The method demonstrated superior noise-resolution tradeoffs compared to filtered back-projection (FBP) and conventional PWLS. For example, with spatial resolution (edge-spread function width) matched at 0.65 mm, reduced variance by 28%–39% and 15%–25% compared to FBP and PWLS, respectively. The method achieved more homogeneous spatial resolution than while maintaining similar variance reduction. These findings were confirmed in clinical studies, which showed ~20% variance reduction in peripheral regions of the brain, potentially improving visual image quality in detection of epidural and/or subdural intracranial hemorrhage. The results are consistent with the general notion that incorporating a more accurate system model improves performance in optimization-based statistical CBCT reconstruction—in this case, a more accurate model for (spatially varying) electronic noise to improve detectability of low-contrast lesions.

Correlation between model observers in uniform background and human observers in patient liver background for a low-contrast detection task in CT.

Channelized Hotelling observer (CHO) has demonstrated strong correlation with human observer (HO) in both single-slice viewing mode and multi-slice viewing mode in low-contrast detection tasks with uniform background. However, it remains unknown if the simplest single-slice CHO in uniform background can be used to predict human observer performance in more realistic tasks that involve patient anatomical background and multi-slice viewing mode. In this study, we aim to investigate the correlation between CHO in a uniform water background and human observer performance at a multi-slice viewing mode on patient liver background for a low-contrast lesion detection task. The human observer study was performed on CT images from 7 abdominal CT exams. A noise insertion tool was employed to synthesize CT scans at two additional dose levels. A validated lesion insertion tool was used to numerically insert metastatic liver lesions of various sizes and contrasts into both phantom and patient images. We selected 12 conditions out of 72 possible experimental conditions to evaluate the correlation at various radiation doses, lesion sizes, lesion contrasts and reconstruction algorithms. CHO with both single and multi-slice viewing modes were strongly correlated with HO. The corresponding Pearson's correlation coefficient was 0.982 (with 95% confidence interval (CI) [0.936, 0.995]) and 0.989 (with 95% CI of [0.960, 0.997]) in multi-slice and single-slice viewing modes, respectively. Therefore, this study demonstrated the potential to use the simplest single-slice CHO to assess image quality for more realistic clinically relevant CT detection tasks.

A Study of Quasi-Monochromatic X-Ray Sources for Breast Cancer Early Detection

Objectives: Early detection and treatment provide the opportunity to decrease the mortality rate from breast cancer. Quasi-monochromatic technique can enhance low contrast lesion detection by eliminating beam-hardening artifacts. We hypothesized that this technique would be feasible and can be used to assist in breast cancer early detection. Methods: The performance of stationary Digital Breast Tomosynthesis with quasi-monochromatic X-ray sources was evaluated using both simulated and physical phantoms. Normalized spectra in the fraction of total photons vs. photon energy were generated. Results: As expected, the peaked energies from sources, from smallest to largest, are Mo/Mo, W/Ag-1000, and W/Ce-10, W/Ce-100. For contrast vs. noise standard deviation on the simulated CIRS phantom, W/Ce-100 and W/Ce-10 have similar performance on both low and high contrast objects. For low contrast object, W/Ce-100 is barely noticeably better than W/Ce-10, and they are better than both W/Ag-1000 and Mo/Mo. For high contrast objects W/Ce-10 is slightly better than W/Ce-100. The spectra of the implemented W/Ce-10 X-ray source were measured, which matched the simulation well. The contrast noise ratios of reconstructed objects in American College of Radiology mammographic phantom with and without using W/Ce-10 combination are 7.1 and 5.4, respectively. Conclusions: The combination of stationary digital breast tomosynthesis and quasi-monochromatic technique can compensate the loss of X-ray flux due to heavy K-edge filtering. This technique can enable the stationary DBT scanners to operate at acceptable scanning times with better low contrast lesion detectability. Advances in Knowledge: The stationary digital breast tomosynthesis can provide high quality images within short scanning time by using X-ray source array, which makes quasi-monochromatic technique feasible.

DİJİTAL RADYOGRAFİ SİSTEMLERİNDE GÜRÜLTÜNÜN KONUM VE FREKANS UZAYINDA İRDELENMESİ

Noise amount in digital radiographic image puts a limitation on detection of low contrast lesions. The aim of this study is to investigate of the noise component in digital radiographic image in both spatial and frequency domain. Noise analysis in spatial and frequency domain was performed by using relative noise variance and noise power spectra measurement methods, respectively.

Erratum to: Impact of model-based iterative reconstruction on low-contrast lesion detection and image quality in abdominal CT: a 12-reader-based comparative phantom study with filtered back projection at different tube voltages.

Erratum to: Impact of model-based iterative reconstruction on low-contrast lesion detection and image quality in abdominal CT: a 12-reader-based comparative phantom study with filtered back projection at different tube voltages.

Modified ideal observer model (MIOM) for high-contrast and high-spatial resolution CT imaging tasks.

Although a variety of mathematical observer models have been developed to predict human observer performance for low contrast lesion detection tasks, their predictive power for high contrast and high spatial resolution discrimination imaging tasks, including those in CT bone imaging, could be limited. The purpose of this work was to develop a modified observer model that has improved correlation with human observer performance for these tasks. The proposed observer model, referred to as the modified ideal observer model (MIOM), uses a weight function to penalize components in the task function that have less contribution to the actual human observer performance for high contrast and high spatial resolution discrimination tasks. To validate MIOM, both human observer and observer model studies were performed, each using exactly the same CT imaging task [discrimination of a connected component in a high contrast (1000 HU) high spatial resolution bone fracture model (0.3 mm)] and experimental CT image data. For the human observer studies, three physicist observers rated the connectivity of the fracture model using a five-point Likert scale; for the observer model studies, a total of five observer models, including both conventional models and the proposed MIOM, were used to calculate the discrimination capability of the CT images in resolving the connected component. Images used in the studies encompassed nine different reconstruction kernels. Correlation between human and observer model performance for these kernels were quantified using the Spearman rank correlation coefficient (ρ). After the validation study, an example application of MIOM was presented, in which the observer model was used to select the optimal reconstruction kernel for a High-Resolution (Hi-Res, GE Healthcare) CT scan technique. The performance of the proposed MIOM correlated well with that of the human observers with a Spearman rank correlation coefficient ρ of 0.88 (P = 0.003). In comparison, the value of ρ was 0.05 (P = 0.904) for the ideal observer, 0.05 (P = 0.904) for the non-prewhitening observer, -0.18 (P = 0.634) for the non-prewhitening observer with eye filter and internal noise, and 0.30 (P = 0.427) for the prewhitening observer with eye filter and internal noise. Using the validated MIOM, the optimal reconstruction kernel for the Hi-Res mode to perform high spatial resolution and high contrast discrimination imaging tasks was determined to be the HD Ultra kernel at the center of the scan field of view (SFOV), or the Lung kernel at the peripheral region of the SFOV. This result was consistent with visual observations of nasal CT images of an in vivo canine subject. Compared with other observer models, the proposed modified ideal observer model provides significantly improved correlation with human observers for high contrast and high spatial resolution CT imaging tasks.

Diagnosis of small posterior fossa stroke on brain CT: effect of iterative reconstruction designed for brain CT on detection performance

In this study, we aimed to determine whether iterative model reconstruction designed for brain CT (IMR-neuro) would improve the accuracy of posterior fossa stroke diagnosis on brain CT. We enrolled 37 patients with ischaemic stroke in the posterior fossa and 37 patients without stroke (controls). Using axial images reconstructed using filtered back-projection (FBP) and IMR-neuro, we compared the CT numbers in infarcted areas, image noise in the pons, and contrast-to-noise ratios (CNRs) of infarcted and non-infarcted areas on scans subjected to IMR-neuro and FBP. To analyse the performance of hypo-attenuation detection, we used receiver-operating characteristic (ROC) curve techniques. The image noise was significantly lower (2.2 ± 0.5 vs. 5.1 ± 0.9 Hounsfield units, p < 0.01) and the difference in CNR between the infarcted and non-infarcted areas was significantly higher with IMR-neuro than with FBP (2.2 ± 1.7 vs. 4.0 ± 3.6, p < 0.01). Furthermore, the average area under the ROC curve was significantly higher with IMR-neuro (0.90 vs. 0.86 for FBP, p = 0.04). IMR-neuro yielded better image quality and improved hypo-attenuation detection in patients with ischaemic stroke. • Iterative model reconstruction of brain CT data can facilitate the diagnosis of ischaemic stroke. • IMR improved the detectability of low-contrast lesions in the posterior fossa. • IMR-neuro yielded better image quality and improved observer performance.

Diagnostic performance of reduced-dose CT with a hybrid iterative reconstruction algorithm for the detection of hypervascular liver lesions: a phantom study.

To investigate the diagnostic performance of reduced-dose CT with a hybrid iterative reconstruction (IR) algorithm for the detection of hypervascular liver lesions. Thirty liver phantoms with or without simulated hypervascular lesions were scanned with a 320-slice CT scanner with control-dose (40 mAs) and reduced-dose (30 and 20 mAs) settings. Control-dose images were reconstructed with filtered back projection (FBP), and reduced-dose images were reconstructed with FBP and a hybrid IR algorithm. Objective image noise and the lesion to liver contrast-to-noise ratio (CNR) were evaluated quantitatively. Images were interpreted independently by 2 blinded radiologists, and jackknife alternative free-response receiver-operating characteristic (JAFROC) analysis was performed. Hybrid IR images with reduced-dose settings (both 30 and 20 mAs) yielded significantly lower objective image noise and higher CNR than control-dose FBP images (P < .05). However, hybrid IR images with reduced-dose settings had lower JAFROC1 figure of merit than control-dose FBP images, although only the difference between 20 mAs images and control-dose FBP images was significant for both readers (P < .01). An aggressive reduction of the radiation dose would impair the detectability of hypervascular liver lesions, although objective image noise and CNR would be preserved by a hybrid IR algorithm. • A half-dose scan with a hybrid iterative reconstruction preserves objective image quality. • A hybrid iterative reconstruction algorithm does not improve diagnostic performance. • An aggressive dose reduction would impair the detectability of low-contrast lesions.

TH-AB-207A-09: Tailoring TCM Schemes to a Task: Evaluating the Impact of Customized TCM Profiles On Detection of Lung Nodules in Simulated CT Lung Cancer Screening

Purpose: Recent work has shown that current TCM profile designs boost detection of low-contrast lung lesions in the lung apices, but yield reduced detection performance in the mid and lower lung regions relative to fixed tube current cases. This observed imbalance suggests that the TCM scheme might be tailored in new ways to maximize nodule detection throughout the entire lung. In this work, we begin a preliminary investigation into custom TCM profiles in an attempt to achieve uniform lesion detection throughout the extent of the lung. Methods: Low-contrast (25HU), 6mm nodules representing ground glass opacities were simulated at 1mm intervals over the length the lungs in a voxelized model of the XCAT phantom, one nodule per lung, per simulated scan. Voxel values represented attenuation values at 80keV. CT projection data was created by simulating a finite focal spot and using Joseph's method for forward projection; scanner geometry was that of the Siemens Sensation 64 and the X-ray source was simulated as an 80keV monochromatic beam. Noise realizations were created using Poisson statistics, a realistic bowtie filter and varying tube current. 500 noise realizations were created for the custom TCM designs. All reconstruction was done with FreeCT_wFBP. An SKE/BKE task was used in conjunction with a 2D Hotelling Observer to calculate area-under-the-curve (AUC) as a proxy for “detectability.” AUC was plotted as a function of nodule Z-location to create a “detectability map.” The detectability map for the custom TCM curve was qualitatively assessed relative to previous results for the fixed TC and clinical TCM cases for uniformity. Results: Detection uniformity was improved throughout the mid and lower lungs, however detection remained disproportionately high in the upper lung region. Conclusion: Detection uniformity was improved with a custom TC profile. Future work will incorporate an analytic, task-specific approach to optimize the TC scheme for nodule detection. J. Hoffman: Part-time intern, Toshiba Medical Research Institute; M. McNitt-Gray: Institutional research agreement, Siemens Healthcare; Past recipient, research grant support, Siemens Healthcare; Consultant, Toshiba America Medical Systems; Consultant, Samsung Electronics; F. Noo: Institutional research agreement, Siemens Healthcare; Receives research funding from Siemens Healthcare

The Detectability of Iterative CT Reconstruction for Low-contrast Lesions in Hyperacute Cerebral Infarction: Assessment with Newly Developed Dedicated Head Phantoms.

Iterative reconstruction techniques, such as adaptive statistical iterative reconstruction (ASiR), improve the contrast-to-noise ratio of computed tomography (CT) images; however, underlying anatomical structures may nevertheless hamper detectability of low-contrast areas in clinical situations, despite using such a technique. We therefore conducted a phantom study to investigate the efficacy of ASiR in improving the detectability of low-contrast areas in the presence of brain anatomical structures. We developed dedicated head phantoms simulating hyperacute cerebral infarction and confirmed that their CT numbers were sufficiently reproducible and that observer performance in detecting low-contrast areas using these phantoms more closely resembled that in clinical situations than that using a simple phantom. The efficacy of ASiR in improving low-contrast detectability was evaluated via receiver operating characteristics analysis. The mean area under the curve (AUC) values at ASiR blend rates of 0%, 30%, 60%, and 100% were 0.57, 0.57, 0.59, and 0.59 at 200 mA; 0.83, 0.84, 0.84, and 0.90 at 500 mA; and 0.79, 0.77, 0.76, and 0.79 at 800 mA, respectively. No significant differences were noted in AUC values among ASiR blend rates at any mA setting, suggesting that ASiR does not improve the detectability of subtle low-contrast lesions seen in hyperacute cerebral infarction in clinical situations.

SU-F-207-11: On the Relationship of Minimum Required Dose to Low Contrast Detectability and Blending Fractions With Adaptive Statistical Iterative Reconstruction (ASIR) in Abdominal CT

Purpose: It is of practical use to obtain the dose for a specified low contrast lesion detectability and a given ASIR fraction before the scan. To achieve the goal, we investigated the relationship of the minimum required dose to low contrast detectability and ASIR blending fraction in abdominal CT scans. Methods: A realistically shaped abdomen phantom (CIRS TE-07) with an embedded vertebral bone rod was utilized. Helical CT scans were performed on a GE 750 HD at two dose levels of 8 mGy and 18 mGy. The images were reconstructed using different ASIR fractions (0–100%). The noise images, obtained using subtraction from adjacent slices, were partitioned into different scales using matrices of square cells sized differently. The noise at each cell size can be quantified by the standard deviation (sd) of the cell mean pixel values distribution. Upon verifying that the distributions follow Gaussian, we proposed that the statistically defined minimum detectable contrast (MDC) can be applied to ASIR reconstructed images, i.e., MDC = 3.29sd. We then defined the normalized MDC (N_MDC) as the ratio of the MDC with ASIR to the MDC with filtered backprojection and postulated that N_MDC is dose independent. N_MDC was obtained at each cell size from different ASIR blending fractions at different dose levels. Results: The matrix cell means distribution was verified following Gaussian. N_MDC was found to linearly decrease as the ASIR blending fraction increases. The relationship is the same with two distinct doses. For the specified MDC at lesion size d and ASIR blending fraction (s), the minimum required dose D = [14.05(1-bs)/(MDC*d^1.12)]^2.03, where b = 0.005/d^0.784. Conclusion: The relationship of the minimum required dose to low contrast detectability and ASIR blending fraction was obtained. It can be used to adjust the scan parameters corresponding to the calculated minimum required dose.

Influence of the adaptive iterative dose reduction 3D algorithm on the detectability of low-contrast lesions and radiation dose repeatability in abdominal computed tomography: a phantom study.

The purpose of the study is to evaluate the influence of the adaptive iterative dose reduction (AIDR 3D) algorithm on the detectability of low-contrast focal liver lesions (FLLs) and the radiation dose repeatability of automatic tube current modulation (ATCM) in abdominal CT scans using anthropomorphic phantoms. Three different sizes of anthropomorphic phantoms, each with 4 low-contrast FLLs, were scanned on a 320-channel CT scanner using the ATCM technique and AIDR 3D, at different radiation doses: full-dose, half-dose, and quarter-dose. Scans were repeated three times and reconstructed with filtered back projection (FBP) and AIDR 3D. Radiation dose repeatability was assessed using the intraclass correlation coefficient (ICC). Image noise, quality, and lesion conspicuity were assessed by four reviewers and the number of invisible FLLs was compared among different radiation doses and reconstruction methods. ICCs of radiation dose among the three CT scans were excellent in all phantoms (0.99). Image noise, quality, and lesion conspicuity in the half-dose group were comparable with full-dose FBP after applying AIDR 3D in all phantoms. In small phantoms, the half-dose group reconstructed with AIDR 3D showed similar sensitivity in visualizing low-contrast FLLs compared to full-dose FBP (P = 0.77-0.84). In medium and large phantoms, AIDR 3D reduced the number of missing low-contrast FLLs [3.1% (9/288), 11.5% (33/288), respectively], compared to FBP [10.4% (30/288), 21.9% (63/288), respectively] in the full-dose group. By applying AIDR 3D, half-dose CT scans may be achievable in small-sized patients without hampering diagnostic performance, while it may improve diagnostic performance in medium- and large-sized patients without increasing the radiation dose.