Gregory Michalak

Validation of proton stopping power ratio estimation based on dual energy CT using fresh tissue samples

Dual energy CT (DECT) has been shown, in theoretical and phantom studies, to improve the stopping power ratio (SPR) determination used for proton treatment planning compared to the use of single energy CT (SECT). However, it has not been shown that this also extends to organic tissues. The purpose of this study was therefore to investigate the accuracy of SPR estimation for fresh pork and beef tissue samples used as surrogates of human tissues. The reference SPRs for fourteen tissue samples, which included fat, muscle and femur bone, were measured using proton pencil beams. The tissue samples were subsequently CT scanned using four different scanners with different dual energy acquisition modes, giving in total six DECT-based SPR estimations for each sample. The SPR was estimated using a proprietary algorithm (syngo.via DE Rho/Z Maps, Siemens Healthcare, Forchheim, Germany) for extracting the electron density and the effective atomic number. SECT images were also acquired and SECT-based SPR estimations were performed using a clinical Hounsfield look-up table. The mean and standard deviation of the SPR over large volume-of-interests were calculated. For the six different DECT acquisition methods, the root-mean-square errors (RMSEs) for the SPR estimates over all tissue samples were between 0.9% and 1.5%. For the SECT-based SPR estimation the RMSE was 2.8%. For one DECT acquisition method, a positive bias was seen in the SPR estimates, having a mean error of 1.3%. The largest errors were found in the very dense cortical bone from a beef femur. This study confirms the advantages of DECT-based SPR estimation although good results were also obtained using SECT for most tissues.

Low-dose CT for the detection and classification of metastatic liver lesions: Results of the 2016 Low Dose CT Grand Challenge

Purpose: Using common datasets, to estimate and compare the diagnostic performance of image‐based denoising techniques or iterative reconstruction algorithms for the task of detecting hepatic metastases. Methods: Datasets from contrast‐enhanced CT scans of the liver were provided to participants in an NIH‐, AAPM‐ and Mayo Clinic‐sponsored Low Dose CT Grand Challenge. Training data included full‐dose and quarter‐dose scans of the ACR CT accreditation phantom and 10 patient examinations; both images and projections were provided in the training data. Projection data were supplied in a vendor‐neutral standardized format (DICOM‐CT‐PD). Twenty quarter‐dose patient datasets were provided to each participant for testing the performance of their technique. Images were provided to sites intending to perform denoising in the image domain. Fully preprocessed projection data and statistical noise maps were provided to sites intending to perform iterative reconstruction. Upon return of the denoised or iteratively reconstructed quarter‐dose images, randomized, blinded evaluation of the cases was performed using a Latin Square study design by 11 senior radiology residents or fellows, who marked the locations of identified hepatic metastases. Markings were scored against reference locations of clinically or pathologically demonstrated metastases to determine a per‐lesion normalized score and a per‐case normalized score (a faculty abdominal radiologist established the reference location using clinical and pathological information). Scores increased for correct detections; scores decreased for missed or incorrect detections. The winner for the competition was the entry that produced the highest total score (mean of the per‐lesion and per‐case normalized score). Reader confidence was used to compute a Jackknife alternative free‐response receiver operating characteristic (JAFROC) figure of merit, which was used for breaking ties. Results: 103 participants from 90 sites and 26 countries registered to participate. Training data were shared with 77 sites that completed the data sharing agreements. Subsequently, 41 sites downloaded the 20 test cases, which included only the 25% dose data (CTDIvol = 3.0 ± 1.8 mGy, SSDE = 3.5 ± 1.3 mGy). 22 sites submitted results for evaluation. One site provided binary images and one site provided images with severe artifacts; cases from these sites were excluded from review and the participants removed from the challenge. The mean (range) per‐lesion and per‐case normalized scores were −24.2% (−75.8%, 3%) and 47% (10%, 70%), respectively. Compared to reader results for commercially reconstructed quarter‐dose images with no noise reduction, 11 of the 20 sites showed a numeric improvement in the mean JAFROC figure of merit. Notably two sites performed comparably to the reader results for full‐dose commercial images. The study was not designed for these comparisons, so wide confidence intervals surrounded these figures of merit and the results should be used only to motivate future testing. Conclusion: Infrastructure and methodology were developed to rapidly estimate observer performance for liver metastasis detection in low‐dose CT examinations of the liver after either image‐based denoising or iterative reconstruction. The results demonstrated large differences in detection and classification performance between noise reduction methods, although the majority of methods provided some improvement in performance relative to the commercial quarter‐dose images with no noise reduction applied.

Utility of single-energy and dual-energy computed tomography in clot characterization: An in-vitro study:

Background and purpose Because computed tomography (CT) is the most commonly used imaging modality for the evaluation of acute ischemic stroke patients, developing CT-based techniques for improving clot characterization could prove useful. The purpose of this in-vitro study was to determine which single-energy or dual-energy CT techniques provided optimum discrimination between red blood cell (RBC) and fibrin-rich clots. Materials and methods Seven clot types with varying fibrin and RBC densities were made (90% RBC, 99% RBC, 63% RBC, 36% RBC, 18% RBC and 0% RBC with high and low fibrin density) and their composition was verified histologically. Ten of each clot type were created and scanned with a second generation dual source scanner using three single (80 kV, 100 kV, 120 kV) and two dual-energy protocols (80/Sn 140 kV and 100/Sn 140 kV). A region of interest (ROI) was placed over each clot and mean attenuation was measured. Receiver operating characteristic curves were calculated at each energy level to determine the accuracy at differentiating RBC-rich clots from fibrin-rich clots. Results Clot attenuation increased with RBC content at all energy levels. Single-energy at 80 kV and 120 kV and dual-energy 80/Sn 140 kV protocols allowed for distinguishing between all clot types, with the exception of 36% RBC and 18% RBC. On receiver operating characteristic curve analysis, the 80/Sn 140 kV dual-energy protocol had the highest area under the curve for distinguishing between fibrin-rich and RBC-rich clots (area under the curve 0.99). Conclusions Dual-energy CT with 80/Sn 140 kV had the highest accuracy for differentiating RBC-rich and fibrin-rich in-vitro thrombi. Further studies are needed to study the utility of non-contrast dual-energy CT in thrombus characterization in acute ischemic stroke.

A comparison of relative proton stopping power measurements across patient size using dual- and single-energy CT

Abstract Purpose: To evaluate the accuracy and precision across phantom size of a dual-energy computed tomography (DECT) technique used to calculate relative proton stopping power (SPR) in tissue-simulating materials and a silicone implant relative to conventional single-energy CT (SECT). Material and methods: A 32 cm lateral diameter (CIRS model 062M, Norfolk, Virginia) electron density phantom containing inserts which simulated the chemical composition of eight tissues in a solid-water background was scanned using SECT and DECT. A liquid water insert was included to confirm CT number accuracy. All materials were also placed in four water tanks, ranging from 15 to 45 cm in lateral width and scanned using DECT and SECT. A silicone breast implant was scanned in the same water phantoms. SPR values were calculated based on commercial software (syngo CT Dual Energy, Siemens Healthcare GmbH) and compared to reference values derived from proton beam measurements. Accuracy and precision were quantified across phantom size using percent error and standard deviation. Graphical and regression analysis were used to determine whether SECT or DECT was superior in estimating SPR across phantom size. Results: Both DECT and SECT SPR data resulted in good agreement with the reference values. Percent error was ±3% for both DECT and SECT in all materials except lung and dense bone. The coefficient of variation (CV) across materials and phantom sizes was 1.12% for SECT and 0.96% for DECT. Material-specific regression and graphical analysis did not reveal size dependence for either technique but did show reduced systematic bias with DECT for dense bone and liver. Mean percent error in SPR for the implant was reduced from 11.46% for SECT to 0.49% for DECT. Conclusions: We demonstrate the superior ability of DECT to mitigate systematic bias in bones and liver and estimate SPR in a silicone breast implant.

Theoretical and experimental analysis of photon counting detector CT for proton stopping power prediction

PURPOSE Photon counting detectors (PCDs) are being introduced in advanced x-ray computed tomography (CT) scanners. From a single PCD-CT acquisition, multiple images can be reconstructed, each based on only a part of the original x-ray spectrum. In this study, we investigated whether PCD-CT can be used to estimate stopping power ratios (SPRs) for proton therapy treatment planning, both by comparing to other SPR methods proposed for single energy CT (SECT) and dual energy CT (DECT) as well as to experimental measurements. METHODS A previously developed DECT-based SPR estimation method was adapted to PCD-CT data, by adjusting the estimation equations to allow for more energy spectra. The method was calibrated directly on noisy data to increase the robustness toward image noise. The new PCD SPR estimation method was tested in theoretical calculations as well as in an experimental setup, using both four and two energy bin PCD-CT images, and through comparison to two other SPR methods proposed for SECT and DECT. These two methods were also evaluated on PCD-CT images, full spectrum (one-bin) or two-bin images, respectively. In a theoretical framework, we evaluated the effect of patient-specific tissue variations (density and elemental composition) and image noise on the SPR accuracy; the latter effect was assessed by applying three different noise levels (low, medium, and high noise). SPR estimates derived using real PCD-CT images were compared to experimentally measured SPRs in nine organic tissue samples, including fat, muscle, and bone tissues. RESULTS For the theoretical calculations, the root-mean-square error (RMSE) of the SPR estimation was 0.1% for the new PCD method using both two and four energy bins, compared to 0.2% and 0.7% for the DECT- and SECT-based method, respectively. The PCD method was found to be very robust toward CT image noise, with a RMSE of 2.7% when high noise was added to the CT numbers. Introducing tissue variations, the RMSE only increased to 0.5%; even when adding high image noise to the changed tissues, the RMSE stayed within 3.1%. In the experimental measurements, the RMSE over the nine tissue samples was 0.8% when using two energy bins, and 1.0% for the four-bin images. CONCLUSIONS In all tested cases, the new PCD method produced similar or better results than the SECT- and DECT-based methods, showing an overall improvement of the SPR accuracy. This study thus demonstrated that PCD-CT scans will be a qualified candidate for SPR estimations.

Determination of optimal image type and lowest detectable concentration for iodine detection on a photon counting detector-based multi-energy CT system

Photon counting detector (PCD) based multi-energy CT is able to generate different types of images such as virtual monoenergetic images (VMIs) and material specific images (e.g., iodine maps) in addition to the conventional single energy images. The purpose of this study is to determine the image type that has optimal iodine detection and to determine the lowest detectable iodine concentration using a PCD-CT system. A 35 cm body phantom with iodine inserts of 4 concentrations and 2 sizes was scanned on a research PCD-CT system. For each iodine concentration, 80 repeated scans were performed and images were reconstructed for each energy threshold. In addition, VMIs at different keVs and iodine maps were also generated. CNR was measured for each type of images. A channelized Hotelling observer was used to assess iodine detectability after being validated with human observer studies, with area under the ROC curve (AUC) as a figure of merit. The agreement between model and human observer performance indicated that model observer could serve as an effective approach to determine optimal image type for the clinical practice and to determine the lowest detectable iodine concentration. Results demonstrated that for all size and concentration combinations, VMI at 70 keV had similar performance as that of threshold low images, both of which outperformed the iodine map images. At the AUC value of 0.8, iodine concentration as low as 0.2 mgI/cc could be detected for an 8 mm object and 0.5 mgI/cc for a 4 mm object with a 5 mm slice thickness.

Intrarenal fat deposition does not interfere with the measurement of single-kidney perfusion in obese swine using multi-detector computed tomography

BACKGROUND Altered vascular structure or function in several diseases may impair renal perfusion. Multi-detector computed tomography (MDCT) is a non-invasive tool to assess single-kidney perfusion and function based on dynamic changes in tissue attenuation during contrast media transit. However, changes in basal tissue attenuation might hamper these assessments, despite background subtraction. Evaluation of iodine concentration using the dual-energy (DECT) MDCT mode allows excluding effects of basal values on dynamic changes in tissue attenuation. We tested whether decreased basal kidney attenuation secondary to intrarenal fat deposition in swine obesity interferes with assessment of renal perfusion using MDCT. METHODS Domestic pigs were fed a standard (lean) or a high-cholesterol/carbohydrate (obese) diet (n = 5 each) for 16 weeks, and both kidneys were then imaged using MDCT/DECT after iodinated contrast injection. DECT images were post-processed to generate iodine and virtual-non-contrast (VNC) datasets, and the MDCT kidney/aorta CT number (following background subtraction) and DECT iodine ratios calculated during the peak vascular phase as surrogates of renal perfusion. Intrarenal fat was subsequently assessed with Oil-Red-O staining. RESULTS VNC maps in obese pigs revealed decreased basal cortical attenuation, and histology confirmed increased renal tissue fat deposition. Nevertheless, the kidney/aorta attenuation and iodine ratios remained similar, and unchanged compared to lean pigs. CONCLUSIONS Despite decreased basal attenuation secondary to renal adiposity, background subtraction allows adequate assessment of kidney perfusion in obese pigs using MDCT. These observations support the feasibility of renal perfusion assessment in obese subjects using MDCT.

Randomized Trial of Spheroid Reservoir Bioartificial Liver in Porcine Model of Post‐Hepatectomy Liver Failure

Acute liver failure (ALF) is a catastrophic condition that can occur after major liver resection. The aim of this study was to determine the effects of the spheroid reservoir bio‐artificial liver (SRBAL) on survival, serum chemistry, and liver regeneration in posthepatectomy ALF pigs. Wild‐type large white swine (20 kg‐30 kg) underwent intracranial pressure (ICP) probe placement followed by 85% hepatectomy. Computed tomography (CT) volumetrics were performed to measure the extent of resection, and at 48 hours following hepatectomy to assess regeneration of the remnant liver. Animals were randomized into three groups based on treatment delivered 24‐48 hours after hepatectomy: Group1—standard medical therapy (SMT, n = 6); Group2—SMT plus bio‐artificial liver treatment using no hepatocytes (0 g, n = 6); and Group3—SMT plus SRBAL treatment using 200 g of primary porcine hepatocyte spheroids (200 g, n = 6). The primary endpoint was survival to 90 hours following hepatectomy. Death equivalent was defined as unresponsive grade 4 hepatic encephalopathy or ICP greater than 20 mmHg with clinical evidence of brain herniation. All animals in both (SMT and 0 g) control groups met the death equivalent before 51 hours following hepatectomy. Five of 6 animals in the 200‐g group survived to 90 hours (P < 0.01). The mean ammonia, ICP, and international normalized ratio values were significantly lower in the 200‐g group. CT volumetrics demonstrated increased volume regeneration at 48 hours following hepatectomy in the 200‐g group compared with the SMT (P < 0.01) and 0‐g (P < 0.01) groups. Ki‐67 staining showed increased positive staining at 48 hours following hepatectomy (P < 0.01). Conclusion: The SRBAL improved survival, reduced ammonia, and accelerated liver regeneration in posthepatectomy ALF. Improved survival was associated with a neuroprotective benefit of SRBAL therapy. These favorable results warrant further clinical testing of the SRBAL.

SU-F-J-75: Accuracy and Stability of Electron Density Measurements Across Patient Size Using Dual Energy CT

PURPOSE Dual energy (DE) CT can be used to characterize tissue composition. One application of DE CT is to measure electron density (ED, rho) and atomic number (Z) for use in radiation therapy treatment planning. This work evaluated the accuracy and stability of ED estimation as patient size varied for both single-energy (SE) and DE CT. METHODS An ED phantom (CIRS) and four torso-shaped water tanks (lateral widths 15, 25, 35 and 45 cm) containing 8 tissue-simulating cylinders of known ED were scanned on a dual-source CT system (Siemens Somatom Force) in SE (120 kV) and DE (90/150Sn) modes. Additional scans were performed on the 15 and 25 cm water tanks using DE techniques of 70/150Sn and 80/150Sn, respectively. CTDIvol was matched for all SE and DE scans for a given phantom size. Images were reconstructed using quantitative kernels to preserve CT number accuracy. ED was estimated in each test cylinder and in solid and liquid water using calibration measurements acquired in the CIRS phantom (SE) and a Rho-Z algorithm (DE). RESULTS ED estimates showed good agreement with the nominal ED values when using Rho-Z (slope = 1.0051, R2 = 0.9982). Mean percent error was similar between SE (1.21%) and DE (1.28%). Mean deviation across patient size decreased 34% (1.43% with SE, 0.95% with DE). When compared to 90/150Sn, DE techniques of 70/150Sn and 80/150Sn showed mean differences in ED of 0.43% and 0.15%, respectively. CONCLUSION While both DE Rho-Z and SE CT number calibration methods are both accurate for estimating ED, Rho-Z offers the advantages of having less variability across patient size, not requiring a phantom calibration, and being able to distinguish between materials of similar attenuation, but different chemical composition. Low kV DE pairs are an option in small patients due to lack of effect on ED accuracy. This research was supported by Siemens Healthcare.

TU‐F‐18A‐09: CT Number Stability Across Patient Sizes Using Virtual‐Monoenergetic Dual‐Energy CT

PURPOSE Virtual-monoenergetic imaging uses dual-energy CT data to synthesize images corresponding to a single photon energy, thereby reducing beam-hardening artifacts. This work evaluated the ability of a commercial virtual-monoenergetic algorithm to achieve stable CT numbers across patient sizes. METHODS Test objects containing a range of iodine and calcium hydroxyapatite concentrations were placed inside 8 torso-shaped water phantoms, ranging in lateral width from 15 to 50 cm, and scanned on a dual-source CT system (Siemens Somatom Force). Single-energy scans were acquired from 70-150 kV in 10 kV increments; dual-energy scans were acquired using 4 energy pairs (low energy: 70, 80, 90, and 100 kV; high energy: 150 kV + 0.6 mm Sn). CTDIvol was matched for all single- and dual-energy scans for a given phantom size. All scans used 128×0.6 mm collimation and were reconstructed with 1-mm thickness at 0.8-mm increment and a medium smooth body kernel. Monoenergetic images were generated using commercial software (syngo Via Dual Energy, VA30). Iodine contrast was calculated as the difference in mean iodine and water CT numbers from respective regions-of-interest in 10 consecutive images. RESULTS CT numbers remained stable as phantom width varied from 15 to 50 cm for all dual-energy data sets (except for at 50 cm using 70/150Sn due to photon starvation effects). Relative to the 15 cm phantom, iodine contrast was within 5.2% of the 70 keV value for phantom sizes up to 45 cm. At 90/150Sn, photon starvation did not occur at 50 cm, and iodine contrast in the 50-cm phantom was within 1.4% of the 15-cm phantom. CONCLUSION Monoenergetic imaging, as implemented in the evaluated commercial system, eliminated the variation in CT numbers due to patient size, and may provide more accurate data for quantitative tasks, including radiation therapy treatment planning. Siemens Healthcare.