Reconstruction Distortion Research Articles

No ground truth needed: unsupervised sinogram inpainting for nanoparticle electron tomography (UsiNet) to correct missing wedges

Complex natural and synthetic materials, such as subcellular organelles, device architectures in integrated circuits, and alloys with microstructural domains, require characterization methods that can investigate the morphology and physical properties of these materials in three dimensions (3D). Electron tomography has unparalleled (sub-)nm resolution in imaging 3D morphology of a material, critical for charting a relationship among synthesis, morphology, and performance. However, electron tomography has long suffered from an experimentally unavoidable missing wedge effect, which leads to undesirable and sometimes extensive distortion in the final reconstruction. Here we develop and demonstrate Unsupervised Sinogram Inpainting for Nanoparticle Electron Tomography (UsiNet) to correct missing wedges. UsiNet is the first sinogram inpainting method that can be realistically used for experimental electron tomography by circumventing the need for ground truth. We quantify its high performance using simulated electron tomography of nanoparticles (NPs). We then apply UsiNet to experimental tomographs, where >100 decahedral NPs and vastly different byproduct NPs are simultaneously reconstructed without missing wedge distortion. The reconstructed NPs are sorted based on their 3D shapes to understand the growth mechanism. Our work presents UsiNet as a potent tool to advance electron tomography, especially for heterogeneous samples and tomography datasets with large missing wedges, e.g. collected for beam sensitive materials or during temporally-resolved in-situ imaging.

Fingerprint Extraction Through Distortion Reconstruction (FEDR): A CNN-Based Approach to RF Fingerprinting

Fingerprint Extraction Through Distortion Reconstruction (FEDR): A CNN-Based Approach to RF Fingerprinting

A brain-inspired quality assessment model for sonar image super-resolution

Sonar sensors are vital in the marine industry for detecting underwater targets in challenging conditions. The imaging distance and image resolution are negatively correlated due to the propagation characteristics of sound waves in water. Although Super-Resolution (SR) techniques alleviate this limitation, they introduce complex distortions that may not fit the desired utility of reconstructed sonar images. Quantifying image quality is essential, yet existing Image Quality Assessment (IQA) algorithms fail to simultaneously consider reconstruction distortions and the sonar image task background. Furthermore, the scarcity of sonar images poses challenges for deep-learning-based algorithms. To address these issues, we propose a brain-inspired model for Super-Resolution reconstructed Sonar Image Quality Assessment (SRSIQA) based on transfer learning. On the one hand, we adopt an effective feature extractor to extract multi-level features that align with the human brain recognition process. On the other hand, we develop an SEA-Block to implement feature weight adjustment and multi-level feature scale matching, given that some transferred features do not fit the IQA task well. Experimental results demonstrate the superiority of the proposed method.

Recording and reconstruction distortion of an off-axis hologram focusing mirror in augmented reality displays

Recording and reconstruction distortion of an off-axis hologram focusing mirror in augmented reality displays

Subsampling of 3D Pixel Blocks as a Video Compression Method for Analog Transmission

Hybrid digital–analog transmission of video signals enables flexibility in dividing video information into two parts to utilize the available bandwidth better. This study proposes a compression scheme to reduce the utilized bandwidth. The scheme uses different subsampling in three-dimensional (3D) blocks, where subsampling factors are selected to minimize reconstruction distortion. The study evaluates various methods for subsampling and reconstruction to find the best combination in terms of reconstruction quality and complexity. Results show that medium-quality reconstructions can be obtained for compression ratios of about 0.125–0.3 samples per pixel.

A Digital Frequency-Dependent I/Q Imbalance and Group Delay Estimation and Compensation Modules for mmWave Single Carrier Baseband Transceivers

This paper proposes the estimation and compensation modules of digital frequency-dependent (FD) I/Q imbalance and group delay for single carrier (SC) mode with 15 Gbps throughput rate in the millimeter-wave (mmWave) band. The proposed compensation module of a digital recursive filter can mitigate FD I/Q imbalance effects caused by the -3 dB corner imbalance of the reconstruction and anti-aliasing filters of I/Q branches. The proposed recursive filter with the appropriate initial coefficients based on the result of Golay sequence estimation can shorten the convergence time and precisely acquire the desired coefficients. The average image rejection ratio (IRR) performance after compensating by the proposed module can be improved from 21.97 dB to 47.54 dB. Besides, the group delay distortion in reconstruction and anti-aliasing filters will cause intersymbol interference to distort the in-band signal. We employ a digital equalization filter with the filter coefficients derived from the correlation results based on the Golay sequence to compensate the in-band group delay distortion. After the proposed group delay is pre-compensation at TX, the system-level performance of error vector magnitude (EVM) performance is improved by 6.21 dB to -27.78 dB to satisfy the IEEE 802.11ad/ay requirement in 64-QAM mode. The inner RX BER performance with the FD I/Q imbalance and the group delay compensation modules can achieve 3 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> at 23 dB and also satisfy the system requirement at SNR of required 25.52 dB. A four-times parallelism architecture is proposed to work at a clock rate of 625 MHz with 64-QAM mode. The gate count and power consumption of the proposed design are 154.4 K and 16.2 mW, respectively.

Phase-based reconstruction optimization method for digital holographic measurement of microstructures.

Digital holography has transformative potential in measuring stacked-chip microstructures due to its noninvasive, single-shot, full-field characteristics. However, uncertainties in reconstruction distance inevitably lead to resolving blur and reconstruction distortion. Herein, we propose a phase-based reconstruction optimization method that consists of a phase-evaluation function and a structured surface-characterization model. Our proposed method involves setting a reconstruction distance range, obtaining phase information using sliced numerical reconstruction, and optimizing the reconstruction distance by finding the extreme value of the function, which identifies the focal plane of the reconstructed image. The structure of the surface topography is then characterized using the characterization model. We perform simulations of the recording, reconstruction, and characterization to verify the effectiveness of the proposed method. To further demonstrate the approach, a simple holographic recording system is constructed to measure a standard resolution target, and the measurement results are compared with a commercial instrument. The simulation and experiment demonstrate, respectively, 31.16% and 34.41% improvement in step-height characterization accuracy.

An irrefutable unambiguous insight into zygomatic air cell defect (ZACD)

Zygomatic air cell defect (ZACD) of the temporal bone, has been characterized as the auxiliary air cells in the zygomatic process and articular eminence of the temporal bone. They present as unresectable findings in the zygomatic process of the temporal bone and articular tubercle. They are crucial, as they signify regions of weak resistance and the spread of disease. Comprehension of the decrepit regions in the zygomatic bones with Cone Beam Computed Tomography (CBCT), and the imaging of choice in maxillofacial disorders for its superior 3D anatomical reconstruction and minimal distortion, are indispensable in the diagnosis, treatment planning and placement of endosseous or basal dental implants to elude complications. 154 CBCT images of subjects aged between18 and 70 years with optimal diagnostic quality and area coverage satisfying the selection criteria without the history of maxillofacial fractures, pathologies, anomalies involving the middle one third of the face were analysed. The overall prevalence of ZACD was in 16 (10.4%) subjects, with 9 in females (56.25%) and in 7 males (43.75%). The men-to-women ratio was 1:1.28. It was also revealed that bilateral incidence of ZACD was more common and the prevalence of ZACD the highest 8 (50%) in the age group of 21- 30 years. This study highlights the importance of research on the prevalence of the ZACD on CBCT, which was exiguous in the past, and compares with similar studies done using panoramic radiographs. This study paves the way for more studies employing CBCT to justify the findings we have expressed.

Quantitative optical coherence microscopy of neuron morphology in human entorhinal cortex.

The size and shape of neurons are important features indicating aging and the pathology of neurodegenerative diseases. Despite the significant advances of optical microscopy, quantitative analysis of the neuronal features in the human brain remains largely incomplete. Traditional histology on thin slices bears tremendous distortions in three-dimensional reconstruction, the magnitude of which are often greater than the structure of interest. Recently development of tissue clearing techniques enable the whole brain to be analyzed in small animals; however, the application in the human remains challenging. In this study, we present a label-free quantitative optical coherence microscopy (OCM) technique to obtain the morphological parameters of neurons in human entorhinal cortex (EC). OCM uses the intrinsic back-scattering property of tissue to identify individual neurons in 3D. The area, length, width, and orientation of individual neurons are quantified and compared between layer II and III in EC. The high-resolution mapping of neuron size, shape, and orientation shows significant differences between layer II and III neurons in EC. The results are validated by standard Nissl staining of the same samples. The quantitative OCM technique in our study offers a new solution to analyze variety of neurons and their organizations in the human brain, which opens new insights in advancing our understanding of neurodegenerative diseases.

Operando characterization and regulation of metal dissolution and redeposition dynamics near battery electrode surface.

Mn dissolution has been a long-standing, ubiquitous issue that negatively impacts the performance of Mn-based battery materials. Mn dissolution involves complex chemical and structural transformations at the electrode-electrolyte interface. The continuously evolving electrode-electrolyte interface has posed great challenges for characterizing the dynamic interfacial process and quantitatively establishing the correlation with battery performance. In this study, we visualize and quantify the temporally and spatially resolved Mn dissolution/redeposition (D/R) dynamics of electrochemically operating Mn-containing cathodes. The particle-level and electrode-level analyses reveal that the D/R dynamics is associated with distinct interfacial degradation mechanisms at different states of charge. Our results statistically differentiate the contributions of surface reconstruction and Jahn-Teller distortion to the Mn dissolution at different operating voltages. Introducing sulfonated polymers (Nafion) into composite electrodes can modulate the D/R dynamics by trapping the dissolved Mn species and rapidly establishing local Mn D/R equilibrium. This work represents an inaugural effort to pinpoint the chemical and structural transformations responsible for Mn dissolution via an operando synchrotron study and develops an effective method to regulate Mn interfacial dynamics for improving battery performance.

A geometry-aware deep network for depth estimation in monocular endoscopy

Monocular depth estimation is critical for endoscopists to perform spatial perception and 3D navigation of surgical sites. However, most of the existing methods ignore the important geometric structural consistency, which inevitably leads to performance degradation and distortion of 3D reconstruction. To address this issue, we introduce a gradient loss to penalize edge fluctuations ambiguous around stepped edge structures and a normal loss to explicitly express the sensitivity to frequently small structures, and propose a geometric consistency loss to spreads the spatial information across the sample grids to constrain the global geometric anatomy structures. In addition, we develop a synthetic RGB-Depth dataset that captures the anatomical structures under reflections and illumination variations. The proposed method is extensively validated across different datasets and clinical images and achieves mean RMSE values of 0.066 (stomach), 0.029 (small intestine), and 0.139 (colon) on the EndoSLAM dataset. The generalizability of the proposed method achieves mean RMSE values of 12.604 (T1-L1), 9.930 (T2-L2), and 13.893 (T3-L3) on the ColonDepth dataset. The experimental results show that our method exceeds previous state-of-the-art competitors and generates more consistent depth maps and reasonable anatomical structures. The quality of intraoperative 3D structure perception from endoscopic videos of the proposed method meets the accuracy requirements of video-CT registration algorithms for endoscopic navigation. The dataset and the source code will be available at https://github.com/YYM-SIA/LINGMI-MR.

Analysis of Diffractive Neural Networks for Seeing Through Random Diffusers

Imaging through diffusive media is a challenging problem, where the existing solutions heavily rely on digital computers to reconstruct distorted images. We provide a detailed analysis of a computer-free, all-optical imaging method for seeing through random, unknown phase diffusers using diffractive neural networks, covering different deep learning-based training strategies. By analyzing various diffractive networks designed to image through random diffusers with different correlation lengths, a trade-off between the image reconstruction fidelity and distortion reduction capability of the diffractive network was observed. During its training, random diffusers with a range of correlation lengths were used to improve the diffractive network's generalization performance. Increasing the number of random diffusers used in each epoch reduced the overfitting of the diffractive network's imaging performance to known diffusers. We also demonstrated that the use of additional diffractive layers improved the generalization capability to see through new, random diffusers. Finally, we introduced deliberate misalignments in training to “vaccinate” the network against random layer-to-layer shifts that might arise due to the imperfect assembly of the diffractive networks. These analyses provide a comprehensive guide in designing diffractive networks to see through random diffusers, which might profoundly impact many fields, such as biomedical imaging, atmospheric physics, and autonomous driving.

Design of hybrid La1-xCexCoO3-δ catalysts for lean methane combustion via creating active Co and Ce species

The catalytic activity of perovskite-typed catalysts depends on both the surface properties and bulk structure. Herein, we report a facile strategy to induce surface reconstruction and lattice distortion of the La1-xCexCoO3-δ catalysts via acid-etching pretreatment, ultimately realizing superior performance for catalytic combustion of lean methane, even in the presence of steam and hydrogen sulfide. The exsolution of small Ce and Co oxides (ca. 5 nm) from the perovskite were found on the surface of the La1-xCexCoO3-δ catalysts to form a hybrid structure after the acid-etching, which interact with the perovskite bulk to create more Co3+ and surface absorbed oxygen as active sites for methane combustion. It is also found that Ce species can be embedded in the crystal lattice and decorated on the surface of the catalysts after the acid-etching, which may enhance the formation of surface absorbed oxygen and the redox ability of Co3+/Co2+ pair. This results in high catalytic activity (T90 = 486 ℃) with a space velocity of 30,000 mL·g−1·h−1. Otherwise, the weak ability of hydroxyl species adsorbed on the surface of acid-etched catalysts restrain further adsorption of water molecules, improving the stability of methane combustion in the presence of water vapour. The increase in CeO2 content on the surface due to acid etching also contributes to the catalyst’s resistance to sulfur poisoning. The present work may provide a new strategy for the synthesis of hybrid perovskite catalysts with excellent catalytic performance, stability and resistance to impurities poisoning.

Induced signals in particle detectors with resistive elements: Numerically modeling novel structures (VCI 2022)

For detectors with resistive elements, the time dependence of the signals is not solely given by the movement of the charges in the drift medium but also by the time-dependent reaction of the resistive materials. In this report, we present a numerical way to capture this contribution by using the extended form of the Ramo–Shockley theorem for conductive media. As an example, the methodology will be applied to the MicroCAT two-dimensional interpolation readout to calculate the center of gravity position reconstruction distortion map of its readout cells.

Two-Way Source-Channel Coding

We propose an adaptive lossy joint source-channel coding (JSCC) scheme for sending correlated sources over two-terminal discrete-memoryless two-way channels (DM-TWCs). The main idea is to couple the independent operations of the terminals via an adaptive coding mechanism, which can mitigate cross-interference resulting from simultaneous channel transmissions and concurrently exploit the sources' correlation to reduce the end-to-end reconstruction distortions. Our adaptive JSCC scheme not only subsumes existing lossy coding methods for two-way simultaneous communication but also improves their performance. Furthermore, we derive outer bounds for our two-way lossy transmission problem and establish complete JSCC theorems in some special settings. In these special cases, a non-adaptive separate source-channel coding (SSCC) scheme achieves the optimal performance, thus simplifying the design of the source-channel communication system.

Deep depth profiling using gas cluster secondary ion mass spectrometry: Micrometer topography development and effects on depth resolution.

Secondary ion mass spectrometry using the argon cluster primary ion beam enables molecular compositional depth profiling of organic thin films with minimal loss of chemical information or changes in sputter rate. However, for depth profiles of thicker organic films (> 10 μm of sputtered depth) we have observed the rapid formation of micron-scale topography in the shape of pillars that significantly affect both the linearity of the sputter yield and depth resolution. To minimize distortions in the 3D reconstruction of the sample due to this topography, a step-wise, staggered sample rotation was employed. By using polymer spheres embedded in an organic film, it was possible to measure the depth resolution at the film-sphere interface as a function of sputtered depth and observe when possible distortions in the 3D image occurred. In this way, it was possible to quantitatively measure the effect of micron-scale topography and sample rotation on the quality of the depth profile.

Joint image and field map estimation for multi-echo hyperpolarized 13 C metabolic imaging of the heart.

Image reconstruction of metabolic images from hyperpolarized 13 C multi-echo data acquisition is sensitive to susceptibility-induced phase offsets, which are particularly challenging in the heart. A model-based framework for joint estimation of metabolite images and field map from echo shift-encoded data is proposed. Using simulations, it is demonstrated that correction of signal spilling due to incorrect decomposition of metabolites and geometrical distortions over a wide range of off-resonance gradients is possible. In vivo feasibility is illustrated using hyperpolarized [1-13 C]pyruvate in the pig heart. The model-based reconstruction for multi-echo, multicoil data was implemented as a nonconvex minimization problem jointly optimizing for metabolic images and B0 . A comprehensive simulation framework for echo shift-encoded hyperpolarized [1-13 C]pyruvate imaging was developed and applied to assess reconstruction performance and distortion correction of the proposed method. In vivo data were obtained in four pigs using hyperpolarized [1-13 C]pyruvate on a clinical 3T MR system with a six-channel receiver coil. Dynamic images were acquired during suspended ventilation using cardiac-triggered multi-echo single-shot echo-planar imaging in short-axis orientation. Simulations revealed that off-resonance gradients up to ±0.26ppm/pixel can be corrected for with reduced signal spilling and geometrical distortions yielding an accuracy of ≥90% in terms of Dice similarity index. In vivo, improved geometrical consistency (10% Dice improvement) compared to image reconstruction without field map correction and with reference to anatomical data was achieved. Joint image and field map estimation allows addressing off-resonance-induced geometrical distortions and metabolite spilling in hyperpolarized 13 C metabolic imaging of the heart.

Information Bottleneck Classification in Extremely Distributed Systems.

We present a new decentralized classification system based on a distributed architecture. This system consists of distributed nodes, each possessing their own datasets and computing modules, along with a centralized server, which provides probes to classification and aggregates the responses of nodes for a final decision. Each node, with access to its own training dataset of a given class, is trained based on an auto-encoder system consisting of a fixed data-independent encoder, a pre-trained quantizer and a class-dependent decoder. Hence, these auto-encoders are highly dependent on the class probability distribution for which the reconstruction distortion is minimized. Alternatively, when an encoding–quantizing–decoding node observes data from different distributions, unseen at training, there is a mismatch, and such a decoding is not optimal, leading to a significant increase of the reconstruction distortion. The final classification is performed at the centralized classifier that votes for the class with the minimum reconstruction distortion. In addition to the system applicability for applications facing big-data communication problems and or requiring private classification, the above distributed scheme creates a theoretical bridge to the information bottleneck principle. The proposed system demonstrates a very promising performance on basic datasets such as MNIST and FasionMNIST.

Head motion during cone-beam computed tomography: Analysis of frequency and influence on image quality.

PurposeImage artifacts caused by patient motion cause problems in cone-beam computed tomography (CBCT) because they lead to distortion of the 3-dimensional reconstruction. This prospective study was performed to quantify patient movement during CBCT acquisition and its influence on image quality.Materials and MethodsIn total, 412 patients receiving CBCT imaging were equipped with a wireless head sensor system that detected inertial, gyroscopic, and magnetometric movements with 6 dimensions of freedom. The type and amplitude of movements during CBCT acquisition were evaluated and image quality was rated in 7 different anatomical regions of interest. For continuous variables, significance was calculated using the Student t-test. A linear regression model was applied to identify associations of the type and extent of motion with image quality scores. Kappa statistics were used to assess intra- and inter-rater agreement. Chi-square testing was used to analyze the impact of age and sex on head movement.ResultsAll CBCT images were acquired in a 10-month period. In 24% of the investigations, movement was recorded (acceleration: >0.10 [m/s2]; angular velocity: >0.018 [°/s]). In all examined regions of interest, head motion during CBCT acquisition resulted in significant impairment of image quality (P<0.001). Movement in the horizontal and vertical axes was most relevant for image quality (R2>0.7).ConclusionRelevant head motions during CBCT imaging were frequently detected, leading to image quality loss and potentially impairing diagnosis and therapy planning. The presented data illustrate the need for digital correction algorithms and hardware to minimize motion artefacts in CBCT imaging.

Underwater Structure from Motion for Cameras Under Refractive Surfaces

Structure from Motion (SfM), as a three-dimensional (3D) reconstruction technique, can estimate the structure of an object by using a single moving camera. Cameras deployed in underwater environments are generally confined to waterproof housings. Thus, the light rays entering the camera are refracted twice; once at the interface between the water and the camera housing, and again at the interface between the camera housing and air. Images captured from scenes in underwater environments are prone to, and deteriorate, from distortion caused by this refraction. Severe distortions in geometric reconstruction would be caused if the refractive distortion is not properly addressed. Here, we propose a SfM approach to deal with the refraction in a camera system including a refractive surface. The impact of light refraction is precisely modeled in the refractive model. Based on the model, a new calibration and camera pose estimation method is proposed. This proposed method assists in accurate 3D reconstruction using the refractive camera system. Experiments, including simulations and real images, show that the proposed method can achieve accurate reconstruction, and effectively reduce the refractive distortion compared to conventional SfM.