Low Noise Level Research Articles

Comparison of deep learning-based denoising methods in cardiac SPECT

BackgroundMyocardial perfusion SPECT (MPS) images often suffer from artefacts caused by low-count statistics. Poor-quality images can lead to misinterpretations of perfusion defects. Deep learning (DL)-based methods have been proposed to overcome the noise artefacts. The aim of this study was to investigate the differences among several DL denoising models.MethodsConvolution neural network (CNN), residual neural network (RES), UNET and conditional generative adversarial neural network (cGAN) were generated and trained using ordered subsets expectation maximization (OSEM) reconstructed MPS studies acquired with full, half, three-eighths and quarter acquisition time. All DL methods were compared against each other and also against images without DL-based denoising. Comparisons were made using half and quarter time acquisition data. The methods were evaluated in terms of noise level (coefficient of variation of counts, CoV), structural similarity index measure (SSIM) in the myocardium of normal patients and receiver operating characteristic (ROC) analysis of realistic artificial perfusion defects inserted into normal MPS scans. Total perfusion deficit scores were used as observer rating for the presence of a perfusion defect.ResultsAll the DL denoising methods tested provided statistically significantly lower noise level than OSEM without DL-based denoising with the same acquisition time. CoV of the myocardium counts with the different DL noising methods was on average 7% (CNN), 8% (RES), 7% (UNET) and 14% (cGAN) lower than with OSEM. All DL methods also outperformed full time OSEM without DL-based denoising in terms of noise level with both half and quarter acquisition time, but this difference was not statistically significant. cGAN had the lowest CoV of the DL methods at all noise levels. Image quality and polar map uniformity of DL-denoised images were also better than reduced acquisition time OSEM’s. SSIM of the reduced acquisition time OSEM was overall higher than with the DL methods. The defect detection performance of full time OSEM measured as area under the ROC curve (AUC) was on average 0.97. Half time OSEM, CNN, RES and UNET provided equal or nearly equal AUC. However, with quarter time data CNN, RES and UNET had an average AUC of 0.93, which was lower than full time OSEM’s AUC, but equal to quarter acquisition time OSEM. cGAN did not achieve the defect detection performance of the other DL methods. Its average AUC with half time data was 0.94 and 0.91 with quarter time data.ConclusionsDL-based denoising effectively improved noise level with slightly lower perfusion defect detection performance than full time reconstruction. cGAN achieved the lowest noise level, but at the same time the poorest defect detection performance among the studied DL methods.

Lowering the vertical component seismic noise-floor in the normal-mode band by frequency-dependent air-pressure corrections

SUMMARY Contrary to popular belief in seismology, the vertical component seismic background level of Earth’s body in the normal-mode band is lower than current low-noise models would indicate. With a correction of atmospherically induced disturbances the low-noise level of the vertical component of the STS-1 at Black Forest Observatory (BFO) is reduced below the Peterson low noise model (NLNM). We demonstrate this with a moving window analysis (MWA) of 10 yr of data in the normal-mode band between 0.2 and 2 mHz. The 5th percentile low-noise level for the data initially is at the level of the NLNM. We lower this by 8 dB at the frequency of 0S2 (0.31 mHz) by partially correcting the seismometer recording in the time domain for disturbances caused by mass fluctuations in the atmosphere. The disturbances are estimated by the frequency dependent ‘improved Bouguer plate model’ (IBPM, defined by asymptotic admittance αDC and notch-frequency fn) from a recording of local air-pressure. In more than 90 per cent of the time windows of the MWA the signal level is lowered when using a priori set IBPM-parameters ($\alpha _{\text{DC}}= -3.5\, {\rm nm\, s}^{-2}\, {\rm hPa}^{-1}$, $f_{\text{n}}=2.25\, {\rm mHz}$). In 50 per cent of the time windows the background level at the frequency of 0S2 is reduced by at least 6 dB. With a priori set IBPM-parameters the level of the 1st percentile is lowered by up to 16 dB for windows of $1\, {\rm d}$ length and by 13 dB for windows of $7\, {\rm d}$ length. Overall, there is little additional benefit from optimizing the IBPM-parameters per time window. If optimized per time window, the parameters show a slight annual variation which apparently compensates for systematic deviations of the dynamics of the atmosphere from the simple considerations of the IBPM-model. By application of the correction procedure and with seismic sensors of sufficiently low self-noise, the range of observations for signals from Earth’s body can be extended to below current low-noise models at frequencies below 2 mHz.

Parallel Factorization to Implement Group Analysis in Brain Networks Estimation.

When dealing with complex functional brain networks, group analysis still represents an open issue. In this paper, we investigated the potential of an innovative approach based on PARAllel FActorization (PARAFAC) for the extraction of the grand average connectivity matrices from both simulated and real datasets. The PARAFAC approach was solved using three different numbers of rank-one tensors (PAR-FACT). Synthetic data were parametrized according to different levels of three parameters: network dimension (NODES), number of observations (SAMPLE-SIZE), and noise (SWAP-CON) in order to investigate the way they affect the grand average estimation. PARAFAC was then tested on a real connectivity dataset, derived from EEG data of 17 healthy subjects performing wrist extension with left and right hand separately. Findings on both synthetic and real data revealed the potential of the PARAFAC algorithm as a useful tool for grand average extraction. As expected, the best performances in terms of FPR, FNR, and AUC were achieved for great values of sample size and low noise level. A crucial role has been revealed for the PAR-FACT parameter, revealing that an increase in the number of rank-one tensors solving the PARAFAC problem leads to an increase in FPR values and, thus, to a worse grand average estimation.

Listening effort for sentence comprehension in noisy classroom: the mediating role of linguistic factors, inhibitory control and noise sensitivity

Children learn in noisy classrooms, where often the main masker is their own babble. Even when the acoustic conditions are favorable (i.e. low noise levels), differences in individual performance and listening effort in complex academic tasks are observed. Personal characteristics such as linguistic and cognitive skills, and sensitivity to noise have been reported as factors supporting students' performance. Moreover, the Framework for Understanding Effortful Listening (FUEL) postulates that the additional individual dimension of children's motivation should be considered when evaluating listening effort besides task-related cognitive demands. This study aims to explore the interplay of the above-mentioned individual factors for primary school children (N=120, grades 3 to 5) doing a sentence comprehension task in a two-talkers background noise. Data on both accuracy and response time, as well as self-ratings of effort and motivation were acquired. In addition, inhibitory control, linguistic competences and self-ratings of noise sensitivity were measured in quiet. Results first highlight how acoustic conditions and linguistic competencies influence child's motivation, and then show how the child's inhibitory control and noise sensitivity mediate behavioral and subjective effort. Thus, individual factors shall be taken into consideration when evaluating the effect of classroom acoustics on the performance in academic tasks.

Connecting Contemporary Trauma Care to Florence Nightingale's Visionary Work.

The impact of Florence Nightingale's visionary work continues to influence the delivery of nursing care in the contemporary emergency department (ED). Her foundational work in the Crimean War resulted in data-based recommendations for using the environment to promote healing and wellness among sick and wounded British soldiers. She advocated for attention to environmental details, including ventilation, air, warmth, drainage, cleanliness, natural light, and low noise levels. These important environmental concepts play a significant role in the nursing management of trauma patients in today's ED. This article features an application of Nightingale's environmental concepts to a trauma patient case exemplar and demonstrates the enduring impact of her work for trauma patients who receive care in the ED.

Optimal Electric Bus Scheduling with Multiple Vehicle Types Considering Bus Crowding Degree

Electric buses are environmentally friendly with the features of zero emissions and low noise levels. Therefore, it is an inevitable trend to replace traditional fuel buses with electric buses. In the urban transportation system, the most common electric bus scheduling mode is single-vehicle-type scheduling, where crowdedness in peak hours could impair the bus service level. In contrast, a high vacancy rate in off-peak hours might lead to a waste of resources. To balance efficiency and the bus service level, this paper built an electric bus scheduling model with multiple vehicle types distinguished by vehicle configurations, such as bus capacity, battery capacity, purchase cost, charging power, and electricity consumption rate. The optimization objective was to minimize the bus system costs, including bus depreciation costs, charging costs, and congestion time costs. Then, a genetic algorithm was designed to solve the model, and the model is verified by a case study of one bus line in Nanjing, China. Compared with the single-vehicle-type scheduling schemes, including large and small vehicle types, the experimental results showed that the proposed model can reduce total bus system costs by 3.52% and 6.85%, respectively. The research results can provide references for bus companies to formulate driving plans and configure vehicles.

Association between aircraft noise levels and deprivation

There is limited evidence on socioeconomic distribution of noise exposure. Noise data (Lden, Laeq24, Lnight, Levening, and Lday) were available for London Heathrow airport for 2014-18. These were linked with different measures of deprivation: the Carstairs deprivation index (UK Census-derived), fuel poverty rate and the avoidable death rate. Using Carstairs, mean noise levels were slightly higher in more deprived areas for most noise metrics, especially for Lnight (~2dB between least and most deprived Carstairs quintiles). However, Leve had slightly lower mean noise levels in the most deprived quintile (~0.5dB difference). A clear pattern of higher mean noise levels with deprivation was seen using avoidable death rates (~4dB difference between lowest and highest quintile for Lnight). Conversely, mean noise levels were slightly lower in more deprived quintiles of fuel poverty. Differences have been further quantified using a random-effects model, accounting for year. Heathrow airport is situated close to highly populated areas, some of which are very wealthy, so may not be representative of other airports. Results will be discussed with community groups near Heathrow prior to Internoise 2022. As air transport increases post-pandemic, information on noise exposures as well as views from community groups can inform future airport policies

Large area hybrid detectors based on Medipix3RX: commissioning and characterization at Sirius beamlines

Hybrid pixel detectors are inherently modular and typically operate at relative low noise levels, while providing high spatial resolution, easing the assembling of large area detectors. This report summarizes the procedures and results regarding the characterization studies performed on the Medipix3RX ASIC based detector, known as PIMEGA, which was developed and is deployed at Sirius synchrotron facility at the Brazilian Synchrotron Light Laboratory. They consist of modular constructions that reach up to 170 × 170 mm2 of active area, with 300 µm thick silicon sensors, and pixel size of 55 × 55 µm2. We investigated the spatial and energy resolutions, structural noise properties, and count-rate linearity up to large fluxes from the synchrotron beamlines. The Modulation Transfer Function presents a considerably slow decay rate as a function of the spatial frequency, reaching 30% at 10.5 lp/mm, for an operation with a threshold setting equivalent to half of the incident energy. We have also found that the energy resolutions of the two highest gain modes are around 15% for 17.5 keV incident photons. In particular, we have found that, for a noise-suppressing set of configurations, the detector presents a dead time of 460 ± 10 ns. Additionally, the pile-up effect leads to a 10% loss of linearity at input fluxes of 2.3 × 105 photons/pixel.s. These results open the possibility for operating these devices with in-depth knowledge of their performance, allowing various scientific experiments, such as X-ray diffraction, computed controlled tomography, and ptychography.

Measuring low noise level in dwellings

DS 490:2018 "Akustik - Lydklassifikation af boliger", translated to "Acoustics - Acoustic Classification of Dwellings" relates to many acoustical parameters, and two of them "Noise from ventilation" and "Reverberation times" are now being tested in real life in the Danish "Voluntary sustainability classification".(Called FBK) Danish building regulation state a sound level limit in dwellings from all kinds of technical systems at 30 dB(A), equal to the Acoustical class C, comparable with the limit in several countries. Research shows that 30 dB(A) does not ensure acoustic satisfying comfort for most people. The FBK have chosen to lower the noise limit to 25 dB(A), equal to acoustical class B. We need to improve the ventilation systems in order to reach such lower level of sound emission, and to be able to measure it implemented in dwellings. The paper describes some methods and in situ experience fighting background noise problems caused by activities around and in the dwelling, as well as in the measurement systems. Special analyses and editing of recordings based on listening and pragmatic judgements are relevant. The goal is: Measure and document low noise without a cost in time and money being a barrier for lower noise levels in dwellings.

Enhancing cosmic shear with the multiscale lensing probability density function

ABSTRACT We quantify the cosmological constraining power of the ‘lensing probability density function (PDF)’ – the one-point probability density of weak lensing convergence maps – by modelling this statistic numerically with an emulator trained on w cold dark matter cosmic shear simulations. After validating our methods on Gaussian and lognormal fields, we show that ‘multiscale’ PDFs – measured from maps with multiple levels of smoothing – offer considerable gains over two-point statistics, owing to their ability to extract non-Gaussian information: For a mock Stage-III survey, lensing PDFs yield 33 per cent tighter constraints on the clustering parameter $S_8=\sigma _8\sqrt{\Omega _{\rm m}/0.3}$ than the two-point shear correlation functions. For Stage-IV surveys, we achieve >90 per cent tighter constraints on S8, but also on the Hubble and dark energy equation-of-state parameters. Interestingly, we find improvements when combining these two probes only in our Stage-III set-up; in the Stage-IV scenario the lensing PDFs contain all information from the standard two-point statistics and more. This suggests that while these two probes are currently complementary, the lower noise levels of upcoming surveys will unleash the constraining power of the PDF.

An HIV stochastic model with cell-to-cell infection, B-cell immune response and distributed delay

In this study, a delayed HIV stochastic model with virus-to-cell infection, cell-to-cell transmission and B-cell immune response is proposed. We first transform the stochastic differential equation with distributed delay into a high-dimensional degenerate stochastic differential equation, and then theoretically analyze the dynamic behaviour of the degenerate model. The unique global solution of the model is given by rigorous analysis. By formulating suitable Lyapunov functions, the existence of the stationary Markov process is obtained if the stochastic B-cell-activated reproduction number is greater than one. We also use the law of large numbers theorem and the spectral radius analysis method to deduce that the virus can be cleared if the stochastic B-cell-inactivated reproduction number is less than one. Through uncertainty and sensitivity analysis, we obtain key parameters that determine the value of the stochastic B-cell-activated reproduction number. Numerically, we examine that low level noise can maintain the number of the virus and B-cell populations at a certain range, while high level noise is helpful for the elimination of the virus. Furthermore, the effect of the cell-to-cell infection on model behaviour, and the influence of the key parameters on the size of the stochastic B-cell-activated reproduction number are also investigated.

Counting statistics of single electron transport in bilayer graphene quantum dots

We measure telegraph noise of current fluctuations in an electrostatically defined quantum dot in bilayer graphene by real-time detection of single electron tunneling with a capacitively coupled neighboring quantum dot. Suppression of the second and third cumulant (related to shot noise) in a tunable graphene quantum dot is demonstrated experimentally. With this method we demonstrate the ability to measure very low current and noise levels. Furthermore, we use this method to investigate the first spin excited state, an essential prerequisite to measure spin relaxation.

Frequency comb-referenced cavity ring-down spectroscopy of natural water between 8041 and 8633 cm−1

The 1.25 µm atmospheric transparency window is of importance for a number of atmospheric applications. As a continuation of our previous works on the improvement of water vapor line parameters in the near infrared, the room temperature absorption spectrum of water vapor in natural isotopic abundance is recorded with unprecedented sensitivity between 8041 and 8633 cm−1, using comb-referenced cavity ring-down spectroscopy. The line positions and intensities of more than 5400 lines were retrieved. Their intensities range between 3.6 × 10−30 and 1.5 × 10−22 cm/molecule. The high sensitivity and low noise level of the recordings (αmin≈ 10−11 cm−1) allow for measuring more than 1600 new lines and determine their positions with an accuracy of about 10−4 cm−1 in the case of isolated features. The rovibrational assignments were performed using known experimental energy levels and calculated spectra based on variational calculations by Schwenke and Partridge. The final line list is assigned to more than 5400 transitions of the first six water isotopologues (H216O, H218O, H217O, HD16O, HD18O and HD17O). The measured line positions allow to determine the energy of 79 new levels of H216O, H218O, H217O, and HD16O, and to correct 139 previously reported term values. Although a good agreement is generally observed, the comparison to the HITRAN2020 spectroscopic database and to the W2020 transition frequencies reveals a number of discrepancies both for line positions and line intensities. The lack of traceability of some HITRAN line parameters and some biases in the derivation procedure of the W2020 energy levels are confirmed in the studied range. Validation tests of the theoretical values of the line intensities against measured values show both band-by-band variations of the deviations on the order of a few % and line-by-line fluctuations within a given band.

Aircraft Noise Reduction Strategies and Analysis of the Effects.

In this study, six aircraft noise reduction strategies including the optimization of aircraft type, regulation of night flight number, optimization of flight procedure, modification of operating runway, land use planning and installation of sound insulation windows were proposed to alleviate the harmful impact of aircraft noise on the local area and population near Guangzhou Baiyun International Airport (BIA) in China. The effects of all proposed strategies except for land use planning and sound insulation windows were simulated and analyzed using CadnaA software. The results indicate that these noise reduction strategies have their own advantages and each of them can serve as an effective noise reduction measure for different applications. For instance, the replacement of noisy aircraft with low-noise aircraft can simultaneously reduce the area and population exposed to a high noise level, while the optimization of flight procedure can only reduce the population exposed under relatively low noise levels (70 ≤LWECPN ≤ 75 dB). Nevertheless, the modification of operating runway is more effective in reducing the population suffering under high noise levels (LWECPN > 85 dB). Among these strategies, reducing the number of night flights is found to be most effective in reducing the overall noise-exposed area and population. Additionally, with the assistance of noise mapping, proper land use planning was suggested according to national standards, and the installation of sound insulation windows with different sound reduction grades can be determined for different areas impacted by the aircraft noise of BIA. It is believed that the results of this study can be applied as a reference in selecting suitable noise reduction strategies to improve the acoustic environment of a specific airport.

Rapid Determination of Selected PFAS in Textiles Entering the Waste Stream.

Due to new European legislation, products entering the waste stream containing some perfluoro alkyl substances (PFAS) are subject to "low persistent organic pollutant concentration limits". Concentrations of restricted PFAS must be below this limit for them to be legally recycled or disposed of. A rapid extraction and clean-up method was developed for the determination of 21 PFAS in various polymers used in soft furnishings and upholstery. The optimised method used vortexing and ultrasonication in methanol (0.1% NH4OH), followed by a dilution and syringe filter clean-up step. PFAS were subsequently determined via UPLC-TripleTOF/MS. Good recoveries (80-120%) of target analytes were obtained with tall and narrow chromatogram peaks. The method was validated using control matrix samples spiked with target analytes. Repeated measurements of concentrations of target compounds showed good agreement with the spiked concentrations demonstrating good accuracy and precision. The resultant extracts provided low noise levels resulting in low limits of quantification ranging from 0.1 to 0.4 mg/kg. The developed method was applied successfully to real consumer products and it provided various advantages over traditional methods, including a substantially reduced analysis time, consumables and solvent consumption, and a high sample throughput which is critical to comply with implemented and proposed legislation.

High-Density Near-Ultraviolet Silicon Photomultipliers: Characterization of photosensors for Cherenkov light detection

In recent years, Silicon Photomultipliers (SiPMs) have proven to be highly suitable devices for applications where high sensitivity to low-intensity light and fast responses are required. Among their many advantages are their low operational voltage when compared with classical photomultiplier tubes, mechanical robustness, and increased photon detection efficiency (PDE).Here we present a full characterization of a SiPM device technology developed in Italy by Fondazione Bruno Kessler, which is suitable for Cherenkov light detection in the Near-Ultraviolet (NUV) band. This device is a High-Density (HD) NUV SiPM, based on a microcell of 40μm×40μm and with an area of 6 × 6mm2, providing low levels of dark noise and high PDE peaking in the NUV band. This particular device has been selected to equip a part of the focal plane of the Schwarzschild–Couder Telescope (SCT) prototype proposed for the Cherenkov Telescope Array (CTA) Observatory.

Hydrodynamic and radial force characteristics with design of a single-channel pump for wastewater treatment based on the similarity law

Single-channel pumps are widely employed for wastewater treatment in Korea. To render possible the application of this type of pump in other countries using different power frequencies, the present study applies the similarity law to redesign the impeller geometry of the single-channel pump optimized in a previous study. To this end, the numerical simulations are conducted using steady and unsteady Reynolds-averaged Navier–Stokes equations and a shear stress transport reattachment modification turbulence model. Numerical simulations are confirmed by an experiment based on the 11th API610 standard. At design flow rate condition, the novel pump model meets the design requirement with the minimum total head coefficient of 3.078. The distance between the center of the radial force and the origin of the new model is notably decreased in comparison with the reference model, especially at the design point with a reduction of 3.418 times. Moreover, the results of the unsteady radial force are consistent with those of the steady radial force. The pressure oscillations in the new model are likewise reduced compared to the reference model, particularly at several measuring points under design flow rate with pressure oscillations dropping from 8.5% and 10.1–5.3% and 1.2%, respectively. The non-uniform high-pressure region caused by the rotor-stator interaction at the outlet of the impeller is evidently suppressed in the new model. The new single-channel pump is feasible for new power frequency and performs better with low vibration and noise levels during the operation.

Impact of hydrogen cells on economic efficiency and the environment according to renewable energy standards

Purpose: The research goal of the work is to determine the levels of profitability of hydrogen cells and their impact on the environment. The authors of the article answer the question under what boundary conditions there is an economic justification for the use of hydrogen cells in various branches of the economy while respecting environmental regulations and standardizing energy production by the principles of a sustainable economy. Design/methodology/approach: The research was carried out using economic measurement model methods. These methods allowed the authors to calculate the market value of the investment with the assumed boundary criteria and to determine the economic effectiveness of the analyzed research problem. Additionally, the authors analyzed the problems of the widespread use of hydrogen in terms of its storage in technical, technological and economic terms. Findings: Research has shown that it is possible to obtain up to 7% by weight. hydrogen relative to the mass of the metal. Carbon nanofibers may become the material of the future for making hydrogen tanks. The use of fuel cells brings many benefits: simple structure and operation, neutral impact on the environment, and low noise level. Moreover, hydrogen fuel cell technology allows for efficient operation for a long time and the possibility of high momentary overloads, which allows for considerable scalability and wide application with economic justification. Practical implications: The presented models have shown that the project of their implementation is fully economically justified and will allow investors to make a rational investment decision. Originality/value: The original contribution of this work is the implication of the data on real research models. This data allowed the authors to make calculations and indicate directions of improvement for the construction of innovative hydrogen cell solutions as part of the standardization of regulations on renewable energy sources in various sectors of the economy. Keywords: Hydrogen, hydrogen storage, fuel cells. Category of the paper: Research paper.

Residual learning with feedback for strong random noise attenuation in seismic data

In random seismic noise attenuation, when the noise energy is higher than or close to a signal, it is difficult to distinguish the signal from the noise. This random noise is relatively strong compared to the signal and is called strong random noise. We have developed a deep learning framework to recover the signal from the strong random noise. The framework is based on a residual learning network and feedback connection and is called the feedback residual network. The residual network (ResNet) suppresses random noise through residual fitting and improves the network’s training efficiency. The feedback connection allows the framework to process data in iterations. In each iteration, the feedback connection proportionally combines the input and output of the ResNet to reconstruct a new input with a lower noise level. This enhances denoising performance by asymptotically decreasing the input noise level and retrieving the remaining signals from the estimated noise, thereby reducing the difficulty of strong random noise attenuation. We terminate the feedback iterations according to the energy change of the estimated noise in each iteration. Synthetic and field examples demonstrate that our network can effectively attenuate the strong random noise.

Cross-Modal Variational Auto-Encoder for Content-Based Micro-Video Background Music Recommendation

In this paper, we propose a cross-modal variational auto-encoder (CMVAE) for content-based micro-video background music recommendation. CMVAE is a hierarchical Bayesian generative model that matches relevant background music to a micro-video by projecting these two multimodal inputs into a shared low-dimensional latent space, where the alignment of two corresponding embeddings of a matched video-music pair is achieved by cross-generation. Moreover, the multimodal information is fused by the product-of-experts (PoE) principle, where the semantic information in visual and textual modalities of the micro-video are weighted according to their variance estimations such that the modality with a lower noise level is given more weights. Therefore, the micro-video latent variables contain less irrelevant information that results in a more robust model generalization. Furthermore, we establish a large-scale content-based micro-video background music recommendation dataset, TT-150k, composed of approximately 3,000 different background music clips associated to 150,000 micro-videos from different users. Extensive experiments on the established TT-150k dataset demonstrate the effectiveness of the proposed method. A qualitative assessment of CMVAE by visualizing some recommendation results is also included.