In order to improve the reconstruction accuracy of magnetic resonance imaging(MRI),an accurate natural image compressed sensing(CS)reconstruction network is proposed,which combines the advantages of model-based and de...In order to improve the reconstruction accuracy of magnetic resonance imaging(MRI),an accurate natural image compressed sensing(CS)reconstruction network is proposed,which combines the advantages of model-based and deep learning-based CS-MRI methods.In theory,enhancing geometric texture details in linear reconstruction is possible.First,the optimization problem is decomposed into two problems:linear approximation and geometric compensation.Aimed at the problem of image linear approximation,the data consistency module is used to deal with it.Since the processing process will lose texture details,a neural network layer that explicitly combines image and frequency feature representation is proposed,which is named butterfly dilated geometric distillation network.The network introduces the idea of butterfly operation,skillfully integrates the features of image domain and frequency domain,and avoids the loss of texture details when extracting features in a single domain.Finally,a channel feature fusion module is designed by combining channel attention mechanism and dilated convolution.The attention of the channel makes the final output feature map focus on the more important part,thus improving the feature representation ability.The dilated convolution enlarges the receptive field,thereby obtaining more dense image feature data.The experimental results show that the peak signal-to-noise ratio of the network is 5.43 dB,5.24 dB and 3.89 dB higher than that of ISTA-Net+,FISTA and DGDN networks on the brain data set with a Cartesian sampling mask CS ratio of 10%.展开更多
Microwave diffraction tomography is a process to infer the internal structure of an objectfrom multiple angle views of microwave diffraction shadow. Being sensitive to variations in refractive index of the object, the...Microwave diffraction tomography is a process to infer the internal structure of an objectfrom multiple angle views of microwave diffraction shadow. Being sensitive to variations in refractive index of the object, the procedure can be used to measure permittivity distributions within dielectric objects and to image soft tissues for biomedical applications. The optimal resolution distance obtainable is half a wavelength, but this can rarely be achieved because of practical limitations. Some procedures, however, are available to improve the practical resolution. One, which is suitable for microwave tomography, is to use multiple angle views data and to combine the resulting images. The other, which is suitable for improving the image reconstruction resolution, is to use the digital filtering technique and the filtered backpropagation algorithm. A system operating over the X-band microwave frequency is described and some experimental results for objects in air are given.展开更多
Background:Medical imaging advancements are constrained by fundamental trade-offs between acquisition speed,radiation dose,and image quality,forcing clinicians to work with noisy,incomplete data.Existing reconstructio...Background:Medical imaging advancements are constrained by fundamental trade-offs between acquisition speed,radiation dose,and image quality,forcing clinicians to work with noisy,incomplete data.Existing reconstruction methods either compromise on accuracy with iterative algorithms or suffer from limited generalizability with task-specific deep learning approaches.Methods:We present LDM-PIR,a lightweight physics-conditioned diffusion multi-model for medical image reconstruction that addresses key challenges in magnetic resonance imaging(MRI),CT,and low-photon imaging.Unlike traditional iterative methods,which are computationally expensive,or task-specific deep learning approaches lacking generalizability,integrates three innovations.A physics-conditioned diffusion framework that embeds acquisition operators(Fourier/Radon transforms)and noise models directly into the reconstruction process.A multi-model architecture that unifies denoising,inpainting,and super-resolution via shared weight conditioning.A lightweight design(2.1M parameters)enabling rapid inference(0.8s/image on GPU).Through self-supervised fine-tuning with measurement consistency losses adapts to new imaging modalities using fewer annotated samples.Results:Achieves state-of-the-art performance on fastMRI(peak signal-to-noise ratio(PSNR):34.04 for single-coil/31.50 for multi-coil)and Lung Image Database Consortium and Image Database Resource Initiative(28.83 PSNR under Poisson noise).Clinical evaluations demonstrate superior preservation of anatomical structures,with SSIM improvements of 8.8%for single-coil and 4.36%for multi-coil MRI over uDPIR.Conclusion:It offers a flexible,efficient,and scalable solution for medical image reconstruction,addressing the challenges of noise,undersampling,and modality generalization.The model’s lightweight design allows for rapid inference,while its self-supervised fine-tuning capability minimizes reliance on large annotated datasets,making it suitable for real-world clinical applications.展开更多
We present a prototype for hybrid Compton and positron emission tomography(PET)imaging aimed at enhancing data utilization and enabling concurrent imaging of multiple radiopharmaceuticals.The prototype comprises two d...We present a prototype for hybrid Compton and positron emission tomography(PET)imaging aimed at enhancing data utilization and enabling concurrent imaging of multiple radiopharmaceuticals.The prototype comprises two detectors that utilize LYSO-SiPM and were available in our laboratory.One detector consists of a 50×50 array of LYSO crystals,each measuring 0.9mm×0.9mm×10mm with 1 mm pitches,whereas the other detector comprises a 25×25 array of LYSO crystals,each measuring 1.9mm×1.9mm×10mm with 2 mm pitches.These detectors are mounted on a rotational stage,which enables them to function as either a Compton camera or a PET detector pair.The 64-channel signals from the SiPMs of each detector are processed through a capacitive multiplexing circuit to yield four position-weighted outputs.Distinct energy windows were used to discriminate Compton events from PET events.Energy resolution and energy-channel relationships were calibrated via multiple sources.The measured average energy resolutions(full widths at half maximum,FWHMs)for the detectors at 511 keV were 17.5%and 15.2%,respectively.The initial experimental results indicate an angular resolution(FWHM)of 8.6◦for the system in Compton imaging mode.A V-shaped tube injected with 18 F solution was clearly reconstructed,which further verified the imaging capabilities of the system in Compton imaging mode.The results of simulation and experimental imaging studies show that the system can detect tumors as small as 1 mm in diameter when working in PET imaging mode.Mouse bone PET imaging was successfully conducted,with the results matching well with the corresponding CT images.This technology holds great potential for advancing the development of physiological function modalities.展开更多
In this paper, we improve traditional generative adversarial networks (GAN) with reference to residual networks and convolutional neural networks to facilitate the reconstruction of complex objects that cannot be reco...In this paper, we improve traditional generative adversarial networks (GAN) with reference to residual networks and convolutional neural networks to facilitate the reconstruction of complex objects that cannot be reconstructed by traditional associative imaging methods. Unlike traditional ghost imaging to reconstruct objects from bucket signals, our proposed method can use simple objects (such as EMNIST) as a training set for GAN, and then recognize objects (such as faces) of completely different complexity than the training set. We use traditional ghost imaging and neural network to reconstruct target objects respectively. According to the research results in this paper, the method based on neural network can reconstruct complex objects very well, but the method based on traditional ghost imaging cannot reconstruct complex objects. The research scheme in this paper is of great significance for the reconstruction of complex object-related imaging under low sampling conditions.展开更多
For the nonuniform microscan system where the interframe translation is no longer equivalent to accurate halfpixel, a 2-dimension non-interpolated subpixel algorithm is proposed to consider arhitrary value of microsca...For the nonuniform microscan system where the interframe translation is no longer equivalent to accurate halfpixel, a 2-dimension non-interpolated subpixel algorithm is proposed to consider arhitrary value of microscanning. The aim of the proposed algorithm is to restore double resolution from 2 × 2 undersampled frames. To solve the ill-posed problem in the process of image reconstruction, the prior boundary condition is introduced, and the proposed subpixel reconstruction algorithm is reformulated into the form of line-by-line backward propagation iteration for reducing the calculation complexity. Since the direction of movement offset relative to the accurate halfpixel determines the transfer matrix of the image degradation process, the recon- struction is classified into 4 types when 2 × 2 mieroscan model is applied. All the simulation results and experiment data demonstrate the double resolution improvement compared with the undersampled images. The focus problem, scarcely any possibility of the operation with accurate halfpixel micromotion, is figured out for en-hancing the feasibility of subpixel reconstruction used in practice.展开更多
By obtaining a feasible filter function,reconstructed images can be got with linear interpolation and liftered backprojection techniques.Considering the gray and spstial correlation neighbour informations of each pixe...By obtaining a feasible filter function,reconstructed images can be got with linear interpolation and liftered backprojection techniques.Considering the gray and spstial correlation neighbour informations of each pixel,a new supervised classification method is put forward for the reconstructed images,and an experiment with noise image is done,the result shows that the method is feasible and accurate compared with ideal phantoms.展开更多
Magnetic particle imaging(MPI)is an emerging molecular imaging technique with high sensitivity and temporal-spatial resolution.Image reconstruction is an important research topic in MPI,which converts an induced volta...Magnetic particle imaging(MPI)is an emerging molecular imaging technique with high sensitivity and temporal-spatial resolution.Image reconstruction is an important research topic in MPI,which converts an induced voltage signal into the image of superparamagnetic iron oxide particles concentration distribution.MPI reconstruction primarily involves system matrix-and x-space-based methods.In this review,we provide a detailed overview of the research status and future research trends of these two methods.In addition,we review the application of deep learning methods in MPI reconstruction and the current open sources of MPI.Finally,research opinions on MPI reconstruction are presented.We hope this review promotes the use of MPI in clinical applications.展开更多
The state-of-the-art approaches for image reconstruction using under-sampled k-space data are compressed sensing based.They are iterative algorithms that optimize objective functions with spatial and/or temporal const...The state-of-the-art approaches for image reconstruction using under-sampled k-space data are compressed sensing based.They are iterative algorithms that optimize objective functions with spatial and/or temporal constraints.This paper proposes a non-iterative algorithm to estimate the un-measured data and then to reconstruct the image with the efficient filtered backprojection algorithm.The feasibility of the proposed method is demonstrated with a patient magnetic resonance imaging study.The proposed method is also compared with the state-of-the-art iterative compressed-sensing image reconstruction method using the total-variation optimization norm.展开更多
Electrical Impedance Tomography (EIT) is a medical imaging technique which can be used to monitor the regional ventilation in patients utilizing voltage measurements made at the thorax. Several reconstruction algorith...Electrical Impedance Tomography (EIT) is a medical imaging technique which can be used to monitor the regional ventilation in patients utilizing voltage measurements made at the thorax. Several reconstruction algorithms have been developed during the last few years. In this manuscript we compare a well-established algorithm and a re-cently developed method for image reconstruction regarding EIT indices derived from the differently reconstructed images.展开更多
Based on a strong inter-diagonal matrix and Taylor series expansions,an oversample reconstruction method was proposed to calibrate the optical micro-scanning error. The technique can obtain regular 2 ×2 microscan...Based on a strong inter-diagonal matrix and Taylor series expansions,an oversample reconstruction method was proposed to calibrate the optical micro-scanning error. The technique can obtain regular 2 ×2 microscanning undersampling images from the real irregular undersampling images,and can then obtain a high spatial oversample resolution image. Simulations and experiments show that the proposed technique can reduce optical micro-scanning error and improve the system＇s spatial resolution. The algorithm is simple,fast and has low computational complexity. It can also be applied to other electro-optical imaging systems to improve their spatial resolution and has a widespread application prospect.展开更多
In this editorial,we review recent progress on the applications of artificial intelligence(AI)in image reconstruction for biomedical imaging.Because it abandons prior information of traditional artificial design and a...In this editorial,we review recent progress on the applications of artificial intelligence(AI)in image reconstruction for biomedical imaging.Because it abandons prior information of traditional artificial design and adopts a completely data-driven mode to obtain deeper prior information via learning,AI technology plays an increasingly important role in biomedical image reconstruction.The combination of AI technology and the biomedical image reconstruction method has become a hotspot in the field.Favoring AI,the performance of biomedical image reconstruction has been improved in terms of accuracy,resolution,imaging speed,etc.We specifically focus on how to use AI technology to improve the performance of biomedical image reconstruction,and propose possible future directions in this field.展开更多
Image super-resolution reconstruction technology is currently widely used in medical imaging,video surveillance,and industrial quality inspection.It not only enhances image quality but also improves details and visual...Image super-resolution reconstruction technology is currently widely used in medical imaging,video surveillance,and industrial quality inspection.It not only enhances image quality but also improves details and visual perception,significantly increasing the utility of low-resolution images.In this study,an improved image superresolution reconstruction model based on Generative Adversarial Networks(SRGAN)was proposed.This model introduced a channel and spatial attention mechanism(CSAB)in the generator,allowing it to effectively leverage the information from the input image to enhance feature representations and capture important details.The discriminator was designed with an improved PatchGAN architecture,which more accurately captured local details and texture information of the image.With these enhanced generator and discriminator architectures and an optimized loss function design,this method demonstrated superior performance in image quality assessment metrics.Experimental results showed that this model outperforms traditional methods,presenting more detailed and realistic image details in the visual effects.展开更多
Multispectral imaging,which simultaneously captures the spatial and spectral information of a scene,is widely used across diverse fields,including remote sensing,biomedical imaging,and agricultural monitoring.We intro...Multispectral imaging,which simultaneously captures the spatial and spectral information of a scene,is widely used across diverse fields,including remote sensing,biomedical imaging,and agricultural monitoring.We introduce a snapshot multispectral imaging approach employing a standard monochrome image sensor with no additional spectral filters or customized components.Our system leverages the inherent chromatic aberration of wavelength-dependent defocusing as a natural source of physical encoding of multispectral information;this encoded image information is rapidly decoded via a deep learning-based multispectral Fourier imager network(mFIN).We experimentally tested our method with six illumination bands and demonstrated an overall accuracy of 98.25%for predicting the illumination channels at the input and achieved a robust multispectral image reconstruction on various test objects.This deep learning-powered framework achieves high-quality multispectral image reconstruction using snapshot image acquisition with a monochrome image sensor and could be useful for applications in biomedicine,industrial quality control,and agriculture,among others.展开更多
For image compression sensing reconstruction,most algorithms use the method of reconstructing image blocks one by one and stacking many convolutional layers,which usually have defects of obvious block effects,high com...For image compression sensing reconstruction,most algorithms use the method of reconstructing image blocks one by one and stacking many convolutional layers,which usually have defects of obvious block effects,high computational complexity,and long reconstruction time.An image compressed sensing reconstruction network based on self-attention mechanism(SAMNet)was proposed.For the compressed sampling,self-attention convolution was designed,which was conducive to capturing richer features,so that the compressed sensing measurement value retained more image structure information.For the reconstruction,a self-attention mechanism was introduced in the convolutional neural network.A reconstruction network including residual blocks,bottleneck transformer(BoTNet),and dense blocks was proposed,which strengthened the transfer of image features and reduced the amount of parameters dramatically.Under the Set5 dataset,when the measurement rates are 0.01,0.04,0.10,and 0.25,the average peak signal-to-noise ratio(PSNR)of SAMNet is improved by 1.27,1.23,0.50,and 0.15 dB,respectively,compared to the CSNet+.The running time of reconstructing a 256×256 image is reduced by 0.1473,0.1789,0.2310,and 0.2524 s compared to ReconNet.Experimental results showed that SAMNet improved the quality of reconstructed images and reduced the reconstruction time.展开更多
Background:Quantum-enhanced medical imaging algorithms–quantum entanglement reconstruction,quantum noise suppression,and quantum beamforming–propose possible remedies for significant constraints in traditional diagn...Background:Quantum-enhanced medical imaging algorithms–quantum entanglement reconstruction,quantum noise suppression,and quantum beamforming–propose possible remedies for significant constraints in traditional diagnostic imaging,such as resolution,radiation efficiency,and real-time processing.Methods:This work used a mixed-methods strategy,including controlled phantom experiments,retrospective multi-center clinical data analysis,and quantum-classical hybrid processing to assess enhancements in resolution,dosage efficiency,and diagnostic confidence.Statistical validation included analysis of variance(ANOVA)and receiver-operating characteristic curve analysis,juxtaposing quantum-enhanced methodologies with conventional and deep learning approaches.Results:Quantum entanglement reconstruction enhanced magnetic resonance imaging spatial resolution by 33.2%(P<0.01),quantum noise suppression facilitated computed tomography scans with a 60%reduction in radiation,and quantum beamforming improved ultrasound contrast by 27%while preserving real-time processing(<2 ms delay).Inter-reader variability(12%in Diagnostic Confidence Scores)showed that systematic training is needed,even if the performance was better.The research presented(1)a reusable clinical quantum imaging framework,(2)enhanced hardware processes(field-programmable gate array/graphics processing unit acceleration),and(3)cost-benefit analyses demonstrating a 22-month return on investment breakeven point.Conclusion:Quantum-enhanced imaging has a lot of promise for use in medicine,especially in neurology and cancer.Future research should focus on multi-modal integration(e.g.,positron emission tomography–magnetic resonance imaging),cloud-based quantum simulations for enhanced accessibility,and extensive trials to confirm long-term diagnostic accuracy.This breakthrough gives healthcare systems a technology roadmap and a reason to spend money on quantum-enhanced diagnostics.展开更多
As a form of discrete representation learning,Vector Quantized Variational Autoencoders(VQ-VAE)have increasingly been applied to generative and multimodal tasks due to their ease of embedding and representative capaci...As a form of discrete representation learning,Vector Quantized Variational Autoencoders(VQ-VAE)have increasingly been applied to generative and multimodal tasks due to their ease of embedding and representative capacity.However,existing VQ-VAEs often perform quantization in the spatial domain,ignoring global structural information and potentially suffering from codebook collapse and information coupling issues.This paper proposes a frequency quantized variational autoencoder(FQ-VAE)to address these issues.The proposed method transforms image features into linear combinations in the frequency domain using a 2D fast Fourier transform(2D-FFT)and performs adaptive quantization on these frequency components to preserve image’s global relationships.The codebook is dynamically optimized to avoid collapse and information coupling issue by considering the usage frequency and dependency of code vectors.Furthermore,we introduce a post-processing module based on graph convolutional networks to further improve reconstruction quality.Experimental results on four public datasets demonstrate that the proposed method outperforms state-of-the-art approaches in terms of Structural Similarity Index(SSIM),Learned Perceptual Image Patch Similarity(LPIPS),and Reconstruction Fréchet Inception Distance(rFID).In the experiments on the CIFAR-10 dataset,compared to the baselinemethod VQ-VAE,the proposedmethod improves the abovemetrics by 4.9%,36.4%,and 52.8%,respectively.展开更多
Super-resolution structured illumination microscopy(SR-SIM)relies heavily on post-processing reconstruction to obtain high-quality SR images from raw data.Although many SIM reconstruction algorithms have been develope...Super-resolution structured illumination microscopy(SR-SIM)relies heavily on post-processing reconstruction to obtain high-quality SR images from raw data.Although many SIM reconstruction algorithms have been developed to recover fine cellular structures with high fidelity even from the noisy data,whether the pixel intensities of reconstructed SR images are still proportional to the original fluorescence intensity has been less explored.The linearity between the intensity before and after reconstruction is de fined as the intensity fidelity.Here,we proposed a method to evaluate the reconstructed SR image intensity fidelity at different spatial frequencies.With the proposed metric,we systematically investigated the impact of the key factors on the intensity fidelity in the standard Wiener-SIM reconstructions with simulated data,then evaluated the intensity fidelity of the SR images reconstructed by representative open-source packages.Our work provides a reference for SR-SIM image intensity fidelity improvement.展开更多
Deep learning(DL)-based image reconstruction methods have garnered increasing interest in the last few years.Numerous studies demonstrate that DL-based reconstruction methods function admirably in optical tomographic ...Deep learning(DL)-based image reconstruction methods have garnered increasing interest in the last few years.Numerous studies demonstrate that DL-based reconstruction methods function admirably in optical tomographic imaging techniques,such as bioluminescence tomography(BLT).Nevertheless,nearly every existing DL-based method utilizes an explicit neural representation for the reconstruction problem,which either consumes much memory space or requires various complicated computations.In this paper,we present a neural field(NF)-based image reconstruction scheme for BLT that uses an implicit neural representation.The proposed NFbased method establishes a transformation between the coordinate of an arbitrary spatial point and the source value of the point with a relatively light-weight multilayer perceptron,which has remarkable computational efficiency.Another simple neural network composed of two fully connected layers and a 1D convolutional layer is used to generate the neural features.Results of simulations and experiments show that the proposed NF-based method has similar performance to the photon density complement network and the two-stage network,while consuming fewer floating point operations with fewer model parameters.展开更多
A new method to accelerate the convergent rate of the space-alternatinggeneralized expectation-maximization (SAGE) algorithm is proposed. The new rescaled block-iterativeSAGE (RBI-SAGE) algorithm combines the RBI algo...A new method to accelerate the convergent rate of the space-alternatinggeneralized expectation-maximization (SAGE) algorithm is proposed. The new rescaled block-iterativeSAGE (RBI-SAGE) algorithm combines the RBI algorithm with the SAGE algorithm for PET imagereconstruction. In the new approach, the projection data is partitioned into disjoint blocks; eachiteration step involves only one of these blocks. SAGE updates the parameters sequentially in eachblock. In experiments, the RBI-SAGE algorithm and classical SAGE algorithm are compared in theapplication on positron emission tomography (PET) image reconstruction. Simulation results show thatRBI-SAGE has better performance than SAGE in both convergence and image quality.展开更多
基金the National Natural Science Foundation of China(No.61962032)。
文摘In order to improve the reconstruction accuracy of magnetic resonance imaging(MRI),an accurate natural image compressed sensing(CS)reconstruction network is proposed,which combines the advantages of model-based and deep learning-based CS-MRI methods.In theory,enhancing geometric texture details in linear reconstruction is possible.First,the optimization problem is decomposed into two problems:linear approximation and geometric compensation.Aimed at the problem of image linear approximation,the data consistency module is used to deal with it.Since the processing process will lose texture details,a neural network layer that explicitly combines image and frequency feature representation is proposed,which is named butterfly dilated geometric distillation network.The network introduces the idea of butterfly operation,skillfully integrates the features of image domain and frequency domain,and avoids the loss of texture details when extracting features in a single domain.Finally,a channel feature fusion module is designed by combining channel attention mechanism and dilated convolution.The attention of the channel makes the final output feature map focus on the more important part,thus improving the feature representation ability.The dilated convolution enlarges the receptive field,thereby obtaining more dense image feature data.The experimental results show that the peak signal-to-noise ratio of the network is 5.43 dB,5.24 dB and 3.89 dB higher than that of ISTA-Net+,FISTA and DGDN networks on the brain data set with a Cartesian sampling mask CS ratio of 10%.
文摘Microwave diffraction tomography is a process to infer the internal structure of an objectfrom multiple angle views of microwave diffraction shadow. Being sensitive to variations in refractive index of the object, the procedure can be used to measure permittivity distributions within dielectric objects and to image soft tissues for biomedical applications. The optimal resolution distance obtainable is half a wavelength, but this can rarely be achieved because of practical limitations. Some procedures, however, are available to improve the practical resolution. One, which is suitable for microwave tomography, is to use multiple angle views data and to combine the resulting images. The other, which is suitable for improving the image reconstruction resolution, is to use the digital filtering technique and the filtered backpropagation algorithm. A system operating over the X-band microwave frequency is described and some experimental results for objects in air are given.
文摘Background:Medical imaging advancements are constrained by fundamental trade-offs between acquisition speed,radiation dose,and image quality,forcing clinicians to work with noisy,incomplete data.Existing reconstruction methods either compromise on accuracy with iterative algorithms or suffer from limited generalizability with task-specific deep learning approaches.Methods:We present LDM-PIR,a lightweight physics-conditioned diffusion multi-model for medical image reconstruction that addresses key challenges in magnetic resonance imaging(MRI),CT,and low-photon imaging.Unlike traditional iterative methods,which are computationally expensive,or task-specific deep learning approaches lacking generalizability,integrates three innovations.A physics-conditioned diffusion framework that embeds acquisition operators(Fourier/Radon transforms)and noise models directly into the reconstruction process.A multi-model architecture that unifies denoising,inpainting,and super-resolution via shared weight conditioning.A lightweight design(2.1M parameters)enabling rapid inference(0.8s/image on GPU).Through self-supervised fine-tuning with measurement consistency losses adapts to new imaging modalities using fewer annotated samples.Results:Achieves state-of-the-art performance on fastMRI(peak signal-to-noise ratio(PSNR):34.04 for single-coil/31.50 for multi-coil)and Lung Image Database Consortium and Image Database Resource Initiative(28.83 PSNR under Poisson noise).Clinical evaluations demonstrate superior preservation of anatomical structures,with SSIM improvements of 8.8%for single-coil and 4.36%for multi-coil MRI over uDPIR.Conclusion:It offers a flexible,efficient,and scalable solution for medical image reconstruction,addressing the challenges of noise,undersampling,and modality generalization.The model’s lightweight design allows for rapid inference,while its self-supervised fine-tuning capability minimizes reliance on large annotated datasets,making it suitable for real-world clinical applications.
基金supported by the National Natural Science Foundation of China(No.12105018)the Beijing Nova Program(Nos.Z211100002121129 and 20230484413)the Beijing Normal University Start-up Grant(No.312232104).
文摘We present a prototype for hybrid Compton and positron emission tomography(PET)imaging aimed at enhancing data utilization and enabling concurrent imaging of multiple radiopharmaceuticals.The prototype comprises two detectors that utilize LYSO-SiPM and were available in our laboratory.One detector consists of a 50×50 array of LYSO crystals,each measuring 0.9mm×0.9mm×10mm with 1 mm pitches,whereas the other detector comprises a 25×25 array of LYSO crystals,each measuring 1.9mm×1.9mm×10mm with 2 mm pitches.These detectors are mounted on a rotational stage,which enables them to function as either a Compton camera or a PET detector pair.The 64-channel signals from the SiPMs of each detector are processed through a capacitive multiplexing circuit to yield four position-weighted outputs.Distinct energy windows were used to discriminate Compton events from PET events.Energy resolution and energy-channel relationships were calibrated via multiple sources.The measured average energy resolutions(full widths at half maximum,FWHMs)for the detectors at 511 keV were 17.5%and 15.2%,respectively.The initial experimental results indicate an angular resolution(FWHM)of 8.6◦for the system in Compton imaging mode.A V-shaped tube injected with 18 F solution was clearly reconstructed,which further verified the imaging capabilities of the system in Compton imaging mode.The results of simulation and experimental imaging studies show that the system can detect tumors as small as 1 mm in diameter when working in PET imaging mode.Mouse bone PET imaging was successfully conducted,with the results matching well with the corresponding CT images.This technology holds great potential for advancing the development of physiological function modalities.
文摘In this paper, we improve traditional generative adversarial networks (GAN) with reference to residual networks and convolutional neural networks to facilitate the reconstruction of complex objects that cannot be reconstructed by traditional associative imaging methods. Unlike traditional ghost imaging to reconstruct objects from bucket signals, our proposed method can use simple objects (such as EMNIST) as a training set for GAN, and then recognize objects (such as faces) of completely different complexity than the training set. We use traditional ghost imaging and neural network to reconstruct target objects respectively. According to the research results in this paper, the method based on neural network can reconstruct complex objects very well, but the method based on traditional ghost imaging cannot reconstruct complex objects. The research scheme in this paper is of great significance for the reconstruction of complex object-related imaging under low sampling conditions.
基金Sponsored by the Ministerial Level Advanced Research Foundation (A1120060884)
文摘For the nonuniform microscan system where the interframe translation is no longer equivalent to accurate halfpixel, a 2-dimension non-interpolated subpixel algorithm is proposed to consider arhitrary value of microscanning. The aim of the proposed algorithm is to restore double resolution from 2 × 2 undersampled frames. To solve the ill-posed problem in the process of image reconstruction, the prior boundary condition is introduced, and the proposed subpixel reconstruction algorithm is reformulated into the form of line-by-line backward propagation iteration for reducing the calculation complexity. Since the direction of movement offset relative to the accurate halfpixel determines the transfer matrix of the image degradation process, the recon- struction is classified into 4 types when 2 × 2 mieroscan model is applied. All the simulation results and experiment data demonstrate the double resolution improvement compared with the undersampled images. The focus problem, scarcely any possibility of the operation with accurate halfpixel micromotion, is figured out for en-hancing the feasibility of subpixel reconstruction used in practice.
文摘By obtaining a feasible filter function,reconstructed images can be got with linear interpolation and liftered backprojection techniques.Considering the gray and spstial correlation neighbour informations of each pixel,a new supervised classification method is put forward for the reconstructed images,and an experiment with noise image is done,the result shows that the method is feasible and accurate compared with ideal phantoms.
基金This work was supported in part by the National Key Research and Development Program of China,Nos.2017YFA0700401 and 2017YFA0205200the National Natural Science Foundation of China,Nos.62027901,81827808,81527805,and 81671851+2 种基金the CAS Youth Innovation Promotion Association,No.2018167the CAS Key Technology Talent Programand the Project of High-Level Talents Team Introduction in Zhuhai City,No.Zhuhai HLHPTP201703。
文摘Magnetic particle imaging(MPI)is an emerging molecular imaging technique with high sensitivity and temporal-spatial resolution.Image reconstruction is an important research topic in MPI,which converts an induced voltage signal into the image of superparamagnetic iron oxide particles concentration distribution.MPI reconstruction primarily involves system matrix-and x-space-based methods.In this review,we provide a detailed overview of the research status and future research trends of these two methods.In addition,we review the application of deep learning methods in MPI reconstruction and the current open sources of MPI.Finally,research opinions on MPI reconstruction are presented.We hope this review promotes the use of MPI in clinical applications.
基金supported by American Heart Association,No.18AJML34280074.
文摘The state-of-the-art approaches for image reconstruction using under-sampled k-space data are compressed sensing based.They are iterative algorithms that optimize objective functions with spatial and/or temporal constraints.This paper proposes a non-iterative algorithm to estimate the un-measured data and then to reconstruct the image with the efficient filtered backprojection algorithm.The feasibility of the proposed method is demonstrated with a patient magnetic resonance imaging study.The proposed method is also compared with the state-of-the-art iterative compressed-sensing image reconstruction method using the total-variation optimization norm.
文摘Electrical Impedance Tomography (EIT) is a medical imaging technique which can be used to monitor the regional ventilation in patients utilizing voltage measurements made at the thorax. Several reconstruction algorithms have been developed during the last few years. In this manuscript we compare a well-established algorithm and a re-cently developed method for image reconstruction regarding EIT indices derived from the differently reconstructed images.
基金Supported by the National Natural Science Foundation of China(NSFC 61501396)the Colleges and Universities under the Science and Technology Research Projects of Hebei Province(QN2015021)
文摘Based on a strong inter-diagonal matrix and Taylor series expansions,an oversample reconstruction method was proposed to calibrate the optical micro-scanning error. The technique can obtain regular 2 ×2 microscanning undersampling images from the real irregular undersampling images,and can then obtain a high spatial oversample resolution image. Simulations and experiments show that the proposed technique can reduce optical micro-scanning error and improve the system＇s spatial resolution. The algorithm is simple,fast and has low computational complexity. It can also be applied to other electro-optical imaging systems to improve their spatial resolution and has a widespread application prospect.
基金Supported by The National Key R&D Program of China,No.2018YFC0910600the National Natural Science Foundation of China No.81627807 and 11727813+2 种基金Shaanxi Science Funds for Distinguished Young Scholars,No.2020JC-27the Fok Ying Tung Education Foundation,No.161104and Program for the Young Topnotch Talent of Shaanxi Province.
文摘In this editorial,we review recent progress on the applications of artificial intelligence(AI)in image reconstruction for biomedical imaging.Because it abandons prior information of traditional artificial design and adopts a completely data-driven mode to obtain deeper prior information via learning,AI technology plays an increasingly important role in biomedical image reconstruction.The combination of AI technology and the biomedical image reconstruction method has become a hotspot in the field.Favoring AI,the performance of biomedical image reconstruction has been improved in terms of accuracy,resolution,imaging speed,etc.We specifically focus on how to use AI technology to improve the performance of biomedical image reconstruction,and propose possible future directions in this field.
文摘Image super-resolution reconstruction technology is currently widely used in medical imaging,video surveillance,and industrial quality inspection.It not only enhances image quality but also improves details and visual perception,significantly increasing the utility of low-resolution images.In this study,an improved image superresolution reconstruction model based on Generative Adversarial Networks(SRGAN)was proposed.This model introduced a channel and spatial attention mechanism(CSAB)in the generator,allowing it to effectively leverage the information from the input image to enhance feature representations and capture important details.The discriminator was designed with an improved PatchGAN architecture,which more accurately captured local details and texture information of the image.With these enhanced generator and discriminator architectures and an optimized loss function design,this method demonstrated superior performance in image quality assessment metrics.Experimental results showed that this model outperforms traditional methods,presenting more detailed and realistic image details in the visual effects.
文摘Multispectral imaging,which simultaneously captures the spatial and spectral information of a scene,is widely used across diverse fields,including remote sensing,biomedical imaging,and agricultural monitoring.We introduce a snapshot multispectral imaging approach employing a standard monochrome image sensor with no additional spectral filters or customized components.Our system leverages the inherent chromatic aberration of wavelength-dependent defocusing as a natural source of physical encoding of multispectral information;this encoded image information is rapidly decoded via a deep learning-based multispectral Fourier imager network(mFIN).We experimentally tested our method with six illumination bands and demonstrated an overall accuracy of 98.25%for predicting the illumination channels at the input and achieved a robust multispectral image reconstruction on various test objects.This deep learning-powered framework achieves high-quality multispectral image reconstruction using snapshot image acquisition with a monochrome image sensor and could be useful for applications in biomedicine,industrial quality control,and agriculture,among others.
基金supported by National Natural Science Foundation of China(Nos.61261016,61661025)Science and Technology Plan of Gansu Province(No.20JR10RA273).
文摘For image compression sensing reconstruction,most algorithms use the method of reconstructing image blocks one by one and stacking many convolutional layers,which usually have defects of obvious block effects,high computational complexity,and long reconstruction time.An image compressed sensing reconstruction network based on self-attention mechanism(SAMNet)was proposed.For the compressed sampling,self-attention convolution was designed,which was conducive to capturing richer features,so that the compressed sensing measurement value retained more image structure information.For the reconstruction,a self-attention mechanism was introduced in the convolutional neural network.A reconstruction network including residual blocks,bottleneck transformer(BoTNet),and dense blocks was proposed,which strengthened the transfer of image features and reduced the amount of parameters dramatically.Under the Set5 dataset,when the measurement rates are 0.01,0.04,0.10,and 0.25,the average peak signal-to-noise ratio(PSNR)of SAMNet is improved by 1.27,1.23,0.50,and 0.15 dB,respectively,compared to the CSNet+.The running time of reconstructing a 256×256 image is reduced by 0.1473,0.1789,0.2310,and 0.2524 s compared to ReconNet.Experimental results showed that SAMNet improved the quality of reconstructed images and reduced the reconstruction time.
文摘Background:Quantum-enhanced medical imaging algorithms–quantum entanglement reconstruction,quantum noise suppression,and quantum beamforming–propose possible remedies for significant constraints in traditional diagnostic imaging,such as resolution,radiation efficiency,and real-time processing.Methods:This work used a mixed-methods strategy,including controlled phantom experiments,retrospective multi-center clinical data analysis,and quantum-classical hybrid processing to assess enhancements in resolution,dosage efficiency,and diagnostic confidence.Statistical validation included analysis of variance(ANOVA)and receiver-operating characteristic curve analysis,juxtaposing quantum-enhanced methodologies with conventional and deep learning approaches.Results:Quantum entanglement reconstruction enhanced magnetic resonance imaging spatial resolution by 33.2%(P<0.01),quantum noise suppression facilitated computed tomography scans with a 60%reduction in radiation,and quantum beamforming improved ultrasound contrast by 27%while preserving real-time processing(<2 ms delay).Inter-reader variability(12%in Diagnostic Confidence Scores)showed that systematic training is needed,even if the performance was better.The research presented(1)a reusable clinical quantum imaging framework,(2)enhanced hardware processes(field-programmable gate array/graphics processing unit acceleration),and(3)cost-benefit analyses demonstrating a 22-month return on investment breakeven point.Conclusion:Quantum-enhanced imaging has a lot of promise for use in medicine,especially in neurology and cancer.Future research should focus on multi-modal integration(e.g.,positron emission tomography–magnetic resonance imaging),cloud-based quantum simulations for enhanced accessibility,and extensive trials to confirm long-term diagnostic accuracy.This breakthrough gives healthcare systems a technology roadmap and a reason to spend money on quantum-enhanced diagnostics.
基金supported by the Interdisciplinary project of Dalian University DLUXK-2023-ZD-001.
文摘As a form of discrete representation learning,Vector Quantized Variational Autoencoders(VQ-VAE)have increasingly been applied to generative and multimodal tasks due to their ease of embedding and representative capacity.However,existing VQ-VAEs often perform quantization in the spatial domain,ignoring global structural information and potentially suffering from codebook collapse and information coupling issues.This paper proposes a frequency quantized variational autoencoder(FQ-VAE)to address these issues.The proposed method transforms image features into linear combinations in the frequency domain using a 2D fast Fourier transform(2D-FFT)and performs adaptive quantization on these frequency components to preserve image’s global relationships.The codebook is dynamically optimized to avoid collapse and information coupling issue by considering the usage frequency and dependency of code vectors.Furthermore,we introduce a post-processing module based on graph convolutional networks to further improve reconstruction quality.Experimental results on four public datasets demonstrate that the proposed method outperforms state-of-the-art approaches in terms of Structural Similarity Index(SSIM),Learned Perceptual Image Patch Similarity(LPIPS),and Reconstruction Fréchet Inception Distance(rFID).In the experiments on the CIFAR-10 dataset,compared to the baselinemethod VQ-VAE,the proposedmethod improves the abovemetrics by 4.9%,36.4%,and 52.8%,respectively.
基金supported by the National Natural Science Foundation of China[Grant Nos.62205367 and 62141506]Suzhou Basic Research Pilot Project[Grant Nos.SSD2023006 and SJC2021013]Jiangsu Provincial Key Research and Development Program[Grant No.BE2020664].
文摘Super-resolution structured illumination microscopy(SR-SIM)relies heavily on post-processing reconstruction to obtain high-quality SR images from raw data.Although many SIM reconstruction algorithms have been developed to recover fine cellular structures with high fidelity even from the noisy data,whether the pixel intensities of reconstructed SR images are still proportional to the original fluorescence intensity has been less explored.The linearity between the intensity before and after reconstruction is de fined as the intensity fidelity.Here,we proposed a method to evaluate the reconstructed SR image intensity fidelity at different spatial frequencies.With the proposed metric,we systematically investigated the impact of the key factors on the intensity fidelity in the standard Wiener-SIM reconstructions with simulated data,then evaluated the intensity fidelity of the SR images reconstructed by representative open-source packages.Our work provides a reference for SR-SIM image intensity fidelity improvement.
基金supported in part by the National Natural Science Foundation of China(62101278,62001379,62271023)Beijing Natural Science Foundation(7242269).
文摘Deep learning(DL)-based image reconstruction methods have garnered increasing interest in the last few years.Numerous studies demonstrate that DL-based reconstruction methods function admirably in optical tomographic imaging techniques,such as bioluminescence tomography(BLT).Nevertheless,nearly every existing DL-based method utilizes an explicit neural representation for the reconstruction problem,which either consumes much memory space or requires various complicated computations.In this paper,we present a neural field(NF)-based image reconstruction scheme for BLT that uses an implicit neural representation.The proposed NFbased method establishes a transformation between the coordinate of an arbitrary spatial point and the source value of the point with a relatively light-weight multilayer perceptron,which has remarkable computational efficiency.Another simple neural network composed of two fully connected layers and a 1D convolutional layer is used to generate the neural features.Results of simulations and experiments show that the proposed NF-based method has similar performance to the photon density complement network and the two-stage network,while consuming fewer floating point operations with fewer model parameters.
文摘A new method to accelerate the convergent rate of the space-alternatinggeneralized expectation-maximization (SAGE) algorithm is proposed. The new rescaled block-iterativeSAGE (RBI-SAGE) algorithm combines the RBI algorithm with the SAGE algorithm for PET imagereconstruction. In the new approach, the projection data is partitioned into disjoint blocks; eachiteration step involves only one of these blocks. SAGE updates the parameters sequentially in eachblock. In experiments, the RBI-SAGE algorithm and classical SAGE algorithm are compared in theapplication on positron emission tomography (PET) image reconstruction. Simulation results show thatRBI-SAGE has better performance than SAGE in both convergence and image quality.
