Simulated Images Research Articles

Interobserver variability of clinical target volume delineation in patients undergoing breast-conserving surgery without surgical clips: a pilot study on preoperative magnetic resonance simulation

BackgroundIn patients undergoing breast-conserving therapy without surgical clip implantation, the accuracy of tumor bed identification and the consistency of clinical target volume (CTV) delineation under computed tomography (CT) simulation remain suboptimal. This study aimed to investigate the feasibility of implementing preoperative magnetic resonance (MR) simulation on delineations by assessing interobserver variability (IOV).MethodsPreoperative MR and postoperative CT simulations were performed in patients who underwent breast-conserving surgery with no surgical clips implanted. Custom immobilization pads were used to ensure the same supine position. Three radiation oncologists independently delineated the CTV of tumor bed on the images acquired from MR and CT simulation registration and CT simulation alone. Cavity visualization score (CVS) was assigned to each patient based on the clarity of the tumor bed on CT simulation images. IOV was indicated by generalized conformity index (CIgen), denoted as CIgen−CT and CIgen−MR/CT, and the distance between the centroid of mass (dCOM), denoted as dCOMCT and dCOMMR/CT. The variation of IOV in different CVS subgroups was analyzed.ResultsA total of 10 patients were enrolled in this study. The median and interquartile range (IQR) of maximum pathological diameter of the tumors in all patients were 1.55 (0.80–1.92) cm. No statistical significance was found between the volumes of CTVs on CT simulation and on MR/CT simulation registration images (p = 0.387). CIgen−MR/CT was significantly larger than CIgen−CT (p = 0.005). dCOMMR/CT was significantly smaller than dCOMCT (p = 0.037). The median and IQR of CVS in all patients were 2.34 (2.00–3.08). The difference of CIgen between CIgen−MR/CT and CIgen−CT was larger in the low CVS group (p = 0.016). The difference of dCOM showed a decreasing trend when CVS was lower, although it did not reach statistical significance (p = 0.095).ConclusionsFor patients who underwent breast-conserving surgery without surgical clip implantation, the use of preoperative MR simulation in delineating the CTV of tumor bed decreased the IOV among observers. The consistency of tumor bed identification was improved especially in cases where the margins of tumor bed were challenging to visualize on CT simulation images. The study findings offer potential benefits in reducing local recurrence and minimizing tissue irritation in the surrounding areas. Future investigation in a larger patient cohort to validate our results is warranted.

Assessing rainfall radar errors with an inverse stochastic modelling framework

Abstract. Weather radar is a crucial tool for rainfall observation and forecasting, providing high-resolution estimates in both space and time. Despite this, radar rainfall estimates are subject to many error sources – including attenuation, ground clutter, beam blockage and drop-size distribution – with the true rainfall field unknown. A flexible stochastic model for simulating errors relating to the radar rainfall estimation process is implemented, inverting standard weather radar processing methods and imposing path-integrated attenuation effects, a stochastic drop-size-distribution field, and sampling and random errors. This can provide realistic weather radar images, of which we know the true rainfall field and the corrected “best-guess” rainfall field which would be obtained if they were observed in a real-world case. The structure of these errors is then investigated, with a focus on the frequency and behaviour of “rainfall shadows”. Half of the simulated weather radar images have at least 3 % of their significant rainfall rates shadowed, and 25 % have at least 45 km2 containing rainfall shadows, resulting in underestimation of the potential impacts of flooding. A model framework for investigating the behaviour of errors relating to the radar rainfall estimation process is demonstrated, with the flexible and efficient tool performing well in generating realistic weather radar images visually for a large range of event types.

Enhanced drag reduction of superhydrophobic coatings with femtosecond laser processing and TEOS/KH-SiO2/SA hybridization: CFD simulation and particle image velocimetry

Superhydrophobic surfaces have gained significant attention due to their potential applications in reducing frictional drag in various fluid flow systems. These surfaces mimic the lotus leaf's natural water-repellent properties, resulting in minimal contact between water droplets and the surface. The F-L@TEOS/KH-SiO2/SA coatings are developed using a combination of femtosecond laser ablation and chemical modification with tetraethyl orthosilicate (TEOS), KH550-modified SiO2 (KH-SiO2), and stearic acid (SA). The micro-nano structures formed on the surface enhanced its hydrophobicity, with a contact angle (CA) of 165.94°and sliding angle (SA) of 2.35°. The drag reduction performance is evaluated through experimental methods, including particle image velocimetry and gravity-driven motion analysis, as well as numerical simulations using computational fluid dynamics (CFD). Results showed that the superhydrophobic surface significantly reduced drag compared to untreated aluminum, with an average drag reduction ratio of 10.4 % in experiments and a calculated reduction rate of 9.86 % in simulations. The presence of micro-nano structures promoted the formation of an air layer at the solid-liquid interface, decreasing friction and enhancing slip flow. Additionally, the biomimetic surface generated turbulence at specific distances, aiding in energy dissipation and further reducing resistance. These findings demonstrate the potential of superhydrophobic coatings for applications requiring reduced fluid resistance, such as in marine vessels, pipelines, and fluid transport systems.

Automatic Compressive Sensing of Shack–Hartmann Sensors Based on the Vision Transformer

Shack–Hartmann wavefront sensors (SHWFSs) are crucial for detecting distortions in adaptive optics systems, but the accuracy of wavefront reconstruction is often hampered by low guide star brightness or strong atmospheric turbulence. This study introduces a new method of using the Vision Transformer model to process image information from SHWFSs. Compared with previous traditional methods, this model can assign a weight value to each subaperture by considering the position and image information of each subaperture of this sensor, and it can process to obtain wavefront reconstruction results. Comparative evaluations using simulated SHWFS light intensity images and corresponding deformable mirror command vectors demonstrate the robustness and accuracy of the Vision Transformer under various guide star magnitudes and atmospheric conditions, compared to convolutional neural networks (CNNs), represented in this study by Residual Neural Network (ResNet), which are widely used by other scholars. Notably, normalization preprocessing significantly improves the CNN performance (improving Strehl ratio by up to 0.2 under low turbulence) while having a varied impact on the Vision Transformer, improving its performance under a low turbulence intensity and high brightness (Strehl ratio up to 0.8) but deteriorating under a high turbulence intensity and low brightness (Strehl ratio reduced to about 0.05). Overall, the Vision Transformer consistently outperforms CNN models across all tested conditions, enhancing the Strehl ratio by an average of 0.2 more than CNNs.

Unlocking the potential of simulated hyperspectral imaging in agro environmental analysis: a comprehensive study of algorithmic approaches

This study focuses on identifying and evaluating the severity of powdery mildew disease in tomato plants. The uniqueness of this work lies in combining the imaging and advanced deep learning methods to develop a technique that transforms Red Green Blue (RGB) images into Simulated Hyperspectral Images (SHSI) to perform spectral and spatial analysis for precise detection and assessment of powdery mildew severity, thereby enhancing disease management. Furthermore, this research evaluates three advanced pre-trained VGG16 models, ResNet50 and EfficientNet-B7 algorithms for image preprocessing and feature extraction. Extracted features are passed to a neural network generator model to convert RGB image features into SHSIs, providing insights into the spectrum. This method enables the image analysis to perform assessments from SHSIs for health classification using Normalized Difference Vegetation Index (NDVI) values, which are meticulously compared with accurate hyperspectral data using metrics like mean absolute error (MAE) and root mean squared error (RMSE). This strategy enhances precision farming, environmental monitoring, and remote sensing accuracy. Results show that ResNet50’s architecture offers a robust framework for this study’s spectral and spatial analysis, making it a suitable choice over VGG16 and EfficientNet-B7 for transforming RGB images into SHSI. These simulated hyperspectral images offer a scalable and affordable approach for precise assessment of crop disease severity.

A novel wafer defocus measurement method for spot-scanning imaging system using laser triangulation

Abstract For spot scanning dark-field scattering technology, the defocus caused by the change in wafer surface height decreases the defect detection rate and size measurement repeatability. The demand for accurately and rapidly measuring the wafer surface height online is becoming increasingly urgent. The defect detection system integrates scattering, reflection, film thickness, and topography defect detection channels. The integration of laser-triangulation-based defocus measurement into the bright-field optical path is proposed. The laser beam is directed onto the surface of the wafer, and the reflected light is transmitted through a lens onto the photosensitive surface of a four-quadrant detector. This detector captures both the strength and position of the reflected signal simultaneously. In this paper, a triangulation-modified model is established to obtain the height of each inspection spot on the wafer surface through the position signal of the reflected image. When the signal-to-noise ratio of the simulated reflection image position is 20 dB, the signal-to-noise ratio of the height measurement resolved by the triangulation-modified model is 41 dB, and the measurement error is less than 0.2 μm, indicating that the model has a noise suppression effect. The spot-scanning imaging system is established, and the height measurement results of the triangulation measurement model are verified using a laser interferometer. Within the travel range of 69 μm, the system has a measurement accuracy better than ±1 μm and a measurement repeatability better than ±0.2 μm, which meet the measurement requirements of the spot-scanning imaging system. The system is utilized for defect detection and defocus measurement of the wafer surface, along with the motion mechanism, to accurately and rapidly obtain the height distribution of the wafer surface.

A simulation framework for preclinical proton irradiation workflow.

&#xD;The integration of proton beamlines with X-ray imaging/irradiation platforms has opened up possibilities for image-guided Bragg peak irradiations in small animals. Such irradiations allow selective targeting of normal tissue substructures and tumours. However, their small size and location pose challenges in designing experiments. This work presents a simulation framework useful for optimizing beamlines, imaging protocols, and design of animal experiments. The usage of the framework is demonstrated, mainly focusing on the imaging part.&#xD;Approach:&#xD;The fastCAT toolkit was modified with Monte Carlo (MC)-calculated primary and scatter data of a small animal imager for the simulation of micro-CT scans. The simulated CT of a mini-calibration phantom from fastCAT was validated against a full MC TOPAS CT simulation. A realistic beam model of a preclinical proton facility was obtained from beam transport simulations to create irradiation plans in matRad. Simulated CT images of a digital mouse phantom were generated using single-energy CT (SECT) and dual-energy CT (DECT) protocols and their accuracy in proton stopping power ratio (SPR) estimation and their impact on calculated proton dose distributions in a mouse were evaluated.&#xD;Main Results:&#xD;The CT numbers from fastCAT agree within 11 HU with TOPAS except for materials at the centre of the phantom. Discrepancies for central inserts are caused by beam hardening issues. The root mean square deviation in the SPR for the best SECT (90kV/Cu) and DECT (50kV/Al-90kV/Al) protocols are 3.7% and 1.0%, respectively. Dose distributions calculated for SECT and DECT datasets revealed range shifts <0.1 mm, gamma pass rates (3%/0.1mm) greater than 99%, and no substantial dosimetric differences for all structures. The outcomes suggest that SECT is sufficient for proton treatment planning in animals.&#xD;Significance:&#xD;The framework is a useful tool for the development of an optimized experimental configuration without using animals and beam time.&#xD.

Improvement of Criminisi’s Stripe Noise Suppression Method for Side-Scan Sonar Images

In response to the problem of stripe noise significantly reducing the clarity and details of side-scan sonar images due to various factors, the authors of this paper propose an improved Criminisi method for stripe noise suppression. To address the issues encountered in the Criminisi algorithm during the suppression of stripe noise in side-scan sonar images, the following steps are suggested: firstly, introduce dynamic weights in the priority calculation to adaptively adjust the confidence and data term weights based on the current patch’s texture complexity; secondly, utilize the Sobel operator in the data term calculation to capture the image edge information more accurately; and, thirdly, optimize the matching block search process by introducing the Manhattan distance in addition to the Sum of Squared Differences (SSD) criterion to further select the best matching block while transitioning from a global search to a local search. Experimental validation was conducted using simulated stripe noise images, comparing the proposed method with four traditional denoising techniques. The results demonstrate that the denoising effectiveness of the proposed method is superior, effectively restoring texture in noisy regions while preserving texture structure integrity. Ablation experiments validate the effectiveness of the proposed improvements. Denoising experiments on real noisy images show satisfactory results with this method, and combining it with Fourier transform for additional smoothing in certain cases may yield even better results.

On-Orbit Geometric Calibration and Accuracy Validation of the Jilin1-KF01B Wide-Field Camera

On-orbit geometric calibration is key to improving the geometric positioning accuracy of high-resolution optical remote sensing satellite data. Grouped calibration with geometric consistency (GCGC) is proposed in this paper for the Jilin1-KF01B satellite, which is the world’s first satellite capable of providing 150-km swath width and 0.5-m resolution data. To ensure the geometric accuracy of high-resolution image data, the GCGC method conducts grouped calibration of the time delay integration charge-coupled device (TDI CCD). Each group independently calibrates the exterior orientation elements to address the multi-time synchronization issues between imaging processing system (IPS). An additional inter-chip geometric positioning consistency constraint is used to enhance geometric positioning consistency in the overlapping areas between adjacent CCDs. By combining image simulation techniques associated with spectral bands, the calibrated panchromatic data are used to generate simulated multispectral reference band image as control data, thereby enhancing the geometric alignment consistency between panchromatic and multispectral data. Experimental results show that the average seamless stitching accuracy of the basic products after calibration is better than 0.6 pixels, the positioning accuracy without ground control points(GCPs) is better than 20 m, the band-to-band registration accuracy is better than 0.3 pixels, the average geometric alignment consistency between panchromatic and multispectral data are better than 0.25 multispectral pixels, the geometric accuracy with GCPs is better than 2.1 m, and the geometric alignment consistency accuracy of multi-temporal data are better than 2 m. The GCGC method significantly improves the quality of image data from the Jilin1-KF01B satellite and provide important references and practical experience for the geometric calibration of other large-swath high-resolution remote sensing satellites.

Investigating the Sim-to-Real Generalizability of Deep Learning Object Detection Models.

State-of-the-art object detection models need large and diverse datasets for training. As these are hard to acquire for many practical applications, training images from simulation environments gain more and more attention. A problem arises as deep learning models trained on simulation images usually have problems generalizing to real-world images shown by a sharp performance drop. Definite reasons and influences for this performance drop are not yet found. While previous work mostly investigated the influence of the data as well as the use of domain adaptation, this work provides a novel perspective by investigating the influence of the object detection model itself. Against this background, first, a corresponding measure called sim-to-real generalizability is defined, comprising the capability of an object detection model to generalize from simulation training images to real-world evaluation images. Second, 12 different deep learning-based object detection models are trained and their sim-to-real generalizability is evaluated. The models are trained with a variation of hyperparameters resulting in a total of 144 trained and evaluated versions. The results show a clear influence of the feature extractor and offer further insights and correlations. They open up future research on investigating influences on the sim-to-real generalizability of deep learning-based object detection models as well as on developing feature extractors that have better sim-to-real generalizability capabilities.

Remote Sensing Image Segmentation Based on Probabilistic Fuzzy Local Information Clustering

Abstract. In this paper, a remote sensing image segmentation based on probabilistic fuzzy local information clustering algorithm is proposed. First, assuming that the spectral measure within the same ground object follows the same probability distribution. The dissimilarity between pixel and object is modeled by the negative logarithm of the Gaussian probability density function. It can improve the noise sensitive problem caused by the Euclidean distance which can only describe the data with isotropic distribution. Then, in order to consider the effect of local spatial constraint, on the one hand, the probability measure is used to modify the local fuzzy factor to establish the dissimilarity measure with spatial constraints. On the other hand, the hidden Markov random field is used to model the prior probability model of pixel membership. Next, the entropy regularization term of the objective function is built by combining the Kullback-Leibler(KL) maximum entropy model to further improve the robustness and noise resistance. The qualitative and quantitative analysis of simulated image and different types of real remote sensing images show that the proposed algorithm can effectively overcome the above problems and further improve the accuracy of image segmentation to over 95%.

Design and 3D simulation of weft-knitted jacquard plush fabrics

Abstract Weft-knitted jacquard plush fabrics are widely used and have a wide variety of varieties, but the design and production process is complicated, time-consuming, and labor-intensive. To reduce the time and cost of fabric sampling and production, a method for designing and simulating these fabrics is proposed based on an analysis of their structural characteristics. Following the design principles of weft-knitted jacquard plush fabrics, mathematical models for pattern design and structural design were established sequentially, leading to the generation of a mathematical model for the knitting diagram. Arrays for raw material editing and threading editing were created to set various parameters. Second, loop models were established based on plush loop shapes, including a forming loop model, a U-shaped loop model, a Henkel plush loop model, a “8”-shaped plush loop model, and a O-shaped plush loop model. Finally, the effects of lighting and number of segments were investigated in THREE.Tube Geometry on the simulation’s effectiveness and speed. By comparing actual fabric images with simulation images, the feasibility of the design and simulation method was verified. This resulted in a computer-aided design process for weft-knitted jacquard plush fabrics, providing an effective method for their design and simulation. The proposed method holds significant theoretical and practical values.

Development, validation, and application of a generic image-based noise addition method for simulating reduced dose computed tomography images.

Major efforts in computed tomography (CT) have been focused on reducing radiation dose to patients while maintaining adequate diagnostic quality. To that end, research tools have been developed to simulate reduced-dose images via either image-based or projection-based methods. The former is limited to fully capturing realistic texture, streak, and non-stationary characteristics of reduced dose, while the latter is impractical clinically. To develop and validate an image-based noise addition method that accounts for such attributes while being practical in clinical settings. A noise addition method was developed to add realistic noise in the image domain. The method first estimates the noise power spectrum (NPS) of CT images, which are also forward-projected to form synthetic projections. The projection data are supplemented with random white noise proportional to their attenuation values. The noise sinogram is then back-projected onto the image, filtered by the NPS, and scaled according to the desired dose reduction level. The tool was evaluated using both phantom images and patient data. The phantom images were acquired using a multi-sized image quality phantom (Mercury Phantom 3.0, Duke University), and a thorax anthropomorphic phantom (Lungman Phantom, Kyoto Kagaku) at different dose levels and reconstruction settings. The patient images consisted of two dose levels of various CT examinations and reconstruction settings. The simulated and real reduced-dose images were compared in terms of the noise magnitude and texture (i.e., NPS average frequency, NPS-fav). The utility of this methodology was also assessed for routine clinical use for CT protocol review. For the phantom images, the percent errors in the noise magnitude between the simulated images and the actual images of the Mercury Phantom and anthropomorphic phantom images were 3.34% and 3.50%, respectively. The difference in fav was 0.07mm-1 for the Mercury Phantom and 0.06mm-1 for the anthropomorphic phantom between the simulated and actual images. The average noise magnitude percent error between the simulated and actual patient images was 4.61% with noise texture judged to be visually comparable with some kernel dependencies. When implemented clinically, the tool proved practical to simplify the process of estimating radiation dose reduction for CT protocols, resulting in a 50% dose reduction of our multiple myeloma protocol. The method generated simulated CT images with realistic noise properties similar to images acquired at the same radiation exposure without needing access to raw projection data.

Lightweight Detection of Train Underframe Bolts Based on SFCA-YOLOv8s

Improving the accuracy and detection speed of bolt recognition under the complex background of the train underframe is crucial for the safety of train operation. To achieve efficient detection, a lightweight detection method based on SFCA-YOLOv8s is proposed. The underframe bolt images are captured by a self-designed track-based inspection robot, and a dataset is constructed by mixing simulated platform images with real train underframe bolt images. By combining the C2f module with ScConv lightweight convolution and replacing the Bottleneck structure with the Faster_Block structure, the SFC2f module is designed for feature extraction to improve detection accuracy and speed. It is compared with FasterNet, GhostNet, and MobileNetV3. Additionally, the CA attention mechanism is introduced, and MPDIoU is used as the loss function of YOLOv8s. LAMP scores are used to rank the model weight parameters, and unimportant weight parameters are pruned to achieve model compression. The compressed SFCA-YOLOv8s model is compared with models such as YOLOv5s, YOLOv7, and YOLOX-s in comparative experiments. The results indicate that the final model achieves an average detection accuracy of 93.3% on the mixed dataset, with a detection speed of 261 FPS. Compared with other classical deep learning models, the improved model demonstrates superior performance in detection effectiveness, robustness, and generalization. Even in the absence of sufficient real underframe bolt images, the algorithm enables the trained network to better adapt to real environments, improving bolt recognition accuracy and detection speed, thus providing technical references and theoretical support for subsequent related research.

Effect of replanning boost radiotherapy plan in locally advanced unresectable middle to lower thoracic esophageal cancer

Thoracic bulky esophageal cancer shrinks during radiotherapy, changing the location and shape of the surrounding heart and lungs. The current study aimed to explore how replanning by volumetric-modulated arc radiotherapy (VMAT) and three-dimensional conformal radiotherapy (3DCRT) influences the target coverage and dose to organs at risk in locally advanced unresectable middle to lower thoracic esophageal cancer. We retrospectively collected CT simulation images of initial and boost radiotherapy plans for locally advanced unresectable thoracic esophageal cancer in 17 consecutive patients. First, we created boost plans of 20 Gy using 3DCRT and VMAT on the initially acquired CT images. Second, we replicated the process on CT images acquired after 20–40 Gy of radiotherapy. We then compared non-replanned boost radiotherapy plans with replanned boost plans. Replanned radiotherapy delivered more conformal doses to the target and reduced heart and lung doses. VMAT reduced more irradiated mean doses to the heart than 3DCRT in the case of replanning (1.7 and 1.1 Gy, p < 0.001). Replanning to accommodate tumor shrinkage during radiotherapy effectively lowers the irradiated doses to the heart and lungs in patients with locally advanced unresectable middle to lower thoracic esophageal cancer, especially those treated with VMAT.

Combining Fusion-Based Thresholding and Non-Linear Diffusion for Improved Speckle Noise Mitigation in SAR Images

The primary concern of synthetic aperture radar (SAR) images is speckle noise, an inherent property. The creation of speckle noise is in a granular form and its nature is multiplicative. To reduce such noise from the radar images, the researchers’ primary motive is to suppress granular pattern while preserving the quality of the obtained images, thereby facilitating easier feature extraction and classification. Existing speckle-noise reduction methods often fail to preserve fine details such as edges and textures. This study proposes a fusion-based method that integrates non-linear transform-based thresholding with advanced noise reduction techniques. The proposed method is implemented on two simulated SAR images at noise variance levels of σ = from 5 to 40. The fundamental and most significant step is to analyze the effect of granular patterns in radar images before despeckling. Different performance metrics, classified into with-reference and without-reference indexes, are considered to investigate the effectiveness of the proposed despeckle method. The Signal-to-Noise Ratio (SNR) for SAR-1 at σ = 20 was observed at 16.22 dB, outperforming the next best result of 12.89 dB from the Log Compression filter. The Universal Image Quality Index (UIQI) reached 0.6987, indicating high visual quality retention across various noise levels. The proposed despeckling method demonstrated superior performance in comparison to different filters, achieving a Peak Signal-to-Noise Ratio (PSNR) improvement of up to 29.37 dB on SAR-2 at a noise variance of σ = 5, significantly higher than the best filter method’s 26.70 dB. Additionally, the method achieved a Structural Similarity Index Measure (SSIM) of 0.6538, indicating superior image quality preservation.

Neural network-assisted localization of clustered point spread functions in single-molecule localization microscopy.

Single-molecule localization microscopy (SMLM), which has revolutionized nanoscale imaging, faces challenges in densely labelled samples due to fluorophore clustering, leading to compromised localization accuracy. In this paper, we propose a novel convolutional neural network (CNN)-assisted approach to address the issue of locating the clustered fluorophores. Our CNN is trained on a diverse data set of simulated SMLM images where it learns to predict point spread function (PSF) locations by generating Gaussian blobs as output. Through rigorous evaluation, we demonstrate significant improvements in PSF localization accuracy, especially in densely labelled samples where traditional methods struggle. In addition, we employ blob detection as a post-processing technique to refine the predicted PSF locations and enhance localization precision. Our study underscores the efficacy of CNN in addressing clustering challenges in SMLM, thereby advancing spatial resolution and enabling deeper insights into complex biologicalstructures.

Design and tests of filtering actions for an AI-based RSOs detection and tracking algorithm

Space debris represent a huge threat for the actual and future space traffic. The continuous monitoring of the space environment around the Earth and the build-up, maintenance of Resident Space Objects (RSOs) catalogue are essential to enhance the space segment safety. In the last few years, trend towards Space Based Space Surveillance (SBSS) missions was relevant. In particular, the monitoring of space debris through on-board Star Trackers (STs) coupled with Artificial Intelligence (AI) based algorithms is being studied, modeling RSOs as clear streaks on a dark background. This work presents the advances in the development of an AI-based algorithm for streaks-like RSOs Detection & Tracking for on-board optical sensors applications like STs. The initial design of the RSOs monitoring algorithm (Mastrofini et al., 2022) is extended and tested for what concerns the tracking part with new filtering actions. The goal is the improvement of the tracking outputs quality and reliability when multiple RSOs passages occur, with different configurations. Tests both on real and simulated images from STs are shown and discussed.

Radiation image reconstruction and uncertainty quantification using a Gaussian process prior

We propose a complete framework for Bayesian image reconstruction and uncertainty quantification based on a Gaussian process prior (GPP) to overcome limitations of maximum likelihood expectation maximization (ML-EM) image reconstruction algorithm. The prior distribution is constructed with a zero-mean Gaussian process (GP) with a choice of a covariance function, and a link function is used to map the Gaussian process to an image. Unlike many other maximum a posteriori approaches, our method offers highly interpretable hyperparamters that are selected automatically with the empirical Bayes method. Furthermore, the GP covariance function can be modified to incorporate a priori structural priors, enabling multi-modality imaging or contextual data fusion. Lastly, we illustrate that our approach lends itself to Bayesian uncertainty quantification techniques, such as the preconditioned Crank–Nicolson method and the Laplace approximation. The proposed framework is general and can be employed in most radiation image reconstruction problems, and we demonstrate it with simulated free-moving single detector radiation source imaging scenarios. We compare the reconstruction results from GPP and ML-EM, and show that the proposed method can significantly improve the image quality over ML-EM, all the while providing greater understanding of the source distribution via the uncertainty quantification capability. Furthermore, significant improvement of the image quality by incorporating a structural prior is illustrated.

Precise Size Determination of Supported Catalyst Nanoparticles via Generative AI and Scanning Transmission Electron Microscopy.

Transmission electron microscopy (TEM) plays a crucial role in heterogeneous catalysis for assessing the size distribution of supported metal nanoparticles. Typically, nanoparticle size is quantified by measuring the diameter under the assumption of spherical geometry, a simplification that limits the precision needed for advancing synthesis-structure-performance relationships. Currently, there is a lack of techniques that can reliably extract more meaningful information from atomically resolved TEM images, like nuclearity or geometry. Here, cycle-consistent generative adversarial networks (CycleGANs) are explored to bridge experimental and simulated images, directly linking experimental observations with information from their underlying atomic structure. Using the versatile Pt/CeO2 (Pt particles centered ≈2nm) catalyst synthesized by impregnation, large datasets of experimental scanning transmission electron micrographs and physical image simulations are created to train a CycleGAN. A subsequent size-estimation network is developed to determine the nuclearity of imaged nanoparticles, providing plausible estimates for ≈70% of experimentally observed particles. This automatic approach enables precise size determination of supported nanoparticle-based catalysts overcoming crystal orientation limitations of conventional techniques, promising high accuracy with sufficient training data. Tools like this are envisioned to be of great use in designing and characterizing catalytic materials with improved atomic precision.