McCollough, Cynthia

Abstract Background To utilize high spatial resolution reconstructions for cardiac imaging at energy‐integrating detector CT (EID)‐CT with comparable noise to similar reconstructions at photon‐counting detector (PCD)‐CT, methods to control EID‐CT image noise are needed. Supervised convolutional neural networks (CNN) have shown promise for denoising, but a challenge remains to efficiently create high‐quality and unbiased estimates of noise without access to dedicated software or proprietary information, such that natural noise texture is retained in CNN‐denoised CT images. Purpose This study aims to develop and test image‐based noise estimation methods that can be used to train a CNN model, and to evaluate denoising performance and noise texture preservation for EID‐CT coronary CT angiography (cCTA) images reconstructed with high‐resolution kernels. Methods U‐net CNN models were trained for denoising. To supervise training, noise‐only images were estimated directly from high‐resolution kernel (Bv59) reconstructed EID‐CT (HR EID‐CT) patient images using two different methods: subtraction of low‐ and high‐strength iterative reconstruction (IR); subtraction of adjacent image slices with the same IR strength. The noise estimates from these methods contain differing noise texture and anatomical information. Networks were trained and validated separately for three data sets: the training data from each of the two noise‐estimation methods, and a 50%–50% partition of training data between the two methods. The trained models were applied to two sets of testing data: CT images of a uniform water phantom to measure noise power spectra (NPS), and an independent cohort of seven patient cCTA HR EID‐CT exams. The denoised patient images were compared to standard resolution EID‐CT reconstructions (Bv40). As a low‐noise reference, patient images acquired on the same day with a PCD‐CT and reconstructed using a similar kernel as HR EID‐CT were used for comparison. Results Models trained with each noise‐image estimation method denoised the HR EID‐CT images by 74%–79% to achieve a comparable noise magnitude to the HR PCD‐CT images. The peak, average, and 10% peak frequencies of the NPS of the input images (6.08, 6.24, and 12.0 cm −1 ) were better approximated by the model trained on adjacent slice subtraction (6.56, 5.87, and 11.5 cm −1 ) than by the model trained on subtraction of low‐ and high‐IR images (4.64, 5.44, and 11.3 cm −1 ). In cCTA images, the IR subtraction model images retained anatomic structures from input images but resulted in undesirable salt‐and‐pepper noise texture and CT number bias. The model trained on adjacent slice subtraction images had more natural texture and no significant bias, but the model sometimes removed small anatomic structures. The model trained on the mixed training dataset preserved both noise texture and anatomy from the model inputs and enabled visualization of small structures seen in PCD‐CT images that were previously unresolved by EID‐CT. Conclusions The noise texture and anatomical accuracy in CT images denoised with an image‐based supervised CNN are greatly influenced by the characteristics and partitioning of training data. With higher‐resolution reconstructions and noise texture‐preserving deep learning denoising, the quality of cCTA images from EID‐CT can be enhanced to enable resolvability of subtle anatomy similar to PCD‐CT.

Abstract Background Multi‐energy CT (MECT) enables quantification of material concentrations by measuring linear attenuation coefficient line integrals with multiple x‐ray spectra. Photon counting detector (PCD)‐CT utilizes a detector‐based approach for MECT that can suffer from substantial spectral overlap, resulting in amplified material quantification noise. Dual‐source dual‐kV approaches for MECT are currently utilized in some energy‐integrating detector (EID)‐CT systems and can potentially be utilized with PCD‐CT for improved spectral separation. Purpose To evaluate the iodine quantification performance of a novel dual‐source (DS)‐PCD‐CT scan mode and compare to single‐source (SS)‐PCD‐CT and DS‐EID‐CT. Materials and Methods A 30 cm × 40 cm solid water phantom with four iodine inserts (2, 5, 10, and 15 mg/mL) was scanned with the three spectral modalities: SS‐PCD‐CT with two energy thresholds, DS‐PCD‐CT (90/Sn150 kV), and DS‐EID‐CT (90/Sn150 kV). For each modality, full‐dose (12 mGy) and half‐dose scans were acquired, and images were reconstructed with filtered back‐projection using a quantitative (Qr40) kernel. When scanning in a DS configuration, the total radiation dose budget is split between two detectors, increasing the strength of a signal‐dependent filter compared to a SS acquisition. To account for this effect, the modulation transfer function (MTF) for each modality was measured from a 0.05 mm tungsten wire. A linear spatial filter was applied to the SS images to match their MTF to that of the DS images. The resulting high‐ and low‐energy images were input into an image‐domain least squares material decomposition algorithm with iodine and water as the two basis materials. Iodine quantification accuracy and noise measured from the iodine basis images were used as figures of merit, and t‐tests used to compare between modalities. Results The 10% MTF cutoffs were 0.56, 0.57, 0.60, and 0.57 lp/mm for DS‐EID‐CT, DS‐PCD‐CT, SS‐PCD‐CT, and SS‐PCD‐CT after MTF‐matching, respectively, with the SS‐PCD‐CT MTF cutoff dropping to 0.58 lp/mm at half‐dose. Without accounting for the signal‐dependent filter by matching the MTFs, the noise in iodine material basis images from SS‐PCD‐CT was 10% higher than that of DS‐EID‐CT. After matching the MTFs, the noise in the SS‐PCD‐CT iodine image was 9%–22% lower than that of the DS‐EID‐CT. The lowest iodine image noise was from the DS‐PCD‐CT, which was 39%–41% lower than the DS‐EID‐CT. The DS‐PCD‐CT noise magnitude was significantly different from the other modalities. Mean iodine quantification accuracy across all measured concentrations was within 5% for all modalities at full dose, but was only below 5% for the DS‐PCD‐CT at half‐dose. Conclusions SS‐PCD‐CT with two energy thresholds outperformed DS‐EID‐CT in terms of image noise in iodine basis images when spatial resolution was matched. DS‐PCD‐CT gave the lowest noise due to the combination of PCD technology and improved spectral separation from the different x‐ray spectra.

AbstractBackgroundInserting noise into existing patient projection data to simulate lower‐radiation‐dose exams has been frequently used in traditional energy‐integrating‐detector (EID)‐CT to optimize radiation dose in clinical protocols and to generate paired images for training deep‐learning‐based reconstruction and noise reduction methods. Recent introduction of photon counting detector CT (PCD‐CT) also requires such a method to accomplish these tasks. However, clinical PCD‐CT scanners often restrict the users access to the raw count data, exporting only the preprocessed, log‐normalized sinogram. Therefore, it remains a challenge to employ projection domain noise insertion algorithms on PCD‐CT.PurposeTo develop and validate a projection domain noise insertion algorithm for PCD‐CT that does not require access to the raw count data.Materials and MethodsA projection‐domain noise model developed originally for EID‐CT was adapted for PCD‐CT. This model requires, as input, a map of the incident number of photons at each detector pixel when no object is in the beam. To obtain the map of incident number of photons, air scans were acquired on a PCD‐CT scanner, then the noise equivalent photon number (NEPN) was calculated from the variance in the log normalized projection data of each scan. Additional air scans were acquired at various mA settings to investigate the impact of pulse pileup on the linearity of NEPN measurement. To validate the noise insertion algorithm, Noise Power Spectra (NPS) were generated from a 30 cm water tank scan and used to compare the noise texture and noise level of measured and simulated half dose and quarter dose images. An anthropomorphic thorax phantom was scanned with automatic exposure control, and noise levels at different slice locations were compared between simulated and measured half dose and quarter dose images. Spectral correlation between energy thresholds T1 and T2, and energy bins, B1 and B2, was compared between simulated and measured data across a wide range of tube current. Additionally, noise insertion was performed on a clinical patient case for qualitative assessment.ResultsThe NPS generated from simulated low dose water tank images showed similar shape and amplitude to that generated from the measured low dose images, differing by a maximum of 5.0% for half dose (HD) T1 images, 6.3% for HD T2 images, 4.1% for quarter dose (QD) T1 images, and 6.1% for QD T2 images. Noise versus slice measurements of the lung phantom showed comparable results between measured and simulated low dose images, with root mean square percent errors of 5.9%, 5.4%, 5.0%, and 4.6% for QD T1, HD T1, QD T2, and HD T2, respectively. NEPN measurements in air were linear up until 112 mA, after which pulse pileup effects significantly distort the air scan NEPN profile. Spectral correlation between T1 and T2 in simulation agreed well with that in the measured data in typical dose ranges.ConclusionsA projection‐domain noise insertion algorithm was developed and validated for PCD‐CT to synthesize low‐dose images from existing scans. It can be used for optimizing scanning protocols and generating paired images for training deep‐learning‐based methods.

Abstract Objective . Photon-counting detector (PCD) CT enables routine virtual-monoenergetic image (VMI) reconstruction. We evaluated the performance of an automatic VMI energy level (keV) selection tool on a clinical PCD-CT system in comparison to an automatic tube potential (kV) selection tool from an energy-integrating-detector (EID) CT system from the same manufacturer. Approach. Four torso-shaped phantoms (20–50 cm width) containing iodine (2, 5, and 10 mg cc −1 ) and calcium (100 mg cc −1 ) were scanned on PCD-CT and EID-CT. Dose optimization techniques, task-based VMI energy level and tube-potential selection on PCD-CT (CARE keV) and task-based tube potential selection on EID-CT (CARE kV), were enabled. CT numbers, image noise, and dose-normalized contrast-to-noise ratio (CNR d ) were compared. Main results . PCD-CT produced task-specific VMIs at 70, 65, 60, and 55 keV for non-contrast, bone, soft tissue with contrast, and vascular settings, respectively. A 120 kV tube potential was automatically selected on PCD-CT for all scans. In comparison, EID-CT used x-ray tube potentials from 80 to 150 kV based on imaging task and phantom size. PCD-CT achieved consistent dose reduction at 9%, 21% and 39% for bone, soft tissue with contrast, and vascular tasks relative to the non-contrast task, independent of phantom size. On EID-CT, dose reduction factor for contrast tasks relative to the non-contrast task ranged from a 65% decrease (vascular task, 70 kV, 20 cm phantom) to a 21% increase (soft tissue with contrast task, 150 kV, 50 cm phantom) due to size-specific tube potential adaptation. PCD-CT CNR d was equivalent to or higher than those of EID-CT for all tasks and phantom sizes, except for the vascular task with 20 cm phantom, where 70 kV EID-CT CNR d outperformed 55 keV PCD-CT images. Significance . PCD-CT produced more consistent CT numbers compared to EID-CT due to standardized VMI output, which greatly benefits standardization efforts and facilitates radiation dose reduction.

AbstractBackgroundObjective and quantitative evaluation for low‐contrast detectability that correlates with human observer performance is lacking for routine CT quality control testing. Channelized Hotelling observer (CHO) is considered a strong candidate to fill the need but has long been deemed impractical to implement due to its requirement of a large number of repeated scans in order to provide accurate and precise estimates of index of detectability (d’). In our previous work, we optimized a CHO model observer on the American College of Radiology (ACR) CT accreditation phantom and achieved accurate measurement of d’ with only 1–3 repeat scans.PurposeIn this work, we aim to validate the repeatability of the proposed CHO‐based low‐contrast evaluation on four scanner models using the ACR CT accreditation phantom.MethodsThe repeatability test was performed on four different scanners from two major CT manufacturers: Siemens Force and Alpha; Canon Prism and Prime SP. An ACR CT phantom was scanned 10 times, each time after repositioning of the phantom. For each repositioning, 3 repeated scans were acquired at 24, 12, and 6 mGy on all four scanner models. CHO was applied at the measured dose levels for different low‐contrast object sizes (4–6 mm). The CHO was also applied to images created using deep learning‐based reconstructions on Canon Prism and to four different scan/reconstruction modes on the Siemens Alpha, a photon‐counting‐detector (PCD)‐CT. The repeatability was evaluated by the probability that a measurement would fall within the ±15% tolerance (P<15%).ResultsWith the CHO setting optimized for the ACR phantom and the use of 3 repeated scans and 9 non‐overlapping slices per scan, the CHO measurement could provide high repeatability with P<15% of 98.8%–99.9% at 12 mGy with IR reconstruction on all four scanners. On scanner A, P<15% were 91.5%–99.9% at the three dose levels and for all three object sizes while the numbers were 93.6%–99.998% on scanner B. P<15% were 96.5%–97.2% for the two deep learning reconstructions and 97.0%–99.97% for the four scan/reconstruction modes on the PCD‐CT.ConclusionThe CHO provided highly repeatable measurements with over 95% probability that a CHO measurement would lie within the ±15% tolerance for most of the dose levels and object sizes on the ACR phantom. The repeatability was maintained when the CHO was applied to images created with a commercial deep learning‐based reconstruction and various scan/reconstruction modes on a PCD‐CT. This study demonstrates that practical implementation of CHO for routine quality control and performance evaluation is feasible.

AbstractObjectiveTo measure the accuracy of material decomposition using a dual‐source photon‐counting‐detector (DS‐PCD) CT operated in the high‐pitch helical scanning mode and compare the results against dual‐source energy‐integrating‐detector (DS‐EID) CT, which requires use of a low‐pitch value in dual‐energy mode.MethodsA DS‐PCD CT and a DS‐EID CT were used to scan a cardiac motion phantom consisting of a 3‐mm diameter iodine cylinder. Iodine maps were reconstructed using DS‐PCD in high‐pitch mode and DS‐EID in low‐pitch mode. Image‐based circularity, diameter, and iodine concentration of the iodine cylinder were calculated and compared between the two scanners. With institutional review board approval, in vivo exams were performed with the DS‐PCD CT in high‐pitch mode. Images were qualitatively compared against patients with similar heart rates that were scanned with DS‐EID CT in low‐pitch dual‐energy mode.ResultsOn iodine maps, the mean circularity was 0.97 ± 0.02 with DS‐PCD in high‐pitch mode and 0.95 ± 0.06 with DS‐EID in low‐pitch mode. The mean diameter was 2.9 ± 0.2 mm with DS‐PCD and 3.1 ± 0.2 mm with DS‐EID, both of which are close to the 3 mm ground truth. For DS‐PCD, the mean iodine concentration was 9.6 ± 0.8 mg/ml and this was consistent with the 9.4 ± 0.6 mg/ml value obtained with the cardiac motion disabled. For DS‐EID, the concentration was 12.7 ± 1.2 mg/ml with motion enabled and 11.7 ± 0.5 mg/ml disabled. The background noise in the iodine maps was 15.1 HU with DS‐PCD and 14.4 HU with DS‐EID, whereas the volume CT dose index (CTDIvol) was 3 mGy with DS‐PCD and 11 mGy with DS‐EID. On comparison of six patients (three on PCD, three on EID) with similar heart rates, DS‐PCD provided iodine maps with well‐defined coronaries even at a high heart rate of 86 beats per minute. Meanwhile, there were substantial motion artifacts in iodine maps obtained with DS‐EID for patients with similar heart rates.ConclusionIn a cardiac motion phantom, DS‐PCD CT can perform accurate material decomposition in high‐pitch mode, providing iodine maps with excellent geometric accuracy and robustness to motion at approximately 38% of the dose for similar noise as DS‐EID CT.

AbstractBackgroundCurrent CT quality control (QC) for low‐contrast detectability relies on visual inspection and measurement of contrast‐to‐noise ratio (CNR). However, CNR numbers become unreliable when it comes to nonlinear methods, such as iterative reconstruction (IR) and deep‐learning‐based techniques. Image quality metrics using channelized Hotelling observer (CHO) have been validated to be well correlated with human observer performance on phantom‐based and patient‐based tasks, but it has not been widely used in routine CT QC mainly because the CHO calculation typically requires a large number of repeated scans in order to provide accurate and precise estimate of index of detectability (d′).PurposeThe main goal of this work is to optimize channel filters and other CHO parameters and accurately estimate the low‐contrast detectability with minimum number of repeated scans for the widely used American College of Radiology (ACR) CT accreditation phantom so that it can become practically feasible for routine CT QC tests.MethodsTo provide a converged d′ value, an ACR phantom was repeatedly scanned 100 times at three dose levels (24, 12, and 6 mGy). Images were reconstructed with two kernels (FBP Br44 and IR Br44‐3). d′ as a function of number of repeated scans was determined for different number of background regions of interest (ROIs), different number of low‐contrast objects, different number of slices per each object, and different channel filter options. A reference d′ was established using the optimized CHO setting, and the bias of d′ was quantified using the d′ calculated from all 100 repeated scans. The variation of d′ at each condition was estimated using a resampling method combining random subsampling among 100 repeated scans and bootstrapping of the ensembles of signal and background ROIs.ResultsOptimized parameters in CHO calculation were determined: two background ROIs per object, four objects per low‐contrast object size, nine non‐overlapping slices per object, and a 4‐channel Gabor filter. The bias and uncertainty were estimated at different numbers of repeated scans using these parameters. When only one single scan was used in the CHO calculation, the bias of d′ was below 6.2% and the uncertainty 15.6‐19.6% for the 6, 5, and 4 mm objects, while with three repeated scans the bias was below 2.0% and uncertainty 8.7‐10.9% for the three object sizes.ConclusionWith optimized parameter settings in CHO, efficient and accurate measurement of low‐contrast detectability on the commonly used ACR phantom becomes feasible, which could potentially lead to adoption of CHO‐based low‐contrast evaluation in routine QC tests.

AbstractPurposeThe purpose of this work is to evaluate the scaled computed tomography (CT) number accuracy of an artificial 120 kV reconstruction technique based on phantom experiments in the context of radiation therapy planning.MethodsAn abdomen‐shaped electron density phantom was scanned on a clinical CT scanner capable of artificial 120 kV reconstruction using different tube potentials from 70 to 150 kV. A series of tissue‐equivalent phantom inserts (lung, adipose, breast, solid water, liver, inner bone, 30%/50% CaCO3, cortical bone) were placed inside the phantom. Images were reconstructed using a conventional quantitative reconstruction kernel as well as the artificial 120 kV reconstruction kernel. Scaled CT numbers of inserts were measured from images acquired at different kVs and compared with those acquired at 120 kV, which were deemed as the ground truth. The relative error was quantified as the percentage deviation of scaled CT numbers acquired at different tube potentials from their ground truth values acquired at 120 kV.ResultsScaled CT numbers measured from images reconstructed using the conventional reconstruction demonstrated a strong kV‐dependence. The relative error in scaled CT numbers ranged from 0.6% (liver insert) to 31.1% (cortical bone insert). The artificial 120 kV reconstruction reduced the kV dependence, especially for bone tissues. The relative error in scaled CT number was reduced to 0.4% (liver insert) and 2.6% (30% CaCO3 insert) using this technique. When tube potential selection was limited to the range of 90 to 150 kV, the relative error was further restrained to <1.2% for all tissue types.ConclusionPhantom results demonstrated that using the artificial 120 kV technique, it was feasible to acquire raw projection data at the desired tube potential and then reconstruct images with scaled CT numbers comparable to those obtained directly at 120 kV. In radiotherapy applications, this technique may allow optimization of tube potential without complicating clinical workflow by eliminating the necessity of maintaining multiple sets of CT calibration curves.

Abstract We report a comprehensive evaluation of a full field-of-view (FOV) photon-counting detector (PCD) computed tomography (CT) system using phantoms, and qualitatively assess image quality in patient examples. A whole-body PCD-CT system with 50 cm FOV, 5.76 cm z-detector coverage and two acquisition modes (standard: 144 × 0.4 mm collimation and ultra-high resolution (UHR): 120 × 0.2 mm collimation) was used in this study. Phantoms were scanned to assess image uniformity, CT number accuracy, noise power spectrum, spatial resolution, material decomposition and virtual monoenergetic imaging (VMI) performance. Four patients were scanned on the PCD-CT system with matched or lower radiation dose than their prior clinical CT scans performed using energy-integrating detector (EID) CT, and the potential clinical impact of PCD-CT was qualitatively evaluated. Phantom results showed water CT numbers within ±5 HU, and image uniformity measured between peripheral and central regions-of-interests to be within ±5 HU. For the UHR mode using a dedicated sharp kernel, the cut-off spatial frequency was 40 line-pairs cm −1 , which corresponds to a 125 μ m limiting in-plane spatial resolution. The full-width-at-half-maximum for the section sensitivity profile was 0.33 mm for the smallest slice thickness (0.2 mm) using the UHR mode. Material decomposition in a multi-energy CT phantom showed accurate material classification, with a root-mean-squared-error of 0.3 mg cc −1 for iodine concentrations (2–15 mg cc −1 ) and 14.2 mg cc −1 for hydroxyapatite concentrations (200 and 400 mg cc −1 ). The average percent error for CT numbers corresponding to the iodine concentrations in VMI (40–70 keV) was 2.75%. Patient PCD-CT images demonstrated better delineation of anatomy for chest and temporal bone exams performed with the UHR mode, which allowed the use of very sharp kernels not possible with EID-CT. VMI and virtual non-contrast images generated from a patient head CT angiography exam using the standard acquisition mode demonstrated the multi-energy capability of the PCD-CT system.

Purpose To determine the accuracy of CT number and calcium score of a kV‐independent technique based on an artificial 120 kV reconstruction, and its potential to reduce radiation dose. Methods Anthropomorphic chest phantoms were scanned on a third‐generation dual‐source CT system equipped with the artificial 120 kV reconstruction. First, a phantom module containing a 20‐mm diameter hydroxyapatite (HA) insert was scanned inside the chest phantoms at different tube potentials (70–140 kV) to evaluate calcium CT number accuracy. Next, three small HA inserts (diameter/length = 5 mm) were inserted into a pork steak and scanned inside the phantoms to evaluate calcium score accuracy at different kVs. Finally, the same setup was scanned using automatic exposure control (AEC) at 120 kV, and then with automatic kV selection (auto‐kV). Phantoms were also scanned at 120 kV using a size‐dependent mA chart. CT numbers of soft tissue and calcium were measured from different kV images. Calcium score of each small HA insert was measured using commercial software. Results The CT number difference from 120 kV was small with tube potentials from 90 to 140 kV for both soft tissue and calcium (maximal difference of 4/5 HU, respectively). Consistent calcium scores were obtained from images of different kVs compared to 120 kV, with a relative difference <8%. Auto‐kV provided a 25–34% dose reduction compared to AEC alone. Conclusion A kV‐independent calcium scoring technique can produce artificial 120 kV images with consistent soft tissue and calcium CT numbers compared to standard 120 kV examinations. When coupled with auto‐kV, this technique can reduce radiation by 25–34% compared to that with AEC alone, while providing consistent calcium scores as that of standard 120 kV examinations.