Consistent Region Of Interest Research Articles

Towards safer imaging: A comparative study of deep learning-based denoising and iterative reconstruction in intraindividual low-dose CT scans using an in-vivo large animal model

PurposeComputed tomography (CT) scans are a significant source of medically induced radiation exposure. Novel deep learning-based denoising (DLD) algorithms have been shown to enable diagnostic image quality at lower radiation doses than iterative reconstruction (IR) methods. However, most comparative studies employ low-dose simulations due to ethical constraints. We used real intraindividual animal scans to investigate the dose-reduction capabilities of a DLD algorithm in comparison to IR. Materials and methodsFourteen veterinarian-sedated alive pigs underwent 2 CT scans on the same 3rd generation dual-source scanner with two months between each scan. Four additional scans ensued each time, with mAs reduced to 50 %, 25 %, 10 %, and 5 %. All scans were reconstructed ADMIRE levels 2 (IR2) and a novel DLD algorithm, resulting in 280 datasets. Objective image quality (CT numbers stability, noise, and contrast-to-noise ratio) was measured via consistent regions of interest. Three radiologists independently rated all possible dataset combinations per time point for subjective image quality (−1 = inferior, 0 = equal, 1 = superior). The points were averaged for a semiquantitative score, and inter-rater agreement was measured using Spearman's correlation coefficient and adequately corrected mixed-effects modeling analyzed objective and subjective image quality. ResultsNeither dose-reduction nor reconstruction method negatively impacted CT number stability (p > 0.999). In objective image quality assessment, the lowest radiation dose achievable by DLD when comparing noise (p = 0.544) and CNR (p = 0.115) to 100 % IR2 was 25 %. Overall, inter-rater agreement of the subjective image quality ratings was strong (r ≥ 0.69, mean 0.93 ± 0.05, 95 % CI 0.92–0.94; each p < 0.001), and subjective assessments corroborated that DLD at 25 % radiation dose was comparable to 100 % IR2 in image quality, sharpness, and contrast (p ≥ 0.281). ConclusionsThe DLD algorithm can achieve image quality comparable to the standard IR method but with a significant dose reduction of up to 75%. This suggests a promising avenue for lowering patient radiation exposure without sacrificing diagnostic quality.

How Real Are Computed Tomography Low Dose Simulations? An Investigational In-Vivo Large Animal Study

CT low-dose simulation methods have gained significant traction in protocol development, as they lack the risk of increased patient exposure. However, in-vivo validations of low-dose simulations are as uncommon as prospective low-dose image acquisition itself. Therefore, we investigated the extent to which simulated low-dose CT datasets resemble their real-dose counterparts. Fourteen veterinarian-sedated alive pigs underwent three CT scans on the same third generation dual-source scanner with 2 months between each scan. At each time, three additional scans ensued, with mAs reduced to 50%, 25%, and 10%. All scans were reconstructed using wFBP and ADMIRE levels 1-5. Matching low-dose datasets were generated from the 100% scans using reconstruction-based and DICOM-based simulations. Objective image quality (CT numbers stability, noise, and signal-to-noise ratio) was measured via consistent regions of interest. Three radiologists independently rated all possible dataset combinations per time point for subjective image quality (-1=inferior, 0=equal, 1=superior). The points were averaged for a semiquantitative score, and inter-rater-agreement was measured using Spearman's correlation coefficient. A structural similarity index (SSIM) analyzed the voxel-wise similarity of the volumes. Adequately corrected mixed-effects analysis compared objective and subjective image quality. Multiple linear regression with three-way interactions measured the contribution of dose, reconstruction mode, simulation method, and rater to subjective image quality. There were no significant differences between objective and subjective image quality of reconstruction-based and DICOM-based simulation on all dose levels (p≥0.137). However, both simulation methods produced significantly lower objective image quality than real-dose images below 25% mAs due to noise overestimation (p<0.001; SSIM≤89±3). Overall, inter-rater-agreement was strong (r≥0.68, mean 0.93±0.05, 95% CI 0.92-0.94; each p<0.001). In regression analysis, significant decreases in subjective image quality were observed for lower radiation doses (b ≤ -0.387, 95%CI -0.399 to -0.358; p<0.001) but not for reconstruction modes, simulation methods, raters, or three-way interactions (p≥0.103). Simulated low-dose CT datasets are subjectively and objectively indistinguishable from their real-dose counterparts down to 25% mAs, making them an invaluable tool for efficient low-dose protocol development.

AI Denoising Significantly Improves Image Quality in Whole-Body Low-Dose Computed Tomography Staging.

(1) Background: To evaluate the effects of an AI-based denoising post-processing software solution in low-dose whole-body computer tomography (WBCT) stagings; (2) Methods: From 1 January 2019 to 1 January 2021, we retrospectively included biometrically matching melanoma patients with clinically indicated WBCT staging from two scanners. The scans were reconstructed using weighted filtered back-projection (wFBP) and Advanced Modeled Iterative Reconstruction strength 2 (ADMIRE 2) at 100% and simulated 50%, 40%, and 30% radiation doses. Each dataset was post-processed using a novel denoising software solution. Five blinded radiologists independently scored subjective image quality twice with 6 weeks between readings. Inter-rater agreement and intra-rater reliability were determined with an intraclass correlation coefficient (ICC). An adequately corrected mixed-effects analysis was used to compare objective and subjective image quality. Multiple linear regression measured the contribution of “Radiation Dose”, “Scanner”, “Mode”, “Rater”, and “Timepoint” to image quality. Consistent regions of interest (ROI) measured noise for objective image quality; (3) Results: With good–excellent inter-rater agreement and intra-rater reliability (Timepoint 1: ICC ≥ 0.82, 95% CI 0.74–0.88; Timepoint 2: ICC ≥ 0.86, 95% CI 0.80–0.91; Timepoint 1 vs. 2: ICC ≥ 0.84, 95% CI 0.78–0.90; all p ≤ 0.001), subjective image quality deteriorated significantly below 100% for wFBP and ADMIRE 2 but remained good–excellent for the post-processed images, regardless of input (p ≤ 0.002). In regression analysis, significant increases in subjective image quality were only observed for higher radiation doses (≥0.78, 95%CI 0.63–0.93; p < 0.001), as well as for the post-processed images (≥2.88, 95%CI 2.72–3.03, p < 0.001). All post-processed images had significantly lower image noise than their standard counterparts (p < 0.001), with no differences between the post-processed images themselves. (4) Conclusions: The investigated AI post-processing software solution produces diagnostic images as low as 30% of the initial radiation dose (3.13 ± 0.75 mSv), regardless of scanner type or reconstruction method. Therefore, it might help limit patient radiation exposure, especially in the setting of repeated whole-body staging examinations.

Diffusion tensor imaging of the substantia nigra in Parkinson's disease revisited.

To analyze diffusion tensor imaging (DTI) data in the substantia nigra (SN) using a more consistent region of interest (ROI) defined by neuromelanin-sensitive MRI in order to assess Parkinson's disease (PD) related changes in diffusion characteristics in the SN. T1 -weighted and DTI data were obtained in a cohort of 37 subjects (17 control subjects and 20 subjects with PD). The subjects in the PD group were clinically diagnosed PD patients with an average Unified Parkinsonian Disease Rating Scale (UPDRS)-III score of 23.2 ± 9.3. DTI data were analyzed using SN ROIs defined by neuromelanin-sensitive MRI and, for comparison, with ROIs defined on T2 -weighted images (b = 0 images). Compared with control subjects, significantly lower fractional anisotropy was observed in PD in the neuromelanin SN ROI but not in the ROI derived from the T2 -weighted image. This decrease was largest in the rostral and lateral portions of the neuromelanin volume, which were found to have more hypointensity in the T2 -weighted image and, presumably, higher iron content in the PD group. In addition, a larger decrease in fractional anisotropy was seen in the SN region of interest on the side contralateral to the side exhibiting more severe symptoms. These results indicate that the use of neuromelanin sensitive MRI to define the ROI in the SN for analyzing DTI data leads to more significant results, enhancing the robustness of DTI study and DTI based biomarkers of PD. Hum Brain Mapp 37:2547-2556, 2016. © 2016 Wiley Periodicals, Inc.

Local dimension-reduced dynamical spatio-temporal models for resting state network estimation.

To overcome the limitations of independent component analysis (ICA), today’s most popular analysis tool for investigating whole-brain spatial activation in resting state functional magnetic resonance imaging (fMRI), we present a new class of local dimension-reduced dynamical spatio-temporal model which dispenses the independence assumptions that severely limit deeper connectivity descriptions between spatial components. The new method combines novel concepts of group sparsity with contiguity-constrained clusterization to produce physiologically consistent regions of interest in illustrative fMRI data whose causal interactions may then be easily estimated, something impossible under the usual ICA assumptions.

359 Intrasubject Variation in Estimates of Neonatal Intrahepatocellular Lipid using 1H Nmr Spectroscopy

Background and AimsTo date there are no reports of intra-subject variability for intrahepatocellular lipid (IHCL) assessed using 1H NMR spectroscopy in neonates. This study utilized a high-field (3-Tesla) MRI scanner...

SU-E-T-95: Imaging Protocol Investigations with a Fan-Beam Optical CT Scanner for 3D Dosimetry

Purpose: To investigate data acquisition protocol choices and their effects on image noise for a prototype fan-beam optical CT scanner intended for evaluating 3D dosimeters. Methods: The prototype scans dosimeters submerged in a refractive index-matching bath. For all tests, uniform samples of scattering solution (synthetic polymer in water) were housed in cylindrical containers (1-litre, 95 mm diameter) and scanned in a water bath. for reference data (I_o), options evaluated were: i) a projection through the bath, ii) an average projection through a “blank” (water-filled container), and iii) an entire sinogram through a “blank”. Scan parameters considered were: i) temporal delay between I_o and I scans, ii) number of projections per rotation, iii) number of samples averaged per projection, and iv) vertical displacement between I_o and I slices. Images were reconstructed using filtered backprojection. Relative noise (st.dev./mean) in a consistent region of interest was used to quantify image quality. Results: Using a reference scan through a “blank” as opposed to one through only a water bath eliminated major ring artefacts. Furthermore, using a full I_o sinogram through a “blank” instead of an averaged profile considerably reduced noise (22.4% to 3.1%). Minimal increases in noise were seen when delaying scans (3% to 5%, one hour later). Increases in noise were seen with fewer than 360 projections per rotation; with > 360 projections, no significant reductions in noise were seen (up to 1800). No benefit was found by acquiring multiple samples per projection. Significant noise increase was seen when I_o and I slices were displaced by even 1 mm (2.6% to 15.5%). Conclusions: Noise reduction is optimized when comparing scan sinograms ray-by-ray to full sinograms of a “blank”. Main sources of image noise were attributed to variations in the surface quality of the containers used. Funding from the Canadian Institutes of Health Research and the University of Victoria supported portions of this work.

SU‐E‐I‐169: Optimizing Image Quality and Accurate SUVs in Four Different PET Scanners within the Same Institution

Purpose: PET performance measurements were done on 4 PET scanners from 3 manufacturers; GE, Siemens, Philips. This presentation compares the test results and evaluates their impact on clinical studies. Accuracy of SUV determination is compared between the different manufacturers. Methods: Testing of PET performance included quantitative analysis of spatial resolution, count rate, sensitivity, uniformity, and accuracy of corrections. Test results were tabulated for comparison. Additional testing included imaging using the Jaszczak Deluxe PET phantom (Data Spectrum, NC). Testing of SUV accuracy was performed using F‐18 activity (range of SUVs, 2 –20) within cylinders and background to simulate typical tumor/normal tissue ratios. Images were acquired using parameters optimized for clinically interpretable image quality. The maximum SUV for each cylinder was recorded using a consistent region of interest. Curves were developed of SUV versus volume to evaluate partial volume effect. The SUV of the largest diameter cylinder (2.5 mm) was used to develop a curve of expected versus measured SUV values. Results: Significant variability is demonstrated between each PET scanner. The image quality and SUV accuracy varied depending on acquisition and reconstruction protocols. The GE 2D, 3D and Philips Non‐ToF were the most accurate in predicting expected SUV for all activity concentrations. The Siemens HD was 9–17% higher than expected, increasing as SUV value increased. The Philips ToF was 5–19% lower than expected, decreasing as SUV value increased (attached curve of measured versus expected values). Conclusions: While image quality is often improved with PET scanner options, the quantitative component of interpretation, SUV, varies with reconstruction algorithm. This can be of concern when patients are scanned on different PET scanners during the course of their disease. Optimizing image quality and developing an SUV calibration curve has allowed us to maintain confidence in the use of the SUV from different PET systems.

Physiologic predictors of lumbar spine bone mass in neonates.

Dual energy x-ray absorptiometry (DXA) of the lumbar spine (LS) was measured in 201 singleton infants with birth weights from 1152 to 3970 g and gestational ages from 27 to 42 wk. All infants were well and studied at a mean (+/-SD) of 2.1 (+/-1.6) days after birth. There were 75 Caucasian (46 males, 29 females) and 126 African American infants (58 males, 68 females). Scan acquisition of the first to fourth lumbar vertebrae was performed with a single beam whole body scanner (Hologic QDR 1000/W densitometer, Hologic Inc, Waltham, MA, U.S.A.) using the infant spine mode. Scan analysis was performed with software version 4.57Q and consistent region of interest. The SD of difference for duplicate LS scans is <1.4% at a mean bone mineral content (BMC) of 2.14 g. Results show that LS BMC, area, bone mineral density (BMD) increased by approximately 550%, 280% and 180%, respectively, between 27 and 42 wk gestational age. Body mass accounted for about 70% and 55% of the variance in BMC and BMD respectively. In contrast, the infant's length appears to be the best determinant of LS area and accounts for about 75% of the variance in LS area. Race, gender or season has little or no effect on LS bone mass. There was progressive increase in BMC and area from first to fourth lumbar vertebra but BMD was significantly higher only at the fourth lumbar vertebra. We conclude that DXA LS can be performed even in small preterm infants. Its excellent precision, low radiation exposure and rapid scan acquisition offers promise as a useful tool for widespread use in pediatrics. Our data may be used as a basis for further studies in physiologic and pathologic situations that may affect bone mineralization in infants.