Image Reconstruction Process Research Articles

Target Recognition Based on Infrared and Visible Image Fusion and Improved YOLOv8 Algorithm.

In response to the issue that the fusion process of infrared and visible images is easily affected by lighting factors, in this paper, we propose an adaptive illumination perception fusion mechanism, which was integrated into an infrared and visible image fusion network. Spatial attention mechanisms were applied to both infrared images and visible images for feature extraction. Deep convolutional neural networks were utilized for further feature information extraction. The adaptive illumination perception fusion mechanism is then integrated into the image reconstruction process to reduce the impact of lighting variations in the fused images. A Median Strengthening Channel and Spatial Attention Module (MSCS) was designed to be integrated into the backbone of YOLOv8. In this paper, we used the fusion network to create a dataset named ivifdata for training the target recognition network. The experimental results indicated that the improved YOLOv8 network saw further enhancements of 2.3%, 1.4%, and 8.2% in the Recall, mAP50, and mAP50-95 metrics, respectively. The experiments revealed that the improved YOLOv8 network has advantages in terms of recognition rate and completeness, while also reducing the rates of false negatives and false positives.

Recent advances and applications of artificial intelligence in 3D bioprinting.

3D bioprinting techniques enable the precise deposition of living cells, biomaterials, and biomolecules, emerging as a promising approach for engineering functional tissues and organs. Meanwhile, recent advances in 3D bioprinting enable researchers to build in vitro models with finely controlled and complex micro-architecture for drug screening and disease modeling. Recently, artificial intelligence (AI) has been applied to different stages of 3D bioprinting, including medical image reconstruction, bioink selection, and printing process, with both classical AI and machine learning approaches. The ability of AI to handle complex datasets, make complex computations, learn from past experiences, and optimize processes dynamically makes it an invaluable tool in advancing 3D bioprinting. The review highlights the current integration of AI in 3D bioprinting and discusses future approaches to harness the synergistic capabilities of 3D bioprinting and AI for developing personalized tissues and organs.

Advancements and Challenges in Algebraic Techniques for Medical Imaging: A Comprehensive Review of MRI and CT Scan Applications in Nigeria

Medical imaging plays a crucial role in modern healthcare by providing essential diagnostic and treatment information. This review examines the advancements and challenges of algebraic techniques in enhancing Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) scans in Nigeria. Algebraic methods, including Algebraic Reconstruction Techniques (ART), Compressed Sensing (CS), and Total Variation (TV) regularization, are integral in refining image reconstruction processes, reducing noise, and improving diagnostic accuracy. These techniques facilitate high-resolution imaging, artifact correction, and effective data fusion across different modalities. Despite the promising advancements, Nigeria faces several challenges, including high implementation costs, limited access to advanced imaging equipment, and a shortage of trained professionals. The economic barriers and disparities in technology access underscore the need for strategic investments, policy support, and collaborative initiatives to broaden the reach of these techniques. Successful case studies from institutions such as Lagos University Teaching Hospital and University College Hospital, Ibadan, highlight the tangible benefits of algebraic imaging techniques, including enhanced diagnostic clarity and reduced scan times. Training and education are essential for leveraging these advanced techniques, necessitating interdisciplinary programs and international collaborations. The integration of algebraic methods holds transformative potential for Nigeria's healthcare system, improving patient outcomes through early disease detection, precise treatment planning, and expanded access to advanced imaging services. Addressing the associated challenges through targeted efforts and continued innovation will further enhance diagnostic practices and healthcare delivery in Nigeria. Keywords: Algebraic Techniques, Medical Imaging, Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Nigeria.

Research on a multi‐scale degradation fusion network in all‐in‐one image restoration

AbstractImage restoration aims to recover high‐quality clean images from degraded low‐quality ones. Deep learning‐based approaches have been a focal point in the field of image restoration. However, most methods focus solely on a single type of degradation and may not extend well to real‐world scenarios with unknown degradation. For this purpose, the present study introduces an all‐in‐one image restoration approach by designing a multi‐scale feature fusion UNet structure (MdfUNet). In summary, the proposed method exhibits two significant advantages. For starters, it implicitly fuses degradation information across multiple scales, enabling the network to extract rich hierarchical features and enhancing its generalization ability towards unknown degradations. Secondly, MdfUnet possesses strong image reconstruction capabilities. It utilizes a simple non‐linear feature optimizer to enhance skip connections, providing rich feature representations for the image reconstruction process, and ultimately generating high‐quality restored images. Extensive experimental results show the proposed method outperforms multiple baselines on deraining, dehazing, and denoising datasets.

Lens-free reflective topography for high-resolution wafer inspection

The demand for high-resolution and large-area imaging systems for non-destructive wafer inspection has grown owing to the increasing complexity and extremely fine nature of semiconductor processes. Several studies have focused on developing high-resolution imaging systems; however, they were limited by the tradeoff between image resolution and field of view. Hence, computational imaging has arisen as an alternative method to conventional optical imaging, aimed at enhancing the aforementioned parameters. This study proposes a method for improving the resolution and field of view of an image in a lens-less reflection-type system. Our method was verified by computationally restoring the final image from diffraction images measured at various illumination positions using a visible light source. We introduced speckle illumination to expand the numerical aperture of the entire system, simultaneously improving image resolution and field of view. The image reconstruction process was accelerated by employing a convolutional neural network. Using the reconstructed phase images, we implemented high-resolution topography and demonstrated its applicability in wafer surface inspection. Furthermore, we demonstrated an ideal diffraction-limited spatial resolution of 1.7 μm over a field of view of 1.8 × 1.8 mm2 for the topographic imaging of targets with various surface roughness. The proposed approach is suitable for applications that simultaneously require high throughput and resolution, such as wafer-wide integrated metrology, owing to its compact design, cost-effectiveness, and mechanical robustness.

A new ECT image reconstruction algorithm based on Vision transformer (ViT)

Aiming at the problem of low accuracy of ECT image reconstruction, this paper proposes an ECT image reconstruction algorithm based on Vision Transformer (ViT). ViT is a method that applies the Transformer to the field of image classification, which is characterized by strong long-distance dependency learning ability and strong multi-modal fusion ability compared to CNN. This paper fully utilizes the characteristics of ViT to transform the image reconstruction process of ECT into the classification process of ViT. Here, a method is proposed to use the object field distribution as the image classification label. This method converts a two-dimensional image of the object field distribution into a one-dimensional vector, which is the label vector of the image classification. These vectors and their corresponding reconstructed images obtained by the Landweber algorithm form the sample set of ViT. Extracting a large number of flow pattern samples with various shapes and distributions through COMSOL is used as a training set. After training ViT, this network model is used to infer the labels of the predicted flow patterns. After post-processing this label, the corresponding reconstructed image can be obtained. Finally, simulation experiments are conducted, and the experiments results show that the image errors and correlation coefficients of the reconstructed images obtained through the algorithm in this paper are better than those of Tikhonov algorithm, Landweber algorithm and Long Short-Term Memory Network(LSTM). And this algorithm has better resistance to noise interference than Tikhonov algorithm, Landweber algorithm, and LSTM. This also provides a new approach and means for ECT image reconstruction.

Multilayer lensless camera for improving the condition number

Computational lensless imaging technology implements the thinning of the optical system by replacing optical imaging with image reconstruction processing. The conventional optical design uses a single coded mask and an image sensor; however, researchers have recently proposed optical designs incorporating multiple stacked coded apertures for multidimensional and wide-field imaging. Here, we investigate the effects of multilayering the coded aperture on the performance of two-dimensional spatial imaging. Through simulations and optical experiments, we demonstrate that multilayering the coded aperture enhances the condition number of the optical system’s transmission matrix and consequently improves the accuracy of image reconstruction in lensless imaging.

Spatial and temporal motion characterization for x-ray CT.

Motion induced image artifacts have been the focus of many investigations for x-ray computed tomography (CT). Methodologies of combating patient motion include the use of gating devices to optimize the data acquisition, reduction in patient scan time via faster gantry rotation and large detector coverage, and the development of advanced reconstruction and post-processing algorithms to minimize motion artifacts. Previously proposed approaches are generally "global" in nature in that motion is characterized for the entire image. It is well known, however, that the presence of motion artifact in a CT image is highly nonuniform. When there is a lack of automated and quantitative local measure indicating the presence and the severity of motion artifacts in a local region, the quality of the reconstructed images depends heavily on the CT operator's rigor and experience. Even when an operator is informed of the presence of motion, little information is provided about the nature of the motion artifact to understand its relevance to the clinical task at hand. In this paper, we propose an image-space spatial- and temporal-consistency metric (CM) to detect and characterize the local motion. In a non-rigid human organ, such as the lung, there are many small and rigid objects (target objects), such as blood vessels and nodules, distributed throughout the organ. If motion can be characterized for these target objects, we obtain a complete motion map for the organ. To accomplish this, a preliminary image reconstruction is carried out to identify the target objects and establish region-of-interests for consistency-metric calculation. The CM is then obtained based on the backprojected intensity difference between the object region and its circular background. For a stationary object, the accumulation of this quantity over views is linear. When a target object moves, nonlinear behavior exhibits and a quantitative measure of linearity indicates the severity of motion. Extensive computer simulation was utilized to confirm the validity of the theory. These tests stress the sensitivity of the proposed CM to the target object size, object shape, in-plane motion, cross-plane motion, cone-beam effect, and complex background. Results confirm that the proposed approach is robust under different testing conditions. The proposed CM is further validated using a cardiac scan of a swine, and the proposed CM correlates well with the visual inspection of the artifact in the reconstructed images. In this paper, we have demonstrated the efficacy of the proposed CM for motion detection. Unlike previously proposed approaches where the consistency condition is derived for the entire image or the entire imaging volume, the proposed metric is well localized so that different zones in a patient anatomy can be individually characterized. In addition, the proposed CM provides a quantitative measure on a view-by-view basis so that the severity of motion is consistently estimated over time. Such information can be used to optimize the image reconstruction process and minimize the motion artifact.

Acto3D: an open-source user-friendly volume rendering software for high-resolution 3D fluorescence imaging in biology.

Advances in fluorescence microscopy and tissue-clearing have revolutionised 3D imaging of fluorescently labelled tissues, organs and embryos. However, the complexity and high cost of existing software and computing solutions limit their widespread adoption, especially by researchers with limited resources. Here, we present Acto3D, an open-source software, designed to streamline the generation and analysis of high-resolution 3D images of targets labelled with multiple fluorescent probes. Acto3D provides an intuitive interface for easy 3D data import and visualisation. Although Acto3D offers straightforward 3D viewing, it performs all computations explicitly, giving users detailed control over the displayed images. Leveraging an integrated graphics processing unit, Acto3D deploys all pixel data to system memory, reducing visualisation latency. This approach facilitates accurate image reconstruction and efficient data processing in 3D, eliminating the need for expensive high-performance computers and dedicated graphics processing units. We have also introduced a method for efficiently extracting lumen structures in 3D. We have validated Acto3D by imaging mouse embryonic structures and by performing 3D reconstruction of pharyngeal arch arteries while preserving fluorescence information. Acto3D is a cost-effective and efficient platform for biological research.

Research on Human Lung Impedance Tomography Based on Soft Thresholding Image Segmentation and Reduced-Order Tikhonov Regularization

The lung is one of the most vital organs in the human body, and its condition is closely correlated with overall health. Electrical impedance tomography (EIT), as a biomedical imaging technique, often produces low-quality reconstructed images due to its inherent ill-posedness in solving the inverse problem. To address this issue, this paper proposes a soft-threshold region segmentation algorithm with a relaxation factor. This algorithm segments the reconstructed lung images into internal regions, edge regions, and background regions, resulting in clearer boundaries in the reconstructed images. This facilitates the intuitive identification of regions of interest by healthcare professionals. Additionally, this segmentation algorithm is suitably combined with a dimension-reduced Tikhonov regularization algorithm. By utilizing the joint capabilities of these algorithms, the partition points belonging to the background region can be excluded from the sought grayscale vector, thereby improving the ill-posedness of the image reconstruction process and enhancing the quality of image reconstruction. Finally, a 16-electrode human lung EIT simulation model is established for the thoracic region and verified through simulation. Experimental validation is conducted using a human lung tank simulation platform to further demonstrate the effectiveness of the proposed method.

Multifrequency Electrical Impedance Tomography Images for Swine Pneumonia Detection Using Conductivity Differences

This article introduces a new MfEIT UDESC Mark I system, which consists of a 32-electrode setup featuring a modified Howland current source, low cost, portability, and non-radiation. The system is capable of reconstructing electrical conductivity tomographic images at a rate of 30.624 frames per second, taking about 5 min for imaging. The current source employs a 0.5 mA adjacent current application pattern with frequencies ranging from 10 kHz to 1 MHz. This article outlines the hardware, firmware, and software design specifications, which include the design of the current source, calibration procedures, and image reconstruction process. Tomographic images of conductivity were reconstructed in ex vivo healthy pig lungs and those with pneumonia, as a proof of concept for future applications in live pigs. The high spectral power density, combined with real-time system calibration provides clinical advantages in veterinary medicine. The goal is to identify lung areas affected by Mycoplasma hyopneumoniae in pigs through the analysis of electrical conductivity difference, offering a valuable tool to assist veterinarians to obtain images of respiratory diseases. The modified reconstruction method GREIT (EIDORS) was evaluated with experimental data and was compared with the Gauss–Newton and Total Variation methods, where GREIT 2D proved to be superior.

Voltage-based separation of respiration and cardiac activity by harmonic analysis in electrical impedance tomography

This study aims to improve the accuracy of Electrical Impedance Tomography (EIT) measurements for monitoring ventilation and cardiac signal in medical imaging by proposing a new signal separation approach that does not require contrast agents. Conventionally, contrast agents like high-conductive saline solutions are used for signal separation in EIT measurements. This study uses a harmonic analysis on EIT raw voltage data to separate the ventilation- and cardiac-related signals (early separation). It evaluates its efficacy with a simulation model at low (1%) and high (10%) superimposed additive noise levels against the already published harmonic analysis at pixel level after EIT image reconstruction (late separation). The findings indicate that the voltage-based harmonic analysis approach, i.e., early separation, provides reliable signal separation, especially under high noise conditions, compared to the late separation. This method enables the possibility of incorporating independent cardiac-specific or ventilation-specific prior knowledge into the image reconstruction process, potentially improving the resulting images.

Heat Transfer-Inspired Network for Image Super-Resolution Reconstruction.

Image super-resolution (SR) is a critical image preprocessing task for many applications. How to recover features as accurately as possible is the focus of SR algorithms. Most existing SR methods tend to guide the image reconstruction process with gradient maps, frequency perception modules, etc. and improve the quality of recovered images from the perspective of enhancing edges, but rarely optimize the neural network structure from the system level. In this article, we conduct an in- depth exploration for the inner nature of the SR network structure. In light of the consistency between thermal particles in the thermal field and pixels in the image domain, we propose a novel heat-transfer-inspired network (HTI-Net) for image SR reconstruction based on the theoretical basis of heat transfer. With the finite difference theory, we use a second-order mixed-difference equation to redesign the residual network (ResNet), which can fully integrate multiple information to achieve better feature reuse. In addition, according to the thermal conduction differential equation (TCDE) in the thermal field, the pixel value flow equation (PVFE) in the image domain is derived to mine deep potential feature information. The experimental results on multiple standard databases demonstrate that the proposed HTI-Net has superior edge detail reconstruction effect and parameter performance compared with the existing SR methods. The experimental results on the microscope chip image (MCI) database consisting of realistic low-resolution (LR) and high-resolution (HR) images show that the proposed HTI-Net for image SR reconstruction can improve the effectiveness of the hardware Trojan detection system.

The super-resolution reconstruction algorithm of multi-scale dilated convolution residual network.

Aiming at the problems of traditional image super-resolution reconstruction algorithms in the image reconstruction process, such as small receptive field, insufficient multi-scale feature extraction, and easy loss of image feature information, a super-resolution reconstruction algorithm of multi-scale dilated convolution network based on dilated convolution is proposed in this paper. First, the algorithm extracts features from the same input image through the dilated convolution kernels of different receptive fields to obtain feature maps with different scales; then, through the residual attention dense block, further obtain the features of the original low resolution images, local residual connections are added to fuse multi-scale feature information between multiple channels, and residual nested networks and jump connections are used at the same time to speed up deep network convergence and avoid network degradation problems. Finally, deep network extraction features, and it is fused with input features to increase the nonlinear expression ability of the network to enhance the super-resolution reconstruction effect. Experimental results show that compared with Bicubic, SRCNN, ESPCN, VDSR, DRCN, LapSRN, MemNet, and DSRNet algorithms on the Set5, Set14, BSDS100, and Urban100 test sets, the proposed algorithm has improved peak signal-to-noise ratio and structural similarity, and reconstructed images. The visual effect is better.

Direct Positron Emission Imaging Using Ultrafast Timing Performance Detectors

This is an explanatory paper on Sun Il Kwon et al., Nat. Photon. 15: 914-918, 2021 and some parts of this manuscript are translated from the paper. Medical imaging modalities such as X-ray computed tomography, Magnetic resonance imaging, positron emission tomography (PET), and single photon emission computed tomography, require image reconstruction processes, consequently constraining them to form cylindrical shapes. However, among them, only PET can use additional information, so called time of flight, on an event-by-event basis. If coincidence time resolution (CTR) of PET detectors improved to 30 ps, which corresponds to spatial resolution of 4.5 mm, directly localizing electron-positron annihilation point is possible, allowing us to circumvent image reconstruction processes and free us from the geometric constraint. We call this concept direct positron emission imaging (dPEI). We have developed ultrafast radiation detectors by focusing on Cherenkov photon detection. Furthermore, the CTR of 32 ps being equivalent to 4.8 mm spatial resolution is achieved by combining deep learning-based signal processing with the detectors. In this article, we explain how we developed the detectors and demonstrated the first dPEI using different types of phantoms, how we will tackle limitations to be addressed to make the dPEI more practical, and how dPEI will emerge as an imaging modality in nuclear medicine.

ReconFormer: Accelerated MRI Reconstruction Using Recurrent Transformer.

The accelerating magnetic resonance imaging (MRI) reconstruction process is a challenging ill-posed inverse problem due to the excessive under-sampling operation in k -space. In this paper, we propose a recurrent Transformer model, namely ReconFormer, for MRI reconstruction, which can iteratively reconstruct high-fidelity magnetic resonance images from highly under-sampled k -space data (e.g., up to 8× acceleration). In particular, the proposed architecture is built upon Recurrent Pyramid Transformer Layers (RPTLs). The core design of the proposed method is Recurrent Scale-wise Attention (RSA), which jointly exploits intrinsic multi-scale information at every architecture unit as well as the dependencies of the deep feature correlation through recurrent states. Moreover, benefiting from its recurrent nature, ReconFormer is lightweight compared to other baselines and only contains 1.1 M trainable parameters. We validate the effectiveness of ReconFormer on multiple datasets with different magnetic resonance sequences and show that it achieves significant improvements over the state-of-the-art methods with better parameter efficiency. The implementation code and pre-trained weights are available at https://github.com/guopengf/ReconFormer.

Integration of an Autoencoder Model with an Actor-Oriented System

Traditional machine learning frameworks often struggle with scalability, modularity, and efficient resource management, especially when dealing with vast data. Actor-Oriented Systems offer a robust framework for building such scalable systems, allowing concurrent processing and efficient handling of large datasets. This study investigated the integration of Autoencoders (AE), which are pivotal in unsupervised learning, with Actor-Oriented Systems to enhance the modularity, scalability, and maintainability of the model training process. The study seeks to leverage the capabilities of AE and Actor-Oriented Systems to achieve high-quality image reconstruction and efficient processing. The study also attempted to understand the underlying patterns in the data, assess the performance of the model, and demonstrate the benefits of modular and scalable systems. Key findings from the results showed significant improvements in training efficiency and performance of the model, especially when using Actor-Oriented Systems. The training time was reduced from 16.96 seconds to 14.21 seconds, and the validation loss improved from 0.2768 to 0.2100, indicating better generalisation and learning. Data augmentation techniques further enhanced the robustness of the model, leading to more accurate reconstructions of the test images. Actor-Oriented Systems facilitated concurrent processing, improved modularity, and enabled the system to scale efficiently with increasing data volume. This study also highlighted the practical benefits of integrating AE with Actor-Oriented Systems, providing valuable insights into building more robust, maintainable, and scalable machine learning workflows.

FDSR: An Interpretable Frequency Division Stepwise Process Based Single-Image Super-Resolution Network.

Deep learning has excelled in single-image super-resolution (SISR) applications, yet the lack of interpretability in most deep learning-based SR networks hinders their applicability, especially in fields like medical imaging that require transparent computation. To address these problems, we present an interpretable frequency division SR network that operates in the image frequency domain. It comprises a frequency division module and a step-wise reconstruction method, which divides the image into different frequencies and performs reconstruction accordingly. We develop a frequency division loss function to ensure that each reconstruction module (ReM) operates solely at one image frequency. These methods establish an interpretable framework for SR networks, visualizing the image reconstruction process and reducing the black box nature of SR networks. Additionally, we revisited the subpixel layer upsampling process by deriving its inverse process and designing a displacement generation module. This interpretable upsampling process incorporates subpixel information and is similar to pre-upsampling frameworks. Furthermore, we develop a new ReM based on interpretable Hessian attention to enhance network performance. Extensive experiments demonstrate that our network, without the frequency division loss, outperforms state-of-the-art methods qualitatively and quantitatively. The inclusion of the frequency division loss enhances the network's interpretability and robustness, and only slightly decreases the PSNR and SSIM metrics by an average of 0.48 dB and 0.0049, respectively.

Digital micromirror device-based fast super-resolution lattice structured light illumination microscopy

Super-resolution structured illumination microscopy (SR-SIM) offers numerous advantages such as high temporal resolution, low photobleaching and phototoxicity, and no special requirements for fluorescent probes. It is particularly suitable for long-term SR imaging of living cells. By using two-dimensional lattice structured light serving as illumination, SR-SIM can achieve faster imaging speed and reduce phototoxicity, however, it is accompanied with system complexity increasing. To address this problem, in this work, we propose a fast SR lattice structured illumination microscopy imaging method based on a digital micromirror device (DMD), called DMD-Lattice-SIM. This method utilizes a DMD and synchronous time-sharing triggering with sCMOS to generate two-dimensional orthogonal lattice structured light. The proposed method only requires the collection of five phase-shifted raw images for SR image reconstruction, reducing the acquisition time by approximately 44.4% compared with the traditional SR-SIM method that requires nine phase-shifted raw images. In this work, we also introduce a rapid SR image reconstruction method called Lattice-JSFR-SIM, which combines the advantages of joint space and frequency reconstruction (JSFR)-SIM and Lattice-SIM. The raw images are pre-filtered in the frequency domain and then undergo SR reconstruction in the spatial domain. This approach reduces reconstruction time by approximately 55.6% compared with traditional frequency domain image reconstruction processing, within an imaging field of view of 512 pixels×512 pixels. The feasibility of the proposed method is demonstrated through experiments on cell microtubules and the observation of mitochondrial division and fusion in living cells. The findings presented in this paper hold great significance and application value for enabling real-time SR imaging of living cells.

Sequential Scan-Based Single-Dimension Multi-Voxel System Matrix Calibration for Open-Sided Magnetic Particle Imaging.

Open-sided magnetic particle imaging (OS-MPI) has garnered significant interest due to its potential for interventional applications. However, the system matrix calibration in OS-MPI using sequential scans is a time-consuming task and susceptible to the low signal-to-noise ratio (SNR) resulting from the small calibration sample size. These challenges have hindered the practical implementation of system matrix-based reconstruction for sequentially scanned OS-MPI. To address these issues, we propose a novel calibration method, named sequential scan-based single-dimension multi-voxel calibration (SS-SDMVC), to efficiently obtain a high-SNR system matrix. This method was implemented in a cylindrical field of view (FOV), where a bar calibration sample parallel to the field-free line (FFL) was shifted along a fixed radial direction. A standard image reconstruction process was also introduced to verify the feasibility of SS-SDMVC. Through simulations, we analyzed the effects of noise levels and scanner imperfections on the SS-SDMVC-based reconstruction and demonstrated its robustness. In experiments, we compared the imaging performance of SS-SDMVC and the sequential scan-based traditional cubic-FOV SMC. The results showed that SS-SDMVC reduced the number of measurements by a factor of 210.94 and achieved higher reconstruction quality. Therefore, SS-SDMVC is expected to improve the reconstruction quality of human-scale or high-gradient FFL MPI scanners.