Magnitude Of Noise Research Articles

Advancements in the performance of nanopore sensing and its implications towards the identification of biomarkers.

Rapid and selective identification of biomarkers is a coveted need recapitulated by the ongoing pandemic that pierced the very fabric of human health and wellbeing. Moreover, there is a constant interest and demand for portable technologies that can meet the above-mentioned. Nanopores—a nanoscale aperture through an otherwise impermeable membrane—have demonstrated tremendous potential to this extent, which is further facilitated by their ability to detect and characterize a wide range of bio/synthetic molecules and particles (e.g., DNA, proteins, viruses). Fundamentally, they register analyte-specific information through resistive (or conductive) pulses as an analyte transverses the pore in response to an applied electric field. However, the presence of solid-state nanopores (SSNs) in the commercial space is still meager. Features such as stability, lifetime, resilience against clogging, pore noise, and signal magnitude are key for the progression of SSNs beyond a laboratory setup. The membrane and solution chemistry during fabrication was found to play an integral role where pores remained open for hours surpassing most recorded statistics concerning the five before-mentioned features. A wide range of solution chemistries (and breakdown conditions) was investigated to understand the chemistry of fabrication and its impact on the pore performance to push the existing boundaries of the technology. Afterward, to uncover finer molecular-scale details, especially those concerning smaller and fast-moving proteins, higher bandwidths (i.e., 10 MHz) were used. This enables the detection of events that last a few hundred nanoseconds which were previously impossible due to electronic limitations. With such advancements, the enormity of available data would enable the seamless coupling of machine learning for the identification of targets in complex mixtures such as serum. Thus, such developments would enable SSNs to perform well under real-world conditions and complex samples.

Task-Based Image Quality Assessment Comparing Classical and Iterative Cone Beam CT Images on Halcyon®.

Despite the development of iterative reconstruction (IR) in diagnostic imaging, CBCT are generally reconstructed with filtered back projection (FBP) in radiotherapy. Varian medical systems, recently released with their latest Halcyon® V2.0 accelerator, a new IR algorithm for CBCT reconstruction. To assess the image quality of radiotherapy CBCT images reconstructed with FBP and an IR algorithm. Three CBCT acquisition modes (head, thorax and pelvis large) available on a Halcyon® were assessed. Five acquisitions were performed for all modes on an image quality phantom and reconstructed with FBP and IR. Task-based image quality assessment was performed with noise power spectrum (NPS), task-based transfer function (TTF) and detectability index (d'). To illustrate the image quality obtained with both reconstruction types, CBCT acquisitions were made on 6 patients. The noise magnitude and the spatial frequency of the NPS peak was lower with IR than with FBP for all modes. For all low and high-contrast inserts, the values for TTF at 50% were higher with IR than with FBP. For all inserts and all modes, the contrast values were similar with FBP and IR. For all low and high-contrast simulated lesions, d' values were higher with IR than with FBP for all modes. These results were also found on the 6 patients where the images were less noisy but smoother with IR-CBCT. Using the IR algorithm for CBCT images in radiotherapy improve image quality and thus could increase the accuracy of online registration and limit positioning errors during processing.

Information-transfer characteristics in network motifs

Information processing in biological systems is realized by the appropriate transmission of information flows over complex networks, such as gene regulatory, signal transduction, and neural networks. These information flows are affected by the input-signal characteristics and structural properties of network systems, such as the network topology, regulation rules, and intrinsic and environmental noise. Many biological networks frequently include several typical patterns called network motifs, which are considered to play important roles in biological functions. However, their information-theoretic properties, particularly the dependence of the information flows in each network on the input signal, remain poorly understood. In our previous study [Mori and Okada, Phys. Rev. Res. 2, 043432 (2020)], we developed a graphical expansion method to describe transfer entropy (TE), a measure of information flow, in Boolean networks in terms of multiple information pathways. There, the input signal was limited to a simple case, and the effect of the input-signal characteristics on TE was not clarified. In this paper, we improve our method to render it applicable to Boolean networks that receive input signals with arbitrary stochastic characteristics. Our formula expresses how TE is determined by the input-signal characteristics, the assignment of Boolean functions, and the noise magnitude. We find that, in both positive and negative feedback loops, TE hardly depends on the signal timescale. In contrast, coherent and incoherent feedforward loops show low- and high-pass filtering properties, respectively, for a time-varying signal, which is consistent with previous reports. The emergence of either low- or high-pass filtering is determined by the Fourier components of the Boolean functions on specific pathways transmitting information flows. Thus our formula reveals the mechanism of information transfer in network motifs and provides insights into the origin of information processing in biological networks.

Controlling thermal noise in phase-field simulation for nucleation during isothermal solidification of a pure metal

It is currently challenging to quantitatively describe the nucleation behavior using the phase-field model because it depends on the computational grid size. In this study, the grid size dependence of nucleation behavior during isothermal solidification in pure Fe was investigated by changing the magnitude of thermal noise inducing the nucleation. To remove such dependence, a neural network was utilized to compute the magnitude of thermal noise needed to describe the desired nucleation rate for a given grid size and degree of undercooling. It was found that the nucleation rate can be controlled using the present approach, and the nucleation rates in pure Fe obtained by molecular dynamics simulations and experiments could be accurately reproduced.

Comparison of image quality of two versions of deep-learning image reconstruction algorithm on a rapid kV-switching CT: a phantom study

BackgroundTo assess the impact of the new version of a deep learning (DL) spectral reconstruction on image quality of virtual monoenergetic images (VMIs) for contrast-enhanced abdominal computed tomography in the rapid kV-switching platform.MethodsTwo phantoms were scanned with a rapid kV-switching CT using abdomen-pelvic CT examination parameters at dose of 12.6 mGy. Images were reconstructed using two versions of DL spectral reconstruction algorithms (DLSR V1 and V2) for three reconstruction levels. The noise power spectrum (NSP) and task-based transfer function at 50% (TTF50) were computed at 40/50/60/70 keV. A detectability index (d') was calculated for enhanced lesions at low iodine concentrations: 2, 1, and 0.5 mg/mL.ResultsThe noise magnitude was significantly lower with DLSR V2 compared to DLSR V1 for energy levels between 40 and 60 keV by -36.5% ± 1.4% (mean ± standard deviation) for the standard level. The average NPS frequencies increased significantly with DLSR V2 by 23.7% ± 4.2% for the standard level. The highest difference in TTF50 was observed at the mild level with a significant increase of 61.7% ± 11.8% over 40−60 keV energy with DLSR V2. The d' values were significantly higher for DLSR V2 versus DLSR V1.ConclusionsThe DLSR V2 improves image quality and detectability of low iodine concentrations in VMIs compared to DLSR V1. This suggests a great potential of DLSR V2 to reduce iodined contrast doses.

Assessment of the impact of noise magnitude and bandwidth variations on a probabilistic inversion of reflection seismic data

AbstractAccounting for an accurate noise model is essential when dealing with real data, which are noisy due to the effect of environmental noise, failures and limitations in data acquisition and processing. Quantifying the noise model is a challenge for practitioners in formulating an inverse problem, and usually, a simple Gaussian noise model is assumed as a white noise model. Here we propose a pragmatic approach to use an estimated seismic wavelet to capture the correlated noise model (coloured noise) for the processed reflection seismic data. We assess the proposed method through a direct inversion of post‐stack seismic data associated with a carbonate reservoir of an oil field in southwest Iran to porosity, using a probabilistic sampling‐based inversion algorithm. In the probabilistic formulation of the inverse problem, we assume eight different noise models with varying bandwidth and magnitude and investigate the corresponding posterior statistics. The results indicate that if the correlated nature of the noise samples is ignored in the noise covariance matrix, some unrealistic features are generated in porosity realizations. In addition, if the noise magnitude is underestimated, the inversion algorithm overfits the data and generates a biased model with low uncertainty. Furthermore, by considering an imperfect bandwidth for the noise model, the error is propagated to the posterior realizations. Assuming the correlated noise in a probabilistic inversion resolves these issues significantly. Therefore, for inverting real seismic data where the estimation of the magnitude and correlations of the noise is not straightforward, the wavelet, which is estimated from the real seismic data, provides a good proxy for describing the correlation of the noise samples or equivalently the bandwidth of the noise model. In addition, it might be better to overestimate the noise magnitude than to underestimate it. This is true especially for an uncorrelated noise model and to a lesser degree also for the correlated noise model.

Hepatic lesion detectability in abdomen computed tomography: Investigation in low kVp single energy and low keV virtual monochromatic images generated from dual energy computed tomography using task-based image quality assessment in phantom

Background: Low tube potential single-energy (SE) and virtual monochromatic image (VMI) dual energy (DE) abdomen CT images both improve the hepatic lesion detection by increasing the liver lesion contrast on images. Objectives: To study the performance of low kVp single energy and low keV virtual monochromatic images (VMI) generated from dual energy acquisition to detect the hepatic lesion on abdominal CT imaging in phantom. Materials and methods: The anthropomorphic liver with nodule inserted phantom with extension rings simulating the small, medium, and large patient was scanned under the SECT acquisition by varying the kVp from 70-120 kVp and for DECT acquisitions, three kVp combinations (80/-,90/-, and 100/Sn150-kVp). The series of 40-,50-,60-, and 70-keV VMI were generated from DECT data set. All images were used to assess the task-based image quality; task transfer function (TTF), noise power spectrum (NPS), and detectability index (d’) with the diagnostic task to detect 15 mm diameter hyperattenuating hepatic lesion. Results: The result showed that the TTF was higher in low kVp SECT while the lesion contrast was higher at low keV VMI -DECT. The noise magnitude remained constant for all kVp values in SECT, but it was dramatically increased as decreased the energy level from 70- to 40-keV in VMI-DECT. The fav of NPS shifted to higher frequency when increasing the kVp and when increasing the energy level of VMI. The obtained d’ was highest in low kVp SECT at 70-or 80-kVp. Conclusion: The low kVp SECT provided the highest d’ than that in low keV VM image from DECT in all phantom sizes where the highest d’ was found at 70 kVp-SECT for small and medium phantom and at 80 kVp for large phantom. For SECT, reduced the kVp to 70- or 80-kVp improved the detectability index. The kVp combination in DECT impacts to the d’ of VMI; at small phantom, the highest d’ for each keV VMI was found at 80/Sn150 kVp acquisition and at larger phantom size, higher kVp on tube “A” is more favored.

Thermorefringent noise in crystalline optical materials

Crystalline materials are increasingly employed to construct precision optical instruments because of their reduced mechanical dissipation and consequent reduction of thermal Brownian noise. However, the anisotropy of the crystalline state implies a fundamental source of thermal noise; depolarization induced by thermal fluctuations of its birefringence. We establish the theory of this effect, which is a generalization of prior treatments of thermo-optic noises in amorphous materials. This theory---in conjunction with poorly understood anisotropic thermal stress coefficients of crystalline coatings---predict that thermo-refringent noise in crystalline mirror coatings may be lurking within an order of magnitude of Brownian noise (below 100 Hz). Thus, in order to appreciate the full promise of crystalline optical materials, a more precise understanding of their anisotropic material constants is necessary. Barring that, we elucidate measurement techniques that can affect partial coherent cancellation of thermorefringent noise. In passing, our general formalism also predicts the existence of thermal beam-pointing noise.

Quantifying noise in survey expectations

Expectations affect economic decisions, and inaccurate expectations are costly. Expectations can be wrong due to either bias (systematic mistakes) or noise (unsystematic mistakes). We develop a framework for quantifying the level of noise in survey expectations. The method is based on the insight that theoretical models of expectation formation predict a factor structure for individual expectations. Using data from professional forecasters, we find that the magnitude of noise is large (10%–30% of forecast MSE) and comparable to bias. We illustrate how our estimates can be applied to calibrate models with incomplete information and bound the effects of measurement error.

Secure State Estimation against Sparse Attacks on a Time-varying Set of Sensors

This paper studies the problem of secure state estimation of a linear time-invariant (LTI) system with bounded noise in the presence of sparse attacks on an unknown, time-varying set of sensors. At each time, the attacker has the freedom to choose an arbitrary set of no more than p sensors and manipulate their measurements without restraint. To this end, we propose a secure state estimation scheme and guarantee a bounded estimation error irrespective of the attack signals subject to 2p-sparse observability and a mild, technical assumption that the system matrix has no degenerate eigenvalues. The proposed scheme comprises a design of decentralized observers for each sensor based on the local observable subspace decomposition. At each time step, the local estimates of sensors are fused by a median operator to obtain a secure estimation, which is then followed by a local detection-and-resetting process of the decentralized observers. The estimation error is shown to be upper-bounded by a constant which is determined only by the system parameters and noise magnitudes. Moreover, we design the detector threshold to ensure that the benign sensors never trigger the detector. The efficacy of the proposed algorithm is demonstrated by its application on a benchmark example of IEEE 14-bus system. We show that our proposed scheme can effectively tolerate sparse attacks on an unknown set of sensors, ensuring a bounded estimation error and effectively detecting and resetting the attacked sensors.

Technical performance of a dual-energy CT system with a novel deep-learning based reconstruction process: Evaluation using an abdomen protocol.

A new tube voltage-switching dual-energy (DE) CT system using a novel deep-learning based reconstruction process has been introduced. Characterizing the performance of this DE approach can help demonstrate its benefits and potential drawbacks. To evaluate the technical performance of a novel DECT system and compare it to that of standard single-kV CT and a rotate/rotate DECT, for abdominal imaging. DE and single-kV images of four different phantoms were acquired on a kV-switching DECT system, and on a rotate/rotate DECT. The dose for the acquisitions of each phantom was set to that selected for the kV-switching DE mode by the automatic tube current modulation (ATCM) at manufacturer-recommended settings. The dose that the ATCM would have selected in single-kV mode was also recorded. Virtual monochromatic images (VMIs) from 40 to 130keV, as well as iodine maps, were reconstructed from the DE data. Single-kV images, acquired at 120kV, were reconstructed using body hybrid iterative reconstruction. All reconstructions were made at 0.5mm section thickness. Task transfer functions (TTFs) were determined for a Teflon and LDPE rod. Noise magnitude (SD), and noise power spectrum (NPS) were calculated using 240 and 320mm diameter water phantoms. Iodine quantification accuracy and contrast-to-noise ratios (CNRs) relative to water for 2, 5, 10, and 15mg I/ml were determined using a multi-energy CT (MECT) phantom. Low-contrast visibility was determined and the presence of beam-hardening artifacts and inhomogeneities were evaluated. The TTFs of the kV-switching DE VMIs were higher than that of the single-kV images for Teflon (20% TTF: 6.8 lp/cm at 40keV, 6.2 lp/cm for single-kV), while for LDPE the DE TTFs at 70keV and above were equivalent or higher than the single-kV TTF. All TTFs of the kV-switching DECT were higher than for the rotate/rotate DECT. The SD was lowest in the 70keV VMI (12.0 HU), which was lower than that of single-kV (18.3 HU). The average NPS frequency varied between 2.3 lp/cm and 4.2 lp/cm for the kV-switching VMIs and was 2.2 lp/cm for single-kV. The error in iodine quantification was at maximum 1mg I/ml (at 5mg I/ml). The highest CNR for all iodine concentrations was at 60keV, 2.5 times higher than the CNR for single-kV. At 70-90keV, the number of visible low contrast objects was comparable to that in single-kV, while other VMIs showed fewer objects. At manufacturer-recommended ATCM settings, the CTDIvol for the DE acquisitions of the water and MECT phantoms were 12.6 and 15.4 mGy, respectively, and higher than that for single-kV. The 70keV VMI had less severe beam hardening artifacts than single-kV images. Hyper- and hypo-dense blotches may appear in VMIs when object attenuation exceeds manufacturer recommended limits. At manufacturer-recommended ATCM settings for abdominal imaging, this DE implementation results in higher CTDIvol compared to single-kV acquisitions. However, it can create sharper, lower noise VMIs with up to 2.5 times higher iodine CNR compared to single-kV images acquired at the same dose.

An improved stacking filtering for extracting the common-mode errors on GNSS coordinate time series in Shanxi

The presence of common mode error (CME) in the coordinate displacement time series of the Global Navigation Satellite System (GNSS) affects geophysical studies using GNSS observations. In order to investigate the effect of CME on the time series in GNSS networks in Shanxi, this paper proposes an improved superposition filtering method by introducing single-day solution accuracy, correlation coefficient, and spherical distance between stations as weights. The filtering effect is evaluated using the GNSS data in Shanxi. By using the improved stacking filtering method, the root mean square (RMS) values for N, E, U are reduced by approximately 27.8%, 29.0%, and 46.0%, respectively. And compared to the traditional stacking filter, our improved method can achieve better results with CME extraction. We investigate the CME spatial-temporal characteristics and its relationship with environmental loading. The results show that the CME between stations decreases as the distance between stations increases. In addition, we analyze the effect of CME on the noise component and velocity estimates. Results show that removing the CME refines the velocity and leads to a significant reduction in the magnitude of noise, indicating that the CME is dominated by the flicker noise in Shanxi Province.

Evaluation of Metal–Semiconductor Contact Quality: Correlation of 1/f Noise and Nonlinearity

Measurement of 1/f noise emerges as a valuable tool for contact quality evaluation, which simultaneously provides information about the 1/f noise magnitude and nonlinearity of the contact. This article investigates experimentally the correlation of current–voltage characteristics nonlinearity and 1/f noise for the metal–semiconductor–metal structure. It was found that the nonlinearity and 1/f noise originate from metal–semiconductor contact. The relative 1/f noise of weak-nonlinear, roughly ohmic contact, is proportional to the parameter describing the nonlinearity. Further increase of contact nonlinearity, by decreasing the temperature, substantially increases the 1/f noise magnitude. The efforts to develop low-noise semiconductor devices should be focused on avoiding even weak nonlinearity of contacts because it results in substantial 1/f noise.

Hyperparameter estimation using a resolution matrix for Bayesian sensing

Bayesian image processing has progressively increased in importance in various fields of the natural sciences. It utilizes prior knowledge and forward models of the observational processes through Bayes’ theorem, enabling the accurate estimation of model parameters that represent the physical quantities of the target. Moreover, using hyperparameter estimation, we can determine the hidden physical parameters that govern the processes in and the structure of the target and sensing systems, such as the spatial continuity of the model parameters and the magnitude of the observational noise. Such a general framework, which uses Bayesian estimation to understand the essential physics of a target system, can be called ‘Bayesian sensing’. This paper discusses the physical meaning of and the mechanism underlying Bayesian sensing using the concept of resolution in the spatial-inversion problem. The spatial resolution of the model parameters can be mapped using a resolution matrix, more rigorously, a model resolution matrix defined as a linear mapping from the true model parameters to the recovered model parameters. We formulate the resolution matrix for Bayesian image processing and also show that in terms of resolution, the optimal hyperparameters are obtained from internally consistent equations that connect the estimated optimal hyperparameters with the actual ones calculated from the estimated model parameters. In addition, we show the equivalence of the internally consistent equations to the expectation-maximization (EM) algorithm and formulate the confidence intervals for the estimated hyperparameters, which indicate the general convergence of the hyperparameter estimates. We also show the effectiveness of the proposed method by performing synthetic numerical tests for two inversion-problem settings: linear travel-time seismic tomography and image deblurring. The resulting equations can contribute to understanding the hidden physical processes in and the structure of the target and observation systems for various problems.

CT image quality evaluation in the age of deep learning: trade-off between functionality and fidelity.

To quantitatively compare DLIR and ASiR-V with realistic anatomical images. CT scans of an anthropomorphic phantom were acquired using three routine protocols (brain, chest, and abdomen) at four dose levels, with images reconstructed at five levels of ASiR-V and three levels of DLIR. Noise power spectrum (NPS) was estimated using a difference image by subtracting two matching images from repeated scans. Using the max-dose FBP reconstruction as the ground truth, the structure similarity index (SSIM) and gradient magnitude (GM) of difference images were evaluated. Image noise magnitude (σ), frequency location of the NPS peak (fpeak), mean SSIM (MSSIM), and mean GM (MGM) were used as quantitative metrics to compare image quality, for each anatomical region, protocol, algorithm, dose level, and slice thickness. Image noise had a strong (R2 > 0.99) power law relationship with dose for all algorithms. For the abdomen and chest, fpeak shifted from 0.3 (FBP) down to 0.15 mm-1 (ASiR-V 100%) with increasing ASiR-V strength but remained 0.3 mm-1 for all DLIR levels. fpeak shifted down for the brain protocol with increasing DLIR levels. Three levels of DLIR produced similar image noise levels as ASiR-V 40%, 80%, and 100%, respectively. DLIR had lower MSSIM but higher MGM than ASiR-V while matching imaging noise. Compared to ASiR-V, DLIR presents trade-offs between functionality and fidelity: it has a noise texture closer to FBP and more edge enhancement, but reduced structure similarity. These trade-offs and unique protocol-dependent behaviors of DLIR should be considered during clinical implementation and deployment. • DLIR reconstructed images demonstrate closer noise texture and lower structure similarity to FBP while producing equivalent noise levels comparable to ASiR-V. • DLIR has additional edge enhancement as compared to ASiR-V. • DLIR has unique protocol-dependent behaviors that should be considered for clinical implementation.

Derivation of New Daily Rainfall Values from TAO 1-Min Rain Gauge Data

Abstract Daily rainfall accumulation estimates have been derived from 1-min volume data collected via self-syphon rain gauges deployed in the Tropical Atmosphere–Ocean (TAO) array of oceanographic buoys. The underlying high-resolution volume data were obtained directly from the Global Tropical Moored Buoy Array (GTMBA) Project Office of NOAA/Pacific Marine Environmental Laboratory. The derived accumulations have been incorporated into the Pacific Rainfall (PACRAIN) database as estimated daily values to augment existing sea level oceanic rainfall records gathered using traditional rain gauges. They have also been included in the PACRAIN historical, monthly gridded rainfall product. The methodology presented, which employs differencing of least squares–regressed sensor levels about 0000 UTC and rain gauge syphon events, is shown to offer improved error characteristics over the methodology used to compute previously published GTMBA rain rates. In particular, the PACRAIN method yields larger coefficients of determination and smaller standard errors than the duplicated GTMBA method when applied to synthetic rainfall data with noise magnitude and decorrelation times encompassing those observed in the real 1-min data. These results are shown to be consistent with mathematical expectations. Sources of instrument and catchment errors, as well as evaporation, are discussed in the context of their potential effects on accumulation estimates for periods of a day or longer. Significance Statement In this paper, we describe the derivation of daily rainfall amounts from raw rain gauge data obtained from buoy-mounted rain gauges. These new accumulation estimates expand the store of rainfall estimates from locations approximating the open-ocean conditions of the tropical Pacific Ocean. The derivation technique we describe exhibits better performance than the method used to generate previously published estimates using the same dataset.

Impact of Iterative Bilateral Filtering on the Noise Power Spectrum of Computed Tomography Images

A bilateral filter is a non-linear denoising algorithm that can reduce noise while preserving the edges. This study explores the characteristics of a bilateral filter in changing the noise and texture within computed tomography (CT) images in an iterative implementation. We collected images of a homogeneous Neusoft phantom scanned with tube currents of 77, 154, and 231 mAs. The images for each tube current were filtered five times with a configuration of sigma space (σd) = 2 pixels, sigma intensity (σr) = noise level, and a kernel of 5 × 5 pixels. To observe the noise texture in each filter iteration, the noise power spectrum (NPS) was obtained for the five slices of each dataset and averaged to generate a stable curve. The modulation-transfer function (MTF) was also measured from the original and the filtered images. Tests on an anthropomorphic phantom image were carried out to observe their impact on clinical scenarios. Noise measurements and visual observations of edge sharpness were performed on this image. Our results showed that the bilateral filter was effective in suppressing noise at high frequencies, which is confirmed by the sloping NPS curve for different tube currents. The peak frequency was shifted from about 0.2 to about 0.1 mm−1 for all tube currents, and the noise magnitude was reduced by more than 50% compared to the original images. The spatial resolution does not change with the number of iterations of the filter, which is confirmed by the constant values of MTF50 and MTF10. The test results on the anthropomorphic phantom image show a similar pattern, with noise reduced by up to 60% and object edges remaining sharp.

Construction and Application of Seismic Real-Time Data Analysis Model Based on GPS Monitoring

GPS (global position system) monitoring has great theoretical and practical significance for understanding regional crustal deformation and disaster prevention and mitigation. Based on the GPS system differential model, this paper builds a seismic real-time data analysis model and discusses and analyzes the GNSS (global navigation satellite system) single-system single-difference model, single-system double-difference model, and dual-system double-difference model. When the model is processing the data of the GNSS dual-satellite positioning system, the proposed short-baseline ambiguity solution is studied to a certain extent and solves the problems that other common methods can only detect large cycle slips or require dual-frequency data. During the simulation process, through the study of the GPS coordinate time series, the stability of the site, the periodic change of the site, the movement law of the plate, and other information can be analyzed. The combined method is used to estimate the station noise, coordinate time series parameters, and the station velocity, and then the influence of the common mode error on the station noise, parameter estimation, and station velocity estimation is studied. The experimental results show that the average magnitude of white noise and colored noise decreases by 31.61% and 42.28%, respectively, which effectively reduces the uncertainty of parameter estimation and the uncertainty of station velocity estimation.

Twilight of Human Judgment

Twilight of Human Judgment

Comparative assessment of noise properties for two deep learning CT image reconstruction techniques and filtered back projection

Two deep learning image reconstruction (DLIR) techniques from two different computed tomography (CT) vendors have recently been introduced into clinical practice. To characterize the noise properties of two DLIR techniques with different training methods, using a phantom containing a simple uniform and a complex non-uniform region. A water-bath phantom with a diameter of 300 mm was used as a base phantom. A textured phantom with a diameter of 128 mm, which was made of two materials, one equivalent to water and the other being 12 mg/ml diluted iodine, irregularly mixed to create a complex texture (non-uniform region), was placed in the base phantom. Thirty repeated phantom scans were performed using two CT scanners (Revolution CT with Apex Edition, GE Healthcare; Aquilion One PRISM Edition, Canon Medical Systems) at two dose levels (CT dose index: 5 and 15 mGy). Images were reconstructed with each CT system's filtered back projection (FBP) and DLIR [TrueFidelity (TF), GE Healthcare; Advanced intelligent Clear-IQ Engine Body Sharp (AC), Canon Medical Systems] for three process strengths. For basic characteristics of noise, the standard deviation (SD) and noise power spectrum (NPS) were measured for the uniform (water) region. A noise magnitude map was generated by calculating the inter-image SD at each pixel position across the 30 images. Then, a noise reduction map (NRM), which visualizes the relative differences in noise magnitude between FBP and DLIR, was calculated. The NRM values ranged from 0.0 to 1.0. A low NRM value represents a less aggressive noise reduction. The histograms of the NRM value were analyzed for the uniform and non-uniform regions. The reduction in noise magnitude compared with FBP tended to be greater with AC (45%-85%) than with TF (32%-65%). The average NPS frequencies of TF and AC were almost comparable to those of FBP, except for the low-dose condition and the high noise reduction strength for AC. The NRM values of TF and AC were higher in the uniform region than in the non-uniform region. In the non-uniform region, TF's average NRM values (0.21-0.48) tended to be lower than AC's (0.39-0.78). The histograms for TF showed a small overlap between the uniform and the non-uniform regions; in contrast, those for AC showed a greater overlap. This difference seems to indicate that TF processes the uniform and non-uniform regions more differently than AC does. This study has revealed a distinct difference in characteristics between the two DLIR techniques: TF tends to offer less aggressive noise reduction in non-uniform regions and preserve the original signals, whereas AC tends to prioritize noise filtering over edge-preservation, especially at the low-dose condition and with the high noise reduction strength.