Validation of CT dose‐reduction simulation

Purpose: Cone-beam computed tomography (CBCT) is commonly used in the operating room to evaluate the placement of surgical implants in relation to critical anatomical structures. A particularly problematic setting, however, is the imaging of metallic implants, where strong artifacts can obscure visualization of both the implant and surrounding anatomy. Such artifacts are compounded when combined with low-dose imaging techniques such as sparse-view acquisition. Approach: This work presents a dual convolutional neural network approach, one operating in the sinogram domain and one in the reconstructed image domain, that is specifically designed for the physics and setting of intraoperative CBCT to address the sources of beam hardening and sparse view sampling that contribute to metal artifacts. The networks were trained with images from cadaver scans with simulated metal hardware. Results: The trained networks were tested on images of cadavers with surgically implanted metal hardware, and performance was compared with a method operating in the image domain alone. While both methods removed most image artifacts, superior performance was observed for the dual-convolutional neural network (CNN) approach in which beam-hardening and view sampling effects were addressed in both the sinogram and image domain. Conclusion: The work demonstrates an innovative approach for eliminating metal and sparsity artifacts in CBCT using a dual-CNN framework which does not require a metal segmentation.

Cross-modal effects occur when subjective opinions of stimuli from one sense (e.g., tactile at steering wheel) are influenced by simultaneous stimuli in another sense (auditory). A steering wheel rig was used to provide specified vibration stimuli to participants' hands, and a recording of vehicle sound was played in accordance with the vibrations. Participants were subjected to test conditions with vibration values between 10 m s−2 and 20 m s−2, and auditory stimuli between 88 dB(A) and 98 dB(A) peak. Participants were neither informed of nor asked about any changes in the noise level: thus the purpose of the experiment was withheld from them. The results show that as noise level increased, subjective ratings of steering wheel vibration increased. Therefore, car cabin noise could be used to enhance the feel of the vibration at the steering wheel.

The aerodynamic performance results indicate that the fan adiabatic efficiency was highest with nominal tip clearance (about 91.8% at design speed) and decreased as fan tip clearance increased. Differences in fan efficiency between tip clearance configurations was small below 77.5% fan design speed, about 0.5% maximum, and got larger as fan speed increased, to around 1% maximum at 100% fan design speed. The decreases in efficiency are due to a lower blade loading and higher temperature rise over the outer 20% of the blade span; the mechanism is the tip leakage flow from pressure to suction surface of the fan blade, which increases as the fan tip gap increases. Farfield acoustic results show that changes in the noise level are primarily aft-radiating and are on the order of 1 to 5 dB in fan broadband noise level as measured for the rotor-alone configuration (which had no stators). There were very small changes in noise level with tip clearance for the 54-vane Baseline Outlet Guide Vane configuration. Small changes in noise were also seen for the other two Outlet Guide Vane configurations tested - the 26-vane radial Low Count Outlet Guide Vanes and the 26-vane aft swept Low Noise Outlet Guide Vanes. The rotor alone acoustic results suggest that tip clearance changes induce broadband noise changes at the fan tip, but that the noise differences are masked by the rotor wake/stator interaction noise generated with the Outlet Guide Vanes installed. The magnitude of the rotor/stator interaction tones showed small increases with increasing fan tip gap at fan speeds below transonic, indicating slightly stronger interactions with the larger tip vortex wakes that formed with larger tip gaps. The broadband levels, for the most part, only showed the effect of Outlet Guide Vane geometry on the magnitude of the noise as the tip gap changed. The Laser Doppler Velocimetry flow field results show that the flow downstream of the tip of the blades changes very little with changes in the tip clearance when operating at the approach condition. At both the cut-back and take-off speeds, significant changes in the tip flow occur with changes in the tip clearance. Since these changes in the tip flow are not *

An idea to filter positron emission tomography (PET) data in an intermediate state between sinogram and image domains, called stackgram domain, has been shown to have advantages over common radial sinogram domain filtering. This method has been approximated with sinogram filter design that utilizes sinusoidal trajectories. However, the performance of the approximation tested with different filter types is not as stable as the performance of these same filter types in the stackgram domain. The reason for this is that the samples used in the stackgram domain are interpolated samples between the radial bins of the sinogram and not exact radial sinogram bins as in the approximation. Additionally, the output samples of the stackgram filter are usually also between two neighboring radial sinogram samples. The main motivation is to understand better how the stackgram filters work by showing what they exactly look like in the sinogram domain. This is accomplished by introducing extended sinogram, i.e. interpolating the radial direction enough for it to contain all the values equivalent to the stackgram domain values. This way the equivalent filters can be shown for different radial locations of the sinogram. Finally, the decimation is performed in order to obtain the same output as the one obtained when the filtered stackgram is transformed back into the sinogram domain.

The multi-source stationary CT, where both the detector and X-ray source are fixed, represents a novel imaging system with high temporal resolution that has garnered significant interest. Limited space within the system restricts the number of X-ray sources, leading to sparse-view CT imaging challenges. Recent diffusion models for reconstructing sparse-view CT have generally focused separately on sinogram or image domains. Sinogram-centric models effectively estimate missing projections but may introduce artifacts, lacking mechanisms to ensure image correctness. Conversely, image-domain models, while capturing detailed image features, often struggle with complex data distribution, leading to inaccuracies in projections. Addressing these issues, the Dual-domain Collaborative Diffusion Sampling (DCDS) model integrates sinogram and image domain diffusion processes for enhanced sparse-view reconstruction. This model combines the strengths of both domains in an optimized mathematical framework. A collaborative diffusion mechanism underpins this model, improving sinogram recovery and image generative capabilities. This mechanism facilitates feedback-driven image generation from the sinogram domain and uses image domain results to complete missing projections. Optimization of the DCDS model is further achieved through the alternative direction iteration method, focusing on data consistency updates. Extensive testing, including numerical simulations, real phantoms, and clinical cardiac datasets, demonstrates the DCDS model's effectiveness. It consistently outperforms various state-of-the-art benchmarks, delivering exceptional reconstruction quality and precise sinogram.

Low-dose computed tomography (LDCT) helps to reduce radiation risks in CT scanning while maintaining image quality, which involves a consistent pursuit of lower incident rays and higher reconstruction performance. Although deep learning approaches have achieved encouraging success in LDCT reconstruction, most of them treat the task as a general inverse problem in either the image domain or the dual (sinogram and image) domains. Such frameworks have not considered the original noise generation of the projection data and suffer from limited performance improvement for the LDCT task. In this paper, we propose a novel reconstruction model based on noise-generating and imaging mechanism in full-domain, which fully considers the statistical properties of intrinsic noises in LDCT and prior information in sinogram and image domains. To solve the model, we propose an optimization algorithm based on the proximal gradient technique. Specifically, we derive the approximate solutions of the integer programming problem on the projection data theoretically. Instead of hand-crafting the sinogram and image regularizers, we propose to unroll the optimization algorithm to be a deep network. The network implicitly learns the proximal operators of sinogram and image regularizers with two deep neural networks, providing a more interpretable and effective reconstruction procedure. Numerical results demonstrate our proposed method improvements of > 2.9 dB in peak signal to noise ratio, > 1.4% promotion in structural similarity metric, and > 9 HU decrements in root mean square error over current state-of-the-art LDCT methods.

Geometric design of forest roads and design of their landscapes can reduce noise pollution and its harmful effects on human health. We investigated the effects of technical and biological parameters such as geometric road design and various roadside tree stands on reducing noise pollution according to the tree density and distance from roadway in Darabkola Forests, Sari, Iran. We recorded the noise generated by a car (Land Rover) relative to changes in longitudinal slope, horizontal curve radius and type of road pavement. We also measured noise levels according to roadside tree density and stand type (coniferous and hardwood) in 40 rectangular plots of three widths (25, 100 and 300 m) and 50 m length that were randomly demarcated along forest roads. The changes in noise level were recorded using a decibel meter with an accuracy of ±1.5 dB and resolution of 0.1 dB. Noise levels were higher alongside unpaved roads than alongside paved roads. There was an inverse relationship between the measured noise level and horizontal curve radius. The rate of noise level on horizontal curve with a radius less than 30, 30–45 m and more than 45 m were 64.8, 70.8 and 75.9 dB, respectively. The noise level increased with the increasing longitudinal slope of the road. There was a significant difference between the noise level on slopes less than 3 % (67 dB) and 3–8 % (71.2 dB) in comparison with slopes greater than 8 % (77.8 dB), (p < 0.05). Pinus brutia L. reduced the noise level more (about 6 dB) in stands of 1/3 density of mixed hardwoods within 25 m from middle of the road. Careful design of geometric properties of forest roads as well as planting coniferous trees with hardwoods is a suitable solution for reducing noise pollution.

This communication compares the previously reported results of the acoustic environment, mainly noise levels at a fixed point, in a quiet residential area in Kobe, Japan, under the declaration of the COVID-19 state of emergency in May 2020 with the results of two follow-up studies in the same area: subsequent follow-up noise measurements in June and July–August 2020, and the present results of measurements in September–October 2020. The results of the comparison among the above three measurements suggest that noise levels were lower during September-October 2020 than during the declaration of the state of emergency in May 2020. In the period from May to October 2020, the noise level was significantly higher in July and August of the same year due to the sound of cicadas, which are common in this area. This suggests that it is difficult to set the target values of the acoustic environment planning by referring to the low noise level at lockdown or similar measures in areas with large seasonal variations in acoustic environment. Although many case studies are necessary to obtain appropriate target values, one case study is presented in this communication to illustrate an example and discuss its difficulty.

Acoustic contamination assessment during the transition between the COVID-19 restrictions and reactivation: A exploratory analysis in Guayaquil

Inspired by great performance of convolutional neural network (CNN), we propose an iterative positron emission tomography (PET) reconstruction using a CNN prior. We used the denoising CNN (DnCNN) method and trained the network using regular dose images as groundtruth and low dose images as input. Poisson thinning method is used for generating the low dose data by downsampling counts. Due to the DnCNN is trained at a certain noise level, the noise level change in each iteration is one of major problems. To address this issue, we propose a local linear fitting (LLF) function incorporated with DnCNN to improve the image quality by preventing unwanted bias. By using LLF function, we demonstrate that the proposed method is robust to noise level changes in iterations. In bias and variance studies in simulations, the proposed method outperforms the conventional iterative methods. We confirm that the proposed method improves the reconstructed image both quantitatively and qualitatively.

Three groups of 12 subjects each performed a mental arithmetic task while exposed to continuous white noise in two experimental sessions. In Session I, each group was exposed to a different noise level [56, 72.5, or 85dB(A)], whereas in Session II all had the medium [72.5 dB(A)] intensity. Arithmetic performance deteriorated with increasing noise intensity in Session I, and in Session II each group retained its position relative to the other two groups. Thus, the higher the noise intensity in the first session, the poorer the performance in the second session. These results suggest that the motivation and energy mobilized in Session II were determined by the cognitive set adopted to meet the demands of Session I. Arousal level, as measured by adrenaline excretion and heart rate, was higher during both experimental sessions than in a control session. The rise in adrenaline output was of the same magnitude regardless of the intensity of the noise, which indicates that the subjects adjusted to the increase in total load by slowing down their performance rather than by raising their energy mobilization. Heart rate rose in response to changes in noise level in Session II relative to Session I, regardless of the direction of the change.

To evaluate die effect of changes in die NICU environment on sound levels. A prospective quasi-experimental design evaluated sound levels in a 43-bed NICU. Decibel levels were monitored utilizing a data-logging dosimeter for 24 hours weekly over 12 months. Sound levels were also measured inside four different incubator models. Forty-four 24-hour decibel recordings were obtained in one of eight randomly selected four-bed pods. In addition, a single 1-hour recording was obtained in four different models of vacant incubators. Ambient sound levels. Decibel levels were analyzed to identify changes in noise levels following alterations in the NICU environment. Installation of motion-sensing motorized paper towel holders significantly increased levels at beds closest to the towel dispensers, as did the trial of a new communication system. Decibel levels in four different incubators revealed varying noise levels. This study suggests that all environmental changes must be monitored to ensure that they reduce rather than increase noise levels.

The aerodynamic noise of a NACA 0012 and NACA 0021 aerofoil is measured and compared in order to determine whether there are differences in their noise signatures with a focus on the onset of stall. Measurements of the self-noise of each aerofoil are measured in an open-jet Anechoic Wind Tunnel at Reynolds numbers of 64,000 and 96,000, at geometric angles of attack from −5° through 40° at a resolution of 1°. Further measurements are taken at Re = 96,000 at geometric angles of attack from −5 through 16° at a resolution of 0.5°. Results show that while the noise generated far into the stall regime is quite similar for both aerofoils the change in noise level at the onset of stall is significantly different between the two aerofoils with the NACA 0021 exhibiting a much sharper increase in noise levels below a chord-based Strouhal number of Stc = 1.1. This behaviour is consistent with the changes in lift of these aerofoils as well as the rate of collapse of the suction peak of a NACA 0012 aerofoil under these flow conditions.

Using noise inconsistencies for blind image forensics

Airport operation is a major source of community concerns due to aircraft noise emissions particularly in locations close to airports and aircraft flight tracks. The effects of aircraft noise on humans can cause hearing damage for people working at airports and speech and sleep interference which can develop into stress for nearby communities. As such, it becomes essential when locating and designing to expand airports to optimize flight paths in a way to reduce noise exposure to nearby communities. For this purpose, it is required to assess changes in noise levels resulting from new runway configurations, evaluate increased aircraft operation demand, new fleet mix, revised routings and airspace structures, alternative flight profiles, and modifications to other operational procedures. This paper presents a noise impact assessment for the proposed expansion at the Beirut International Airport, Lebanon. Field noise measurements were collected at sensitive noise receptors along selected locations around the airport and the corresponding airport traffic flow was characterized. A mathematical model was then applied to simulate noise levels as a function of the different airplane classes, operations, flight path, and landing characteristics. The model was also used to predict future noise levels, optimize airplane flight path, and assess mitigation measures to minimize potential aircraft noise impacts.