Improved noise reduction in photon-counting detector CT using prior knowledge-aware iterative denoising neural network.

Purpose: We aim to reduce image noise in high-resolution (HR) virtual monoenergetic images (VMIs) from photon-counting detector (PCD) CT scans by developing a prior knowledge-aware iterative denoising neural network (PKAID-Net) that efficiently exploits the unique noise characteristics of VMIs at different energy (keV) levels. Approach: PKAID-Net offers two major features: first, it utilizes a lower-noise VMI (e.g., 70 keV) as a prior input; second, it iteratively constructs a refined training dataset to improve the neural network's denoising performance. In each iteration, the denoised image from the previous module serves as an updated target image, which is included in the dataset for the subsequent training iteration. Our study includes 10 patient coronary CT angiography exams acquired on a clinical dual-source PCD-CT (NAEOTOM Alpha, Siemens Healthineers). The HR VMIs were reconstructed at 50, 70, and 100 keV, using a sharp vascular kernel (Bv68) and thin (0.6 mm) slice thickness (0.3 mm increment). PKAID-Net's performance was evaluated in terms of image noise, spatial detail preservation, and quantitative accuracy. Results: PKAID-Net achieved a noise reduction of 96% compared to filtered back projection and 65% relative to iterative reconstruction, all while preserving spatial and spectral fidelity and maintaining a natural noise texture. The iterative refinement of PCD-CT data during the training process substantially enhanced the robustness of deep learning-based denoising compared to the original method, which resulted in some spatial detail loss. Conclusions: The PKAID-Net provides substantial noise reduction while maintaining spatial and spectral fidelity of the HR VMIs from PCD-CT.

Technical note: Exploring the detectability of coronary calcification using ultra-high-resolution photon-counting-detector CT.

BACKGROUND: Coronary calcification is a strong indicator of coronary artery disease, and patients with a "zero" coronary calcification score have a much lower risk of future cardiac events than those with even small amounts of calcium. However, false-negative (incorrect zero scores) may occur if small calcifications are missed at CT due to limited spatial resolution. PURPOSE: To demonstrate lower limits of detection for coronary calcification using an ultra-high-resolution (UHR) mode on a clinical photon-counting-detector CT (PCD-CT), compared to a conventional energy-integrating-detector CT (EID-CT). METHODS: Chicken eggshell fragments (0.4-0.8 mm) mimicking coronary calcifications were scanned on a clinical PCD-CT (NAEOTOM Alpha) in UHR mode and a conventional EID-CT (SOMATOM Force) with matched tube potential and radiation dose levels to the PCD-CT. PCD-CT images were reconstructed with a sharp kernel (Qr68) and a quantum iterative algorithm (QIR-3). Two sets of EID-CT images were reconstructed: routine clinical kernel (Qr36, ADMIRE-3) and a sharper kernel (Qr54) with similar noise to PCD-CT images. With institutional review board approval, in vivo exams performed with the PCD-CT in UHR mode were compared against patients' clinical EID-CT exams. The visibility of calcifications on PCD-CT and EID-CT images was assessed and compared qualitatively. RESULTS: PCD-CT images visualized all calcified fragments, while EID-CT failed to detect those below 0.6 mm using a routine protocol. EID-CT with Qr54 improved visibility but distorted boundaries. Calcifications were less visible on EID-CT than PCD-CT as phantom sizes increased. 0.6- and 0.7-mm calcified fragments were barely visible on 35- and 40-cm phantom EID-CT images. Patient cases showed small calcifications missed on EID-CT but detected on PCD-CT. CONCLUSION: At matched radiation dose, PCD-CT in UHR mode provided higher spatial resolution and improved the detectability of small calcified fragments for different phantom/patient sizes in comparison to EID-CT.

PKAID-Net: Prior Knowledge Aware Iterative Denoising Neural Network for Photon Counting Detector CT.

An important feature enabled by Photon-Counting Detector (PCD) CT is the simultaneous acquisition of multi-energy data, which can produce virtual monoenergetic images (VMIs) at a high spatial resolution. However, noise levels observed in the high-resolution (HR) VMIs are markedly increased. Recent work involving deep learning methods has shown great potential in CT image denoising. Many CNN applications involve training using spatially co-registered low- and high-dose CT images featuring high and low image noise, respectively. However, this is implausible in routine clinical practice. Further, typical denoising methods treat each VMI energy level independently, without consideration of the valuable information in the spectral domain. To overcome these obstacles, we propose a prior knowledge-aware iterative denoising neural network (PKAID-Net). PKAID-Net offers two major benefits: first, it utilizes spectral information by including a lower-noise VMI as a prior input; and second, it iteratively constructs refined datasets for neural network training to improve the denoising performance. This study includes 10 patient coronary CT angiography (CTA) exams acquired on a clinical HR PCD-CT (NAEOTOM Alpha, Siemens Healthineers). The HR VMIs were reconstructed at 50 and 70 keV, using a sharp kernel (Bv68) and thin (0.6 mm, 0.3 mm increment) slice thickness. Results showed that the PKAID-Net provided a noise reduction of 96% and 70% relative to FBP and iterative reconstruction, respectively while maintaining spatial and spectral fidelity and a natural noise texture. These results demonstrate the noise reduction capacity of PKAID-Net as applied to cutting-edge PCD-CT data to enable HR, multi-energy cardiac CT imaging.

Ex vivo coronary calcium volume quantification using a high-spatial-resolution clinical photon-counting-detector computed tomography.

Purpose: Coronary artery calcification (CAC) is an important indicator of coronary disease. Accurate volume quantification of CAC is challenging using computed tomography (CT) due to calcium blooming, which is a consequence of limited spatial resolution. Ex vivo coronary specimens were scanned on an ultra-high-resolution (UHR) clinical photon-counting detector (PCD) CT scanner, and the accuracy of CAC volume estimation was compared with a state-of-the-art conventional energy-integrating detector (EID) CT, a previous-generation investigational PCD-CT, and micro-CT. Approach: CAC specimens (n=13) were scanned on EID-CT and PCD-CT using matched parameters (120 kV, 9.3 mGy CTDIvol). EID-CT images were reconstructed using our institutional routine clinical protocol for CAC quantification. UHR PCD-CT data were reconstructed using a sharper kernel. An image-based denoising algorithm was applied to the PCD-CT images to achieve similar noise levels as EID-CT. Micro-CT images served as the volume reference standard. Calcification images were segmented, and their volume estimates were compared. The CT data were further compared with previous work using an investigational PCD-CT. Results: Compared with micro-CT, CT volume estimates had a mean absolute percent error of 24.1%±25.6% for clinical PCD-CT, 60.1%±48.2% for EID-CT, and 51.1%±41.7% for previous-generation PCD-CT. Clinical PCD-CT absolute percent error was significantly (p<0.01) lower than both EID-CT and previous generation PCD-CT. The mean calcification CT number and contrast-to-noise ratio were both significantly (p<0.01) higher in clinical PCD-CT relative to EID-CT. Conclusions: UHR clinical PCD-CT showed reduced calcium blooming artifacts and further enabled improved accuracy of CAC quantification beyond that of conventional EID-CT and previous generation PCD-CT systems.

Measurement of enhanced vasa vasorum density in a porcine carotid model using photon counting detector CT.

Purpose: The onset of atherosclerosis is preceded by changes in blood perfusion within the arterial wall due to localized proliferation of the vasa vasorum. The purpose of this study was to quantify these changes in spatial density of the vasa vasorum using a research whole-body photon-counting detector CT (PCD-CT) scanner and a porcine model. Approach: Vasa vasorum angiogenesis was stimulated in the left carotid artery wall of anesthetized pigs ( n = 5 ) while the right carotid served as a control. After a 6-week recovery period, the animals were scanned on the PCD-CT prior to and after injection of iodinated contrast. Annular regions of interest were used to measure wall enhancement in the injured and control arteries. The exact Wilcoxon-signed rank test was used to determine whether a significant difference in contrast enhancement existed between the injured and control arterial walls. Results: The greatest arterial wall enhancement was observed following contrast recirculation. The wall enhancement measurements made over these time points revealed that the enhancement was greater in the injured artery for 13/16 scanned arterial regions. Using an exact Wilcoxon-signed rank test, a significantly increased enhancement ratio was found in injured arteries compared with control arteries ( p = 0.013 ). Vasa vasorum angiogenesis was confirmed in micro-CT scans of excised arteries. Conclusions: Whole-body PCD-CT scanners can be used to detect and quantify the increased perfusion occurring within the porcine carotid arterial wall resulting from an increased density of vasa vasorum.

Pie-Net: Prior-information-enabled deep learning noise reduction for coronary CT angiography acquired with a photon counting detector CT.

BACKGROUND: Photon-counting-detector CT (PCD-CT) enables the production of virtual monoenergetic images (VMIs) at a high spatial resolution (HR) via simultaneous acquisition of multi-energy data. However, noise levels in these HR VMIs are markedly increased. PURPOSE: To develop a deep learning technique that utilizes a lower noise VMI as prior information to reduce image noise in HR, PCD-CT coronary CT angiography (CTA). METHODS: Coronary CTA exams of 10 patients were acquired using PCD-CT (NAEOTOM Alpha, Siemens Healthineers). A prior-information-enabled neural network (Pie-Net) was developed, treating one lower-noise VMI (e.g., 70 keV) as a prior input and one noisy VMI (e.g., 50 keV or 100 keV) as another. For data preprocessing, noisy VMIs were reconstructed by filtered back-projection (FBP) and iterative reconstruction (IR), which were then subtracted to generate "noise-only" images. Spatial decoupling was applied to the noise-only images to mitigate overfitting and improve randomization. Thicker slice averaging was used for the IR and prior images. The final training inputs for the convolutional neural network (CNN) inside the Pie-Net consisted of thicker-slice signal images with the reinsertion of spatially decoupled noise-only images and the thicker-slice prior images. The CNN training labels consisted of the corresponding thicker-slice label images without noise insertion. Pie-Net's performance was evaluated in terms of image noise, spatial detail preservation, and quantitative accuracy, and compared to a U-net-based method that did not include prior information. RESULTS: Pie-Net provided strong noise reduction, by 95 ± 1% relative to FBP and by 60 ± 8% relative to IR. For HR VMIs at different keV (e.g., 50 keV or 100 keV), Pie-Net maintained spatial and spectral fidelity. The inclusion of prior information from the PCD-CT data in the spectral domain was able to improve a robust deep learning-based denoising performance compared to the U-net-based method, which caused some loss of spatial detail and introduced some artifacts. CONCLUSION: The proposed Pie-Net achieved substantial noise reduction while preserving HR VMI's spatial and spectral properties.

Accuracy of Nodule Volume and Airway Wall Thickness Measurement Using Low-Dose Chest CT on a Photon-Counting Detector CT Scanner.

OBJECTIVES: A comparison of high-resolution photon-counting detector computed tomography (PCD-CT) versus energy-integrating detector (EID) CT via a phantom study using low-dose chest CT to evaluate nodule volume and airway wall thickness quantification. MATERIALS AND METHODS: Twelve solid and ground-glass lung nodule phantoms with 3 diameters (5 mm, 8 mm, and 10 mm) and 2 shapes (spherical and star-shaped) and 12 airway tube phantoms (wall thicknesses, 0.27-1.54 mm) were placed in an anthropomorphic chest phantom. The phantom was scanned with EID-CT and PCD-CT at 5 dose levels (CTDI vol = 0.1-0.8 mGy at Sn-100 kV, 7.35 mGy at 120 kV). All images were iteratively reconstructed using matched kernels for EID-CT and medium-sharp kernel (MK) PCD-CT and an ultra-sharp kernel (USK) PCD-CT kernel, and image noise at each dose level was quantified. Nodule volumes were measured using semiautomated segmentation software, and the accuracy was expressed as the percentage error between segmented and reference volumes. Airway wall thicknesses were measured, and the root-mean-square error across all tubes was evaluated. RESULTS: MK PCD-CT images had the lowest noise. At 0.1 mGy, the mean volume accuracy for the solid and ground-glass nodules was improved in USK PCD-CT (3.1% and 3.3% error) compared with MK PCD-CT (9.9% and 10.2% error) and EID-CT images (11.4% and 9.2% error), respectively. At 0.2 mGy and 0.8 mGy, the wall thickness root-mean-square error values were 0.42 mm and 0.41 mm for EID-CT, 0.54 mm and 0.49 mm for MK PCD-CT, and 0.23 mm and 0.16 mm for USK PCD-CT. CONCLUSIONS: USK PCD-CT provided more accurate lung nodule volume and airway wall thickness quantification at lower radiation dose compared with MK PCD-CT and EID-CT.

Characterization of single- and multi-energy CT performance of an oral dark borosilicate contrast media using a clinical photon-counting-detector CT platform.

BACKGROUND: The feasibility of oral dark contrast media is under exploration in abdominal computed tomography (CT) applications. One of the experimental contrast media in this class is dark borosilicate contrast media (DBCM), which has a CT attenuation lower than that of intra-abdominal fat. PURPOSE: To evaluate the performances of DBCM using single- and multi-energy CT imaging on a clinical photon-counting-detector CT (PCD-CT). METHODS: Five vials, three with iodinated contrast agent (5, 10, and 20 mg/mL; Omnipaque 350) and two with DBCM (6% and 12%; Nextrast, Inc.), and one solid-water rod (neutral contrast agent) were inserted into two multi-energy CT phantoms, and scanned on a clinical PCD-CT system (NAEOTOM Alpha) at 90, 120, 140, Sn100, and Sn140 kV (Sn: tin filter) in multi-energy mode. CARE keV IQ level was 180 (CTDIvol: 3.0 and 12.0 mGy for the small and large phantoms, respectively). Low-energy threshold images were reconstructed with a quantitative kernel (Qr40, iterative reconstruction strength 2) and slice thickness/increment of 2.0/2.0 mm. Virtual monoenergetic images (VMIs) were reconstructed from 40 to 140 keV at 10 keV increments. On all images, average CT numbers for each vial/rod were measured using circular region-of-interests and averaged over eight slices. The contrast-to-noise ratio (CNR) of iodine (5 mg/mL) against DBCM was calculated and plotted against tube potential and VMI energy level, and compared to the CNR of iodine against water. Similar analyses were performed on iodine maps and VNC images derived from the multi-energy scan at 120 kV. RESULTS: With increasing kV or VMI keV, the negative HU of DBCM decreased only slightly, whereas the positive HU of iodine decreased across all contrast concentrations and phantom sizes. CT numbers for DBCM decreased from -178.5 ± 9.6 to -194.4 ± 6.3 HU (small phantom) and from -181.7 ± 15.7 to -192.1 ± 11.9 HU (large phantom) for DBCM-12% from 90 to Sn140 kV; on VMIs, the CT numbers for DBCM decreased minimally from -147.1 ± 15.7 to -185.1 ± 9.2 HU (small phantom) and -158.8 ± 28.6 to -188.9 ± 14.7 HU (large phantom) from 40 to 70 keV, but remained stable from 80 to 140 keV. The highest iodine CNR against DBCM in low-energy threshold images was seen at 90 or Sn140 kV for the small phantom, whereas all CNR values from low-energy threshold images for the large phantom were comparable. The CNR values of iodine against DBCM computed on VMIs were highest at 40 or 70 keV depending on iodine and DBCM concentrations. The CNR values of iodine against DBCM were consistently higher than iodine to water (up to 460% higher dependent on energy level). Further, the CNR of iodine compared to DBCM is less affected by VMI energy level than the identical comparison between iodine and water: CNR values at 140 keV were reduced by 46.6% (small phantom) or 42.6% (large phantom) compared to 40 keV; CNR values for iodine compared to water were reduced by 86.3% and 83.8% for similar phantom sizes, respectively. Compared to 70 keV VMI, the iodine CNR against DBCM was 13%-79% lower on iodine maps and VNC. CONCLUSIONS: When evaluated at different tube potentials and VMI energy levels using a clinical PCD-CT system, DBCM showed consistently higher CNR compared to iodine versus water (a neutral contrast).

Deep-learning lesion and noise insertion for virtual clinical trial in Chest CT.

Accurate and objective image quality assessment is essential for the task of radiation dose optimization in clinical CT. Standard method relies on multi-reader multi-case (MRMC) studies in which radiologists are tasked to interpret diagnostic image quality of many carefully-collected positive and negative cases. The efficiency of such MRMC studies is frequently challenged by the lengthy and expensive procedure of case collection and the establishment of clinical reference standard. To address this challenge, multiple methods of virtual clinical trial to synthesize patient cases at different conditions have been proposed. Projection-domain lesion- / noise-insertion methods require the access to patient raw data and vendor-specific proprietary tools which are frequently not accessible to most users. The conventional image-domain noise-insertion methods are often challenged by the over-simplified lesion models and CT system models which may not represent conditions in real scans. In this work, we developed deep-learning lesion and noise insertion techniques that can synthesize chest CT images at different dose levels with and without lung nodules using existing patient cases. The proposed method involved a nodule-insertion convolutional neural network (CNN) and a noise-insertion CNN. Both CNNs demonstrated comparable quality to our previously-validated projection domain lesion- / noise-insertion techniques: mean structural similarity index (SSIM) of inserted nodules 0.94 (routine dose), and mean percent noise difference ~5% (50% of routine dose). The proposed deep-learning techniques for chest CT virtual clinical trial operate directly on image domain, which is more widely applicable than projection-domain techniques.

Deep-learning-based direct synthesis of low-energy virtual monoenergetic images with multi-energy CT.

Purpose: We developed a deep learning method to reduce noise and beam-hardening artifact in virtual monoenergetic image (VMI) at low x-ray energy levels. Approach: An encoder-decoder type convolutional neural network was implemented with customized inception modules and in-house-designed training loss (denoted as Incept-net), to directly estimate VMI from multi-energy CT images. Images of an abdomen-sized water phantom with varying insert materials were acquired from a research photon-counting-detector CT. The Incept-net was trained with image patches ( 64 × 64 pixels ) extracted from the phantom data, as well as synthesized, random-shaped numerical insert materials. The whole CT images ( 512 × 512 pixels ) with the remaining real insert materials that were unseen in network training were used for testing. Seven contrast-enhanced abdominal CT exams were used for preliminary evaluation of Incept-net generalizability over anatomical background. Mean absolute percentage error (MAPE) was used to evaluate CT number accuracy. Results: Compared to commercial VMI software, Incept-net largely suppressed beam-hardening artifact and reduced noise (53%) in phantom study. Incept-net presented comparable CT number accuracy at higher-density ( P -value [0.0625, 0.999]) and improved it at lower-density inserts ( P - value = 0.0313 ) with overall MAPE: Incept-net [2.9%, 4.6%]; commercial-VMI [6.7%, 10.9%]. In patient images, Incept-net suppressed beam-hardening artifact and reduced noise (up to 50%, P - value = 0.0156 ). Conclusion: In this preliminary study, Incept-net presented the potential to improve low-energy VMI quality.

Multi-energy CT imaging for large patients using dual-source photon-counting detector CT.

Multi-energy CT imaging of large patients with conventional dual-energy (DE)-CT using an energy-integrating-detector (EID) is challenging due to photon starvation-induced image artifacts, especially in lower tube potential (80-100 kV) images. Here, we performed phantom experiments to investigate the performance of DECT for morbidly obese patients, using an iodine and water material decomposition task as an example, on an emulated dual-source (DS)-photon-counting-detector (PCD)-CT, and compared its performance with a clinical DS-EID-CT. An abdominal CT phantom with iodine inserts of different concentrations was wrapped with tissue-equivalent gel layers to emulate a large patient (50 cm lateral size). The phantom was scanned on a research whole-body single-source (SS)-PCD-CT (140 kV tube potential), a DS-PCD-CT (100/Sn140 kV; Sn140 indicates 140 kV with Sn filter), and a clinical DS-EID-CT (100/Sn140 kV) with the same radiation dose. Phantom scans were repeated five times on each system. The DS-PCD-CT acquisition was emulated by scanning twice on the SS-PCD-CT using different tube potentials. The multi-energy CT images acquired on each system were then reconstructed, and iodine- and water-specific images were generated using material decomposition. The root-mean-square-error (RMSE) between true and measured iodine concentrations were calculated for each system and compared. The images acquired on the DS-EID-CT showed severe artifacts, including ringing, reduced uniformity, and photon starvation artifacts, especially for low-energy images. These were largely reduced in DS-PCD-CT images. The CT number difference that was measured using regions-of-interest across field-of-view were reduced from 20.3 ± 0.9 (DS-EID-CT) to 2.5 ± 0.4 HU on DS-PCD-CT, showing improved image uniformity using DS-PCD-CT. Iodine RMSE was reduced from 3.42 ± 0.03 mg ml-1 (SS-PCD-CT) and 2.90 ± 0.03 mg ml-1 (DS-EID-CT) to 2.39 ± 0.05 mg ml-1 using DS-PCD-CT. DS-PCD-CT out-performed a clinical DS-EID-CT for iodine and water-based material decomposition on phantom emulating obese patients by reducing image artifacts and improving iodine quantification (RMSE reduced by 20%). With DS-PCD-CT, multi-energy CT can be performed on large patients that cannot be accommodated with current DECT.

Evaluation of a photon counting Medipix3RX cadmium zinc telluride spectral x-ray detector.

We assess the performance of a cadmium zinc telluride (CZT)-based Medipix3RX energy-resolving and photon-counting x-ray detector as a candidate for spectral microcomputed tomography (micro-CT) imaging. It features an array of 128 × 128 , 110 - µ m 2 pixels, each with four simultaneous threshold counters that utilize real-time charge summing. Each pixel's response is assessed by imaging with a range of incident x-ray intensities and detector integration times. Energy-related assessments are made by exposing the detector to the emission from an I-125 radioisotope brachytherapy seed. Long-term stability is assessed by repeating identical exposures over the course of 1 h. The high yield of properly functioning pixels (98.8%), long-term stability (linear regression of whole-chip response over 1 h of acquisitions: y = - 0.0038 x + 2284 ; standard deviation: 3.7 counts), and energy resolution [2.5 keV full-width half-maximum (FWHM) (single pixel), 3.7 keV FWHM (across the full image)] make this device suitable for spectral micro-CT.