Noise Correction in Pairwise Document Preferences for Learning to Rank

BackgroundInformation transmission and processing in the nervous system has stochastic nature. Multiple factors contribute to neuronal trial-to-trial variability. Noise and variations are introduced by the processes at the molecular and cellular level (thermal noise, channel current noise, membrane potential variations, biochemical and diffusion noise at synapses etc). The stochastic processes are affected by different physical (temperature, electromagnetic field) and chemical (drugs) factors. The aim of this study was experimental investigation of hypotheses that increase in the noise level in the brain affects processing of visual information. Change in the noise level was introduced by an external factor producing excess noise in the brain.MethodsAn exposure to 450 MHz low-frequency modulated microwave radiation was applied to generate excess noise. Such exposure has been shown to increase diffusion, alter membrane resting potential, gating variables and intracellular Calcium efflux. Nine healthy volunteers passed the experimental protocol at the lower (without microwave) and the higher (with microwave) noise level. Two photos (visual stimuli) of unfamiliar, young male faces were presented to the subjects, one picture after another. The task was to identify later the photos from a group of six photos and to decide in which order they were presented. Each subject had a total of eight sessions at the lower and eight at the higher noise level. Each session consisted of 50 trials; altogether a subject made 800 trials, 400 at the lower and 400 at the higher noise level. Student t-test was applied for statistical evaluation of the results.ResultsCorrect recognition of both stimuli in the right order was better at the lower noise level. All the subjects under investigation showed higher numbers of right answers in trials at the lower noise level. Average number of correct answers from n=400 trials with microwave exposure was 50.3, without exposure 54.4, difference 7.5%, p<0.002. No difference between results at the lower and the higher noise level was revealed in the case of only partly correct or incorrect answers.ConclusionsOur experimental results showed that introduced excess noise reduced significantly ability of the nervous system in correct processing of visual information.

CT perfusion imaging (CTP) plays an important role in decision making for the treatment of acute ischemic stroke with large vessel occlusion. Since the CT perfusion scan time is approximately one minute, the patient is exposed to a non-negligible dose of ionizing radiation. However, further dose reduction increases the level of noise in the data and the resulting perfusion maps. We present a method for reducing noise in perfusion data based on dimension reduction of time attenuation curves. For dimension reduction, we use either the fit of the first five terms of the trigonometric polynomial or the first five terms of the SVD decomposition of the time attenuation profiles. CTP data from four patients with large vessel occlusion and three control subjects were studied. To compare the noise level in the perfusion maps, we use the wavelet estimation of the noise standard deviation implemented in the scikit-image package. We show that both methods significantly reduce noise in the data while preserving important information about the perfusion deficits. These methods can be used to further reduce the dose in CT perfusion protocols or in perfusion studies using C-arm CT, which are burdened by high noise levels.

To assess acoustic noise levels for ambient noise and magnetic resonance (MR) imaging at 1.5 T during use of six "worst-case" pulse sequences. Acoustic noise measurements were obtained at the entrance, center, and exit of the magnet bore of the MR system by means of a specially modified device that is unaffected by electromagnetic radiation. The highest ambient noise level was 73 dB at both the entrance and exit of the magnet bore (A-weighted scale). The highest noise levels during MR imaging occurred during use of a gradient-echo (GRE) pulse sequence and was 102 dB at the entrance and exit of the magnet bore and 103 dB at the center (A-weighted scale). MR imaging performed with the worst-case pulse sequences did not produce noise levels that exceeded federal guidelines. Noise levels were, however, high enough to impair oral communication and annoy patients. Thus, techniques to attenuate acoustic noise and allow improved operator-patient communication should be used during MR imaging, especially during use of GRE pulse sequences.

The transport sector has been moving toward electrification due to the significant advancement in E-mobility technology. This prioritizes reliable and safe battery energy storage system (BESS) operation. Therefore, accurate battery state-of-charge (SoC) estimation is essential in effectively monitoring and controlling the BESS stability. Many studies have been conducted to estimate the BESS SoC and improve the estimation accuracy. Nevertheless, considering system complexity and computational efforts, the suggested SoC estimate techniques fall short of providing optimal filtering performance with high noise levels. In this regard, this article introduces SoC estimation using the triple forgetting factor adaptive extended Kalman filter (TFF-AEKF) to provide better SoC estimation accuracy and faster convergence considering the high measurement noise levels and environmental circumstances encountered by the operation of electric buses (EBs). The performance of the proposed TFF-AEKF is evaluated and compared to the conventional adaptive extended Kalman filter (AEKF) and the dual forgetting factor AEKF (DFF-AEKF), considering low and high measurement noise levels. It has been validated that the proposed algorithm can provide faster convergence and better accuracy when considering a high measurement noise level. In addition, the three filters are evaluated using four performance indicators, namely, maximum absolute error (MaxAE), mean absolute error (MAE), root mean square error (RMSE), and convergence time. It is concluded that the presented method offers faster convergence and lower error. Results have demonstrated that the proposed algorithm provides an RMSE of 0.3%, an MAE of 0.01%, and a MaxAE of 1.7% for SoC estimation.

Magnetic resonance imaging (MRI) coils are typically associated with high acoustic noise levels, which can pose problems for the patient and attending medical staff. Lorentz forces are generated by gradient coils, which are switched rapidly. The Lorentz forces displace the coil former and promote a pressure wave that is responsible for acoustic noise up to 130 dB. The high noise levels are an unavoidable consequence of requiring the gradient coils to generate highly homogeneous linear magnetic fields. However, in this article, we present a mathematical method for designing coils that achieve a drastic reduction in noise level by deliberately sacrificing field linearity. We show that useful images can nevertheless be obtained from the nonlinear fields associated with these quiet coils. Examples are given that demonstrate the robustness of the method. © 2011 Wiley Periodicals, Inc. Concepts Magn Reson Part B (Magn Reson Engineering) 39B: 191–205, 2011

Sleep fragmentation and deprivation in critically ill patients – Is noise a factor?

In Vilnius and in other bigger towns the noise generated by mobile sources accounts for up to 90 % of the total noise level in a town. A high noise level is detected at motor vehicle parking lots. To measure this noise level, testing was carried out at guarded parking lots which were divided into certain types. One lot of each type was chosen for testing. The number of measuring spots was conditioned by the size of a lot. The data obtained during testing reveal that the highest noise level is generated not at a parking lot but on a nearby street. Such conclusions are proved by the results of testing carried out at a parking lot on E. Sugihara street. Although this lot parks over 100 vehicles, the equivalent noise level is 47 dBA. Basing on the testing data, the numeral simulation of the noise dispersion in areas in the vicinity of parking lots has been run with the help of the numeral simulation software IMMI.

This study describes the effects of broadband background noise on the average discharge rate to best-frequency (BF) tones of auditory nerve fibers in the cat. The effects of exposure to long-term continuous noise are compared to the effects of noise gated on and off simultaneously with test tones. Addition of background noise causes a shift of the dynamic portion of tone-evoked rate versus level functions to higher tone intensities. The shift occurs at a mean rate of 0.61 dB of shift for each 1-dB increment in noise level. The rate of shift is independent of best frequency and spontaneous discharge rate. The noise level at which the shift begins is frequency dependent and is consistent with the frequency-dependent bandwidths of auditory nerve fiber tuning curves. The adjustment of the dynamic range shows many similarities to two-tone suppression. Therefore, it is most likely that it is caused by suppression of the response to the BF test tone by energy present in the noise at surrounding frequencies. At high noise levels, the ability of auditory nerve fibers to respond to test tones is limited by the rate response to the noise. As noise level increases, the discharge rate it evokes approaches a fiber's saturation rate and ultimately eliminates the fiber's ability to respond to test tones. Low spontaneous rate fibers, which have been shown to have higher thresholds and wider dynamic range (17,29), are significantly more resistant to saturation by high noise levels. Exposure to broadband noise prior to onset of test tones produces an overall decrement in response rate. This phenomenon is similar to the effects of short-term adaptation (32) and seems to develop independently of the shift of dynamic range. At high noise levels, previous exposure to the noise produces a small dynamic range shift. This effect is similar to that produced by suppression but is smaller. The effect is occluded in continuous noise backgrounds by the adjustment of sensitivity produced by suppression.

Previous research in cochlear implant (CI) recipients with bilateral severe-to-profound sensorineural hearing loss showed improvements in speech recognition in noise using remote wireless microphone systems. However, to our knowledge, no previous studies have addressed the benefit of these systems in CI recipients with single-sided deafness. The objective of this study was to evaluate the potential improvement in speech recognition in noise for distant speakers in single-sided deaf (SSD) CI recipients obtained using the digital remote wireless microphone system, Roger. In addition, we evaluated the potential benefit in normal hearing (NH) participants gained by applying this system. Speech recognition in noise for a distant speaker in different conditions with and without Roger was evaluated with a two-way repeated-measures design in each group, SSD CI recipients, and NH participants. Post hoc analyses were conducted using pairwise comparison t-tests with Bonferroni correction. Eleven adult SSD participants aided with CIs and eleven adult NH participants were included in this study. All participants were assessed in 15 test conditions (5 listening conditions × 3 noise levels) each. The listening conditions for SSD CI recipients included the following: (I) only NH ear and CI turned off, (II) NH ear and CI (turned on), (III) NH ear and CI with Roger 14, (IV) NH ear with Roger Focus and CI, and (V) NH ear with Roger Focus and CI with Roger 14. For the NH participants, five corresponding listening conditions were chosen: (I) only better ear and weaker ear masked, (II) both ears, (III) better ear and weaker ear with Roger Focus, (IV) better ear with Roger Focus and weaker ear, and (V) both ears with Roger Focus. The speech level was fixed at 65 dB(A) at 1 meter from the speech-presenting loudspeaker, yielding a speech level of 56.5 dB(A) at the recipient's head. Noise levels were 55, 65, and 75 dB(A). Digitally altered noise recorded in school classrooms was used as competing noise. Speech recognition was measured in percent correct using the Oldenburg sentence test. In SSD CI recipients, a significant improvement in speech recognition was found for all listening conditions with Roger (III, IV, and V) versus all no-Roger conditions (I and II) at the higher noise levels (65 and 75 dB[A]). NH participants significantly benefited from the application of Roger in noise for higher levels, too. In both groups, no significant difference was detected between any of the different listening conditions at 55 dB(A) competing noise. There was also no significant difference between any of the Roger conditions III, IV, and V across all noise levels. The application of the advanced remote wireless microphone system, Roger, in SSD CI recipients provided significant benefits in speech recognition for distant speakers at higher noise levels. In NH participants, the application of Roger also produced a significant benefit in speech recognition in noise.

The deleterious effects of noise pollution on public health have been well documented, with traffic noise being identified as a significant contributor to stress and adverse impacts on the human body and mind. In this study, sound levels at 12 different points in Al-Oqailat Park in Buraydah, Saudi Arabia, were measured using a sound level meter (SLM), with the study’s primary objective being to conduct this measurement. The experimental results were then compared with perception measurements collected from users who frequently visited Al-Oqailat park. Sound measurements were taken in four different zones (A, B, C, and D) during rush hours between 1:30 p.m. and 5:20 p.m. It was found that noise levels at point A1 peaked at 79 dBA at 4:40 p.m., while the lowest level recorded was 41.1 dBA at point D2 at 2:35 p.m. The range of noise levels varied between 79 and 41 dBA, with a rate of decline of 48.10%. Zones A and B seemed to have the highest noise levels during rush hours, since they were located closest to King Fahd Road and Al-Adl Street, while zone D exhibited the lowest noise levels due to its location as a parking lot for Buraydah Court. An intermediate noise level was found in zone C, in the middle of Al-Oqailat park. The people perception results, completed by 84 park visitors, showed that zone A was identified as having exceptionally high noise levels compared to the other zones, with zone D having the lowest levels. These results were consistent with the experimental findings and reflected that the points along King Fahd Road and Al-Adl Street had the highest noise levels. Overall, the research highlighted the dominance of car traffic and horns as the primary sources of noise pollution in and around Al-Oqailat Park, emphasizing the significance of meticulous site selection for parks in urban areas.

Alternative approach procedures are currently under investigation to evaluate their ground noise reduction potential. One such procedure involves approaching the airport at a considerably higher altitude compared with standard landing trajectories, followed by a spiraling descent (helix flight path) shortly before the runway threshold. In this way, high ground noise levels by approaching aircraft are dislocated away from the common approach path and concentrated in the area near the helix path, that is, in direct vicinity of the airport. Ground noise levels along the entire flight path before the helix are significantly reduced. The effectiveness of this procedure, referred to as A¢Â€Âœhelical noise abatement procedure,A¢Â€Â has been quantified by means of computational simulation analyses. These analyses also focused on aspects such as increased fuel burn and the occurrence of multiple noise events below the helix. In June 2009, a new autopilot by the DLR, German Aerospace Center, especially capable of tracking curved flight-path trajectories, was flight tested. Three helical noise abatement procedures were included in the flight plan, as well as standard and steep landing approaches. In addition, dedicated flyover noise measurements were organized. Twelve ground microphones were placed along the common approach path and the helical flight segment. The measured data confirm the predicted noise dislocation effects. High noise levels were found to be limited to observer locations around the helix. Results from computational noise prediction have been compared with the experimental data. Predicted trends and noise dislocation effects are in good agreement with the measurements, whereas the absolute numerical values show discrepancies. The flight test was closely accompanied by a research and development member of the German air navigation service provider DFS Deutsche Flugsicherung GmbH to study the impact of spiraling procedures on air traffic management integration aspects and air traffic controller workload, for example, increased interaction with the pilots. Obviously, a spiraling approach procedure would not be implemented into the existing air traffic scenario with its common approach paths and more frequented airports. The operational and economic environment still need more detailed investigation. Helical approaches become more feasible for implementation at small, less-frequented regional airports or during night hours to avoid possible noise related curfews.

ABSTRACTThe objective of this work was to characterise the bases of columns belonging to an industrial plant. We had to design a proper methodology for characterising these structures, in a place with limited available space and time for the deployment of sensors, and with other environmental constraints, such as high electromagnetic noise level, soil vibrations due to operating machines, and complex soil composition. In particular, the floor had a stainless steel cover about 1‐cm thick, below which there was first a layer of concrete about 30‐cm thick and then clay material used to absorb the oils of the machines. Finally, there was the natural soil on which the column bases stood.Given these characteristics, we performed dipole–dipole geoelectric profiles. Holes were drilled through the metallic cover and the concrete layer, with fixed spacings, to insert the electrodes. Because of this and the space and time restrictions, only one profile could be performed for each column. Since we knew that the data could have high noise levels and other errors negatively impacting the data quality, and also because there was a limitation to the length of the profiles, we optimised the measurement configuration by jointly using several electrode apertures, to ensure the best possible resolution and sufficient penetration depth. To obtain adequate error estimations, we performed direct and reciprocal measurements. As an alternative technique, we also carried out a second set of direct measurements, after removing and reinstalling the electrodes.Then, for each profile, we performed 2D inversions of the three datasets separately, without considering data errors. Besides, we inverted the datasets obtained by combining the first direct and the reciprocal measurements, and the first and second direct measurements, considering the two obtained error estimates. In addition, numerical simulations were made, combining direct 3D modelling with 2D inversions. To analyse the results, we took into account the knowledge we had in advance about the main characteristics of the structures.The first conclusion obtained is that, at least for this type of structures, inverting each dataset without considering data errors provided more realistic images. However, although the best results were obtained by inverting without considering errors, it was still fundamental to have different sets of data, for evaluating the reliability of the images and for discarding possible spurious artefacts. This is especially important in environmentally complex urban sites, where errors could be particularly high. Finally, through the numerical simulations, a more rigorous interpretation could be made, and also, the effect of data errors in the quality of the images could be assessed.

A characteristic of neurological signal processing is high levels of noise from subcellular ion channels up to whole-brain processes. In this paper, we propose a new model of electroencephalogram (EEG) background periodograms, based on a family of functions which we call generalized van der Ziel-McWhorter (GVZM) power spectral densities (PSDs). To the best of our knowledge, the GVZM PSD function is the only EEG noise model that has relatively few parameters, matches recorded EEG PSD's with high accuracy from 0 to over 30 Hz, and has approximately 1/fθ behavior in the midfrequencies without infinities. We validate this model using three approaches. First, we show how GVZM PSDs can arise in a population of ion channels at maximum entropy equilibrium. Second, we present a class of mixed autoregressive models, which simulate brain background noise and whose periodograms are asymptotic to the GVZM PSD. Third, we present two real-time estimation algorithms for steady-state visual evoked potential (SSVEP) frequencies, and analyze their performance statistically. In pairwise comparisons, the GVZM-based algorithms showed statistically significant accuracy improvement over two well-known and widely used SSVEP estimators. The GVZM noise model can be a useful and reliable technique for EEG signal processing. Understanding EEG noise is essential for EEG-based neurology and applications such as real-time brain-computer interfaces, which must make accurate control decisions from very short data epochs. The GVZM approach represents a successful new paradigm for understanding and managing this neurological noise.

The purpose of this study was to compare threshold-matched ears with and without suspected cochlear dead regions in terms of the speech perception benefit from high-frequency amplification. The Threshold Equalizing Noise Test (TEN) was used to assess the presence of dead regions. Speech perception was measured while participants were wearing a hearing aid fit to approximate DSL[i/o] targets. Consonant identification of nonsense vowel-consonant-vowel combinations was measured in quiet using a forced-choice procedure. Phoneme recognition was measured at signal-to-noise ratios ranging from 0 to +15 dB using the Computer-Assisted Speech Perception Assessment test (CASPA). Recognition scores were obtained for unfiltered stimuli and stimuli that were low-pass filtered at the estimated boundary of the suspected dead regions, 1/2 octave above and 1 octave above the boundary. Filter settings for the ears without suspected dead regions were the same as settings of the threshold-matched counterpart. In quiet and in low levels of noise, speech perception scores were significantly higher for the wide-band (unfiltered) condition than for the filtered conditions, and performance was similar for the ears with and without suspected dead regions. In high levels of noise, mean scores were highest in the wide-band condition for the ears without suspected dead regions, but performance reached an asymptote for the ears with suspected dead regions. These results suggest that patients with cochlear dead regions may experience speech perception benefit from wide-band high-frequency gain in quiet and low levels of noise, but not in high levels of noise.

Canalization is a form of network robustness found in genetic regulatory networks that results in a reduction of the variation of phenotypic expression relative to the complexity of the genome. Recently, it was discovered that canalization can evolve in a complex network through a self-organization of node (gene) behavior resulting from a competition of a network's nodes that selects for a diversity of behavior [K.E. Bassler, C. Lee, and Y. Lee, Phys. Rev. Lett. 93, 038101 (2004)]. Previously, this mechanism for the evolution of canalization has been studied only in deterministic model systems. This paper considers the effects of stochastic noise in the signals exchanged between nodes on the self-organized evolution of canalization. We find that small levels of stochastic noise increase the amount of canalization produced. At higher levels of noise, the amount of canalization produced levels off and reaches a maximum value, before it reduces at large levels of noise.