Abstract The x-ray attenuation coefficient of most materials is energy-dependent, and this dependence varies among materials. Moreover, the x-ray beam in a typical computed tomography (CT) scanner has a broad energy spectrum. Therefore, image reconstruction from x-ray CT data is a nonlinear problem. If the nonlinear nature of CT data is ignored and a linear reconstruction formula is used, which is frequently the case, the resulting image contains beam-hardening artifacts such as streaks. In this work, we describe the nonlinearity of CT data using a conventional and widely accepted model. Our main contribution is the characterization of streak artifacts arising from this nonlinearity. We also obtain an explicit expression for the leading singular behavior of the reconstruction near the streaks. Finally, numerical experiments are conducted to validate the theoretical results.

Time-difference electrical impedance tomography (EIT) is gaining widespread use for bedside lung monitoring in intensive care patients suffering from lung-related diseases. It involves collecting voltage measurements from electrodes placed on the patient's thorax, which are then used to reconstruct impedance images. This study investigates how incorporating anatomical information from CT data into the widely used GREIT reconstruction algorithm affects EIT images and improves their interpretability. Based on clinically motivated lung state scenarios, we simulated EIT measurements to assess how the GREIT parameters influence the result of EIT image reconstruction, particularly with respect to noise performance and image accuracy. We introduce quality measures that allow us to perform a quantitative assessment of reconstruction quality. We incorporate the anatomical features of a patient from CT data by customizing the background conductivity and the distribution of GREIT training targets. Our analysis confirmed that unphysiological background conductivity assumptions can lead to misleading EIT images, whereas physiological values, although more accurate, come with higher noise sensitivity. By increasing the number of GREIT training targets inside the lung and adapting the respective weighting radius, we significantly improved the anatomical accuracy of the EIT images. When applied to clinical EIT data from a representative ARDS patient, these adjustments in the reconstruction setup substantially enhanced the interpretability of the resulting EIT images. Incorporating CT-based anatomical data in the GREIT reconstruction significantly enhances the clinical applicability of EIT in lung monitoring. The improved interpretability of EIT images facilitates better-informed clinical decisions and the individualized adjustment of ventilation strategies for critically ill patients.

Microwave imaging (MWI) is a non-invasive technique for visualizing the anomalies of biological tissues. The imaging process is accomplished by comparing the electrical parameters of healthy tissues and malignant tissues. This work introduces a microwave imaging system for tumor detection in breast tissue. The experiment is performed in a homogeneous background medium, where a high dielectric contrast material is used to mimic the tumor. The proposed imaging system is experimentally evaluated for multiple tumor locations and sizes using a horn antenna. Reflection coefficients obtained from the monostatic configuration of the horn antenna are used for image reconstruction. The evaluation metrics, such as localization error, absolute area error, DICE score, Intersection over Union (IoU), precision, accuracy, sensitivity and specificity, are computed from the reconstructed image. A modified version of the beamforming algorithm improves the quality of reconstructed images by providing a minimum accuracy of 96% for all test cases, with an evaluation time of less than 48 s. The proposed methodology shows promising results under a controlled environment and can be implemented for clinical applications after adequate biological studies. This methodology can be used to calibrate any antenna system or phantom, as it has high contrast in conductivity, leading to better imaging. The present study contributes to Sustainable Development Goal (SDG) 3 by ensuring healthy lives and promoting wellbeing for all ages.

Carotid CT angiography (CTA) is valuable for diagnosing carotid artery disease but involves radiation and contrast agent risks. Deep Learning Image Reconstruction (DLIR-H) shows potential for maintaining image quality in low-dose protocols. In this prospective study, 180 patients undergoing dual-energy CTA were divided into three groups: a control group (ASIR-V 50%, NI = 4, contrast = 0.5mL/kg), a low-dose group (DLIR-H, NI = 11, contrast = 0.5mL/kg), and an ultra-low-dose group (DLIR-H, NI = 13, contrast = 0.4mL/kg). Objective (CTV[CT values], noise, SNR, CNR) and subjective (5-point Likert scale) image quality were evaluated. The ultra-low-dose group achieved a 20.3% reduction in contrast volume and a 53.3% reduction in effective dose compared to the control group (P < 0.001). Both experimental groups showed lower noise and higher CNR/SNR (except at aortic arch) than controls. However, the ultra-low-dose group had significantly lower CNR/SNR than the low-dose group (P < 0.05). Subjective image quality was superior in both experimental groups (P < 0.001), with high inter-rater agreement. DLIR-H outperformed ASIR-V in low and ultra-low-dose protocols but could not fully compensate for image quality degradation when radiation and contrast were further reduced.

Abstract In this paper, we introduce snnTrans-DHZ, a lightweight spiking transformer architecture that incorporates a learnable threshold membrane potential mechanism for underwater image dehazing. Despite its compact design with 0.57 M parameters, the method substantially improves underwater image clarity and visibility. Leveraging the temporal dynamics of SNNs, snnTrans-DHZ efficiently processes time-dependent raw image sequences while maintaining low power consumption. The raw underwater images are first converted into time-dependent image sequences by repeatedly passing the static image to a user-defined timestep value. The RGB sequences are then converted into LAB color space representations and processed simultaneously. The architecture integrates three primary modules: (i) K estimator module to extract features from different color space representations, (ii) Background Light Estimator module to jointly estimate the background light component from the RGB-LAB color space representations, and (iii) Soft Image Reconstruction to reconstruct the haze-free, visibility-enhanced image. The snnTrans-DHZ model is directly trained using surrogate gradient-based backpropagation through time (BPTT) strategy. In this research, a combined loss function is designed and used. Our model is trained and tested on the UIEB and EUVP, the two publicly available benchmark dataset for image dehazing. Our algorithm achieves a PSNR of 21.6773 dB and SSIM of 0.8795 on UIEB and, on EUVP, it achieves 23.4562 dB and 0.8439. snnTrans-DHZ algorithm achieves this algorithmic performance with fewer operations (7.42 GSOPs) and lower energy consumption of 0.0151 J compared to existing state-of-the-art image enhancement methods. It provides a 3.3× improvement in energy efficiency over the lightest state-of-the-art transformer-based method, making it suitable for underwater robotics and environmental monitoring. The source code is available at snnTrans-DHZ.

Next-generation neuromorphic systems require hardware platforms that seamlessly integrate sensing, memory, and computation. Here, we present a light-programmable optoelectronic memristor based on an ITO/IGZO/W structure, capable of emulating a broad spectrum of synaptic and neuronal functions under purely optical stimulation through the transparent ITO top electrode. The device exhibits short-term plasticity, including excitatory postsynaptic current (EPSC), paired-pulse facilitation (PPF), and spike-dependent learning behaviors (SADP, SWDP, SNDP). It also replicates nociceptive responses such as threshold activation, no adaptation, relaxation, and sensitization. Pavlovian associative learning is demonstrated using optical stimuli, showing acquisition, extinction, and recovery behaviors driven by persistent photoconductivity. Furthermore, a 4-bit optical pulse-driven reservoir computing architecture achieves 97.005% MNIST classification accuracy through a convolutional neural network readout. A light-induced stochastic activation function, extracted from threshold-switching behavior, is applied in a Restricted Boltzmann Machine to model probabilistic neurons, reaching 96.35% image reconstruction accuracy. Postforming optical modulation enables light-intensity-dependent trap/detrap dynamics and fine-tuning of the conductive filament. These results highlight the proposed IGZO-based optoelectronic memristor as a versatile and energy-efficient platform for multifunctional neuromorphic computing, combining sensory, deterministic, and probabilistic intelligence in a single reconfigurable device.

The use of high-performance gradient coils results in stronger spatially dependent second-order concomitant magnetic fields, which can lead to signal dropout, blurring artifacts and phase errors that become more significant at locations farther from the gradient isocenter. A correction coil-based method for prospectively compensating second-order concomitant fields in higher-performance gradient systems is described. An insertable, axially symmetric second-order field coil to prospectively correct for second-order concomitant field-induced phase errors on a high-performance head-only gradient system at 3.0T was developed. The efficacy of the second-order concomitant-field correction was demonstrated in phantom and healthy volunteer scans using 2D phase contrast (PC) and spiral gradient echo (GRE) imaging. By employing the correction coil, there was a significant reduction in second-order concomitant field-induced blurring in the 2D spiral images, and reduction in phase errors and signal degradation in the GRE PC images. In the single-sided PC, the z- and radially-directed second-order concomitant phase accrued in the coronal and axial PC acquisition was reduced by 100% and 83%, respectively. Signal enhancement up to 968.9% was obtained in the two-sided PC acquisitions. In spiral GRE images, blurring was reduced by ˜40.2% at 60 mm from the gradient isocenter in a phantom. Correspondingly, the reduction in concomitant field-induced blurring in in-vivo spiral GRE images was noted with the correction coil. The described second-order correction coil insert prospectively compensates erroneous phase accruals due to second-order concomitant fields on a high-performance gradient system at the source, complementing or replacing software corrections/compensations during image reconstruction.

Attention-driven complementary information fusion network for sparse photoacoustic image reconstruction.

The structured illumination microscopy (SIM) technique, when applied under low photon efficiency, provides an effective solution for rapid live-cell imaging, thereby enabling the investigation of dynamic cellular processes. However, noise interference during the acquisition process significantly hinders the reconstruction of SIM images, leading to substantial artifacts. To address this challenge, we propose a zero-shot learning-based SIM image denoising method (ZS-SIM). This approach relies solely on a single acquisition of noisy SIM data and achieves accurate denoising through neural network training. The original SIM image stack is downsampled and interpolated to complete the resampling process, while the traditional Wiener-SIM reconstruction method is integrated to ensure physical fidelity. We introduce a symmetric reconstruction loss and a mutual constraint SSIM loss that jointly enhance training stability and accelerate convergence, as demonstrated by our convergence analysis. ZS-SIM further achieves a favorable balance between denoising quality and computational efficiency, with low model complexity and fast inference speed, making it well-suited for practical deployment in microscopy workflows. Experimental results demonstrate that ZS-SIM efficiently and rapidly achieves artifact-free, high-fidelity denoising reconstruction, making it particularly well-suited for low-photon efficiency live-cell imaging and scenarios with limited computational resources. Furthermore, by extending the method to scanning electron microscopy (SEM) data, we validate the effectiveness of ZS-SIM for SEM data denoising, significantly enhancing the performance of downstream segmentation tasks. We anticipate that ZS-SIM will play a pivotal role in low-photon efficiency imaging, driving advancements in this field and providing crucial support for rapid validation in biomedical research, thereby overcoming the challenges posed by acquisition noise.

Automated 3D reconstruction of high-resolution images for tunnel inspection and monitoring purposes

Factorization model with total variation regularizer for image reconstruction and subgradient algorithm

DRNT2LRNet: A new model-driven deep unrolling network for high-quality compressive hyperspectral image reconstruction

Enhancing and accelerating brain MRI through deep learning reconstruction using prior subject-specific imaging.

Multi-contrast magnetic resonance image reconstruction joint with one-dimensional or two-dimensional sampling patterns optimization

Impact of a new deep-learning image reconstruction algorithm on potential dose reduction and quality of chest CT images: a phantom study.

This study explores cross-cultural narrative reconstruction of female images in the 2021 Uzbek Liaozhai translation via Mona Baker’s theory.

Volume electron microscopy (vEM) is a powerful technique that enables 3D visualization of biological structures at the nanometer scale. However, vEM imaging relies on sequential scanning of 2D images, and due to sectionthickness limitations, the axial resolution is significantly lower than the lateral resolution. In this paper, we propose the vEMINR, an ultra-fast isotropic reconstruction method based on implicit neural representation (INR). This method enhances the reconstruction quality of vEM images by learning the true degradation patterns of low-resolution images, and significantly accelerates the reconstruction process by utilizing the efficient parameterization and a continuous function representation of INR. In experiments on 11 public datasets, vEMINR outperforms mainstream methods with over tenfold faster reconstruction and higher accuracy. vEMINR substantially improved the accuracy of organelle and neuron reconstruction from vEM. Overall, the excellent reconstruction time efficiency of vEMINR enables high-throughput processing of terabyte-scale vEM datasets while maintaining reconstruction accuracy. We believe that it will play a significant role in large-scale vEM image reconstruction and related research fields.

This manuscript is a formal response to the Comment by Wagner etal. regarding our publication "De Novo Reconstruction of 3D Human Facial Images from DNA Sequence." We clarify the methodological rationale, analytical procedures, and scientific scope of our original study, and we address several misconceptions arising in the Comment. We further highlight the intended purpose and limitations of Difface and discuss ongoing efforts to advance the rigor, interpretability, and ethical governance of DNA-based 3D facial prediction.

Objective.Normalized metal artifact reduction (NMAR) is a robust and widely used method for reducing metal artifacts in computed tomography (CT). However, conventional NMAR requires at least two forward projections, one for metal trace detection and the other for prior sinogram generation, resulting in redundant computation and limited efficiency. This study aims to reformulate NMAR into a single forward projection-based framework that maintains artifact reduction performance while improving computational efficiency and structural simplicity.Approach.We show that the two separate forward projections in NMAR can be unified into a single operation by leveraging deep learning (DL) priors, thereby eliminating the explicit forward projection for metal trace. The metal trace is inferred directly from localized discrepancies between the original sinogram and the forward projection of the DL prior image, allowing both interpolation and trace identification within a unified forward projection. Simulations and cadaver experiments were performed to compare the proposed method with NMAR, DL reconstruction, and conventional DL-NMAR.Main results.The proposed method reduced metal artifacts with image quality comparable to conventional DL-NMAR while improving computational efficiency. By reducing the number of forward projections from two to one, the proposed method achieved the lowest number of projection operations among all compared methods, highlighting its computational advantage.Significance.This study demonstrates that DL priors can be seamlessly integrated into physics-based NMAR frameworks to simplify image reconstruction pipelines and enhance computational performance. The proposed unified forward projection provides an efficient solution to accelerate MAR in CT imaging.

With the rapid progression of photoelectrical technology, the development of multi-mode photodetectors is highly desirable for complex application scenarios. Here, we fabricate 1T-2H mixed-phase MoS2 films via electrochemical intercalation technique, which intrinsically integrates a 2H-phase photosensitive network with a 1T-phase dispersed capacitance. By modulating the external bias and phase distribution, a family of photodetectors with tailored tri-mode operation was achieved among reconfigured photocapacitive (PCC), photovoltaic (PV), photoconductive (PC) modes. Breaking the conventional perception of capacitors as parasitic components in optoelectronic devices, our work harnessed the capacitive effect as an effective signal source, enabling self-powered operation and high-sensitivity detection. It exhibits efficient charging pulses and superior self-resetting characteristics with a response speed reaching 19.2ms and an average charge-discharge efficiency of 87.6%. From device design to application prototype, our proof-of-concept demonstrations in non-contact sensing, flame monitoring, and image reconstruction highlight the positive potential in intelligent sensing technologies. This study not only provides a novel design paradigm for phase engineering in multi-functional photodetection, but also its successful implementations hold significant promising for advancing intelligent optoelectronics, such as optoelectronic chips and biosensing.