Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems,yet still faces many challenges.Building on convolutional neural networks(CNNs),medical image segmentation has achieved tremend...Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems,yet still faces many challenges.Building on convolutional neural networks(CNNs),medical image segmentation has achieved tremendous progress.However,owing to the locality of convolution operations,CNNs have the inherent limitation in learning global context.To address the limitation in building global context relationship from CNNs,we propose LGNet,a semantic segmentation network aiming to learn local and global features for fast and accurate medical image segmentation in this paper.Specifically,we employ a two-branch architecture consisting of convolution layers in one branch to learn local features and transformer layers in the other branch to learn global features.LGNet has two key insights:(1)We bridge two-branch to learn local and global features in an interactive way;(2)we present a novel multi-feature fusion model(MSFFM)to leverage the global contexture information from transformer and the local representational features from convolutions.Our method achieves state-of-the-art trade-off in terms of accuracy and efficiency on several medical image segmentation benchmarks including Synapse,ACDC and MOST.Specifically,LGNet achieves the state-of-the-art performance with Dice's indexes of 80.15%on Synapse,of 91.70%on ACDC,and of 95.56%on MOST.Meanwhile,the inference speed attains at 172 frames per second with 224-224 input resolution.The extensive experiments demonstrate the effectiveness of the proposed LGNet for fast and accurate for medical image segmentation.展开更多
The 9th Chinese–Russian Workshop on Biophotonics and Biomedical Optics was held online on 28–30 September 2020.The bilateral workshop brought together both Russian and Chinese scientists,engineers,and clinical resea...The 9th Chinese–Russian Workshop on Biophotonics and Biomedical Optics was held online on 28–30 September 2020.The bilateral workshop brought together both Russian and Chinese scientists,engineers,and clinical researchers from a variety of disciplines engaged in applying optical science,photonics,and imaging technologies to problems in biology and medicine.During the workshop,2 plenary lectures,35 invited presentations,5 oral presentations,and 8 internet reports were presented.This special issue selects some papers from the attendees and includes both research and review articles.展开更多
The 9th Chinese-Russian Workshop on Biopho-tonics and Biomedical Optics was held online on 28-30 September 2020.The bilateral workshop brought together both Russian and Chinese scientists,engineers,and clinical resear...The 9th Chinese-Russian Workshop on Biopho-tonics and Biomedical Optics was held online on 28-30 September 2020.The bilateral workshop brought together both Russian and Chinese scientists,engineers,and clinical researchers from a variety of disciplines engaged in applying optical science,photonics,and imaging technologies to problems in biology and medicine.During the workshop,2 ple-nary lectures,35 invited presentations,5 oral pre-sentations,and 8 internet reports were presented.This special issue selects some papers from the attendees and includes both research and review articles.展开更多
This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualiz...This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement.However,fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging.To deal with this problem,we propose a 3D multi-scale attention VoxelMorph(MAVoxelMorph)registration network.To alleviate the large deformation problem,a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction,and position-aware axial attention for long-distance dependencies between pixels capture.To further improve the large deformation and fine structure registration results,a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers.Our method was evaluated on four public lung datasets(DIR-Lab dataset,Creatis dataset,Learn2Reg dataset,OASIS dataset)and a local dataset.Results proved that the proposed method achieved better registration performance than current state-of-the-art methods,especially in handling the registration of large deformations and fine structures.It also proved to be fast in 3D image registration,using about 1.5 s,and faster than most methods.Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.展开更多
Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a chall...Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a challenge for accurate segmentation.In this paper,we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue.Using an encoding-decoding structure,we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module(MSAW)after each encoding block.The MSAW module can not only emphasize the fine structures via an upsampling strategy,but also provide pixel-wise weights to measure the importance of the multi-scale features.Additionally,a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions.The proposed MSAW-based semantic segmentation network(MSAW-Net)was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain,demonstrating the efficiency of the proposed method.It achieved an F1 score of 91.8%on Fezf2-2A-CreER dataset,97.1%on LSL-H2B-GFP dataset,82.8%on Thy1-EGFP-Mline dataset,and 86.9%on macaque dataset,achieving improvements over the 3D U-Net model by 3.1%,3.3%,3.9%,and 2.3%,respectively.展开更多
Deep learning networks are increasingly exploited in the field of neuronal soma segmentation.However,annotating dataset is also an expensive and time-consuming task.Unsupervised domain adaptation is an effective metho...Deep learning networks are increasingly exploited in the field of neuronal soma segmentation.However,annotating dataset is also an expensive and time-consuming task.Unsupervised domain adaptation is an effective method to mitigate the problem,which is able to learn an adaptive segmentation model by transferring knowledge from a rich-labeled source domain.In this paper,we propose a multi-level distribution alignment-based unsupervised domain adaptation network(MDA-Net)for segmentation of 3D neuronal soma images.Distribution alignment is performed in both feature space and output space.In the feature space,features from different scales are adaptively fused to enhance the feature extraction capability for small target somata and con-strained to be domain invariant by adversarial adaptation strategy.In the output space,local discrepancy maps that can reveal the spatial structures of somata are constructed on the predicted segmentation results.Then thedistribution alignment is performed on the local discrepancies maps across domains to obtain a superior discrepancy map in the target domain,achieving refined segmentation performance of neuronal somata.Additionally,after a period of distribution align-ment procedure,a portion of target samples with high confident pseudo-labels are selected as training data,which assist in learning a more adaptive segmentation network.We verified the superiority of the proposed algorithm by comparing several domain adaptation networks on two 3D mouse brain neuronal somata datasets and one macaque brain neuronal soma dataset.展开更多
The internal radiation dose calculations based on Chinese models is important in nuclear medicine.Most of the existing models are based on the physical and anatomical data of Caucasian,whose anatomical structure and p...The internal radiation dose calculations based on Chinese models is important in nuclear medicine.Most of the existing models are based on the physical and anatomical data of Caucasian,whose anatomical structure and physiological parameters are quite different from the Chinese,may lead significant effect on internal radiation. Therefore,it is necessary to establish the model based on the Chinese ethnic characteristics,and applied to radiation dosimetry calculation.In this study,a voxel model was established based on the high resolution Visible Chinese Human(VCH).The transport procedure of photon and electron was simulated using the MCNPX Monte Carlo code. Absorbed fraction(AF)and specific absorbed fraction(SAF)were calculated and S-factors and mean absorbed doses for organs with ^(99m)Tc located in liver were also obtained.In comparison with those of VIP-Man and MIRD models, discrepancies were found to be correlated with the racial and anatomical differences in organ mass and inter-organ distance.The internal dosimetry data based on other models that were used to apply to Chinese adult population are replaced with Chinese specific data.The obtained results provide a reference for nuclear medicine,such as dose verification after surgery and potential radiation evaluation for radionuclides in preclinical research,etc.展开更多
Since its inception, endoscopy has aimed to establish an immediate diagnosis that is virtually consistent with a histologic diagnosis. In the past decade, confocal laser scanning microscopy has been brought into endos...Since its inception, endoscopy has aimed to establish an immediate diagnosis that is virtually consistent with a histologic diagnosis. In the past decade, confocal laser scanning microscopy has been brought into endoscopy, thus enabling in vivo microscopic tissue visualization with a magnification and resolution comparable to that obtained with the ex vivo microscopy of histological specimens. The major challenge in the development of instrumentation lies in the miniaturization of a fiber-optic probe for microscopic imaging with micron-scale resolution. Here, we present the design and construction of a confocal endoscope based on a fiber bundle with 1.4-μm lateral resolution and 8-frames per second(fps) imaging speed. The fiber-optic probe has a diameter of 2.6 mm that is compatible with the biopsy channel of a conventional endoscope. The prototype of a confocal endoscope has been used to observe epithelial cells of the gastrointestinal tracts of mice and will be further demonstrated in clinical trials. In addition, the confocal endoscope can be used for translational studies of epithelial function in order to monitor how molecules work and how cells interact in their natural environment.展开更多
For many tller crops,the plant archit ecture(PA),including the plant fresh weight,plant height,number of tllrs,tller angle and stem diameter,sigificantly afects the grain yield.In this study,we propose a method based ...For many tller crops,the plant archit ecture(PA),including the plant fresh weight,plant height,number of tllrs,tller angle and stem diameter,sigificantly afects the grain yield.In this study,we propose a method based on volumetric reconstruction for high-throughput three-dimensional(3D)wheat PA studies.The proposed methodology involves plant volumetric reconst ruction from multiple images,plant model processing and phenotypic parameter estimation and analysis.This study was performed on 80 Triticum aestium plants,and the results were analyzed.Comparing the automated measurements with manual measurements,the mean absolute per-centage error(MAPE)in the plant height and the plant fresh weight was 2.71%(1.08cm with an average plant height of 40.07cm)and 10.06%(1.41g with an average plant fresh weight of 14.06 g),respectively.The root mean square error(RMSE)was 137 cm and 1.79g for the plant height and plant fresh weight,respectively.The correlation cofficients were 0.95 and 0.96 for the plant height and plant fresh weight,respectively.Additionally,the proposed methodology,in-cluding plant reconstruction,model processing and trait ext raction,required only approximately 20s on average per plant using parallel computing on a graphics processing unit(GPU),dem-onstrating that the methodology would be valuable for a high-throughput phenotyping platform.展开更多
Wavefront shaping(WFS)techniques have been used as a powerful tool to control light propagation in complex media,including multimode fibers.In this paper,we propose a new application of WFS for multimode fber-based se...Wavefront shaping(WFS)techniques have been used as a powerful tool to control light propagation in complex media,including multimode fibers.In this paper,we propose a new application of WFS for multimode fber-based sensors.The use of a single multimode fiber alone,without any special fabrication,as a sensor based on the light intensity variations is not an easy task.The twist effect on multimode fiber is used as an example herein.Experimental results show that light intensity through the multimode fiber shows no direct relationship with the twist angle,but the correlation coefficient(CC)of speckle patterns does.Moreover,if WFS is applied to transform the spatially seemingly random light pattern at the exit of the multimode fiber into an optical focus.The focal pattern correlation and intensity both can serve to gauge the twist angle,with doubled measurement range and allowance of using a fast point detector to provide the feedback.With further development,WFS may find potentials to facilitate the development of multimode fber-based sensors in a variety of scenarios.展开更多
Self-occlusions are common in rice canopy images and strongly influence the calculation accuracies of panicle traits. Such interference can be largely eliminated if panicles are phenotyped at the 3 D level.Research on...Self-occlusions are common in rice canopy images and strongly influence the calculation accuracies of panicle traits. Such interference can be largely eliminated if panicles are phenotyped at the 3 D level.Research on 3 D panicle phenotyping has been limited. Given that existing 3 D modeling techniques do not focus on specified parts of a target object, an efficient method for panicle modeling of large numbers of rice plants is lacking. This paper presents an automatic and nondestructive method for 3 D panicle modeling. The proposed method integrates shoot rice reconstruction with shape from silhouette, 2 D panicle segmentation with a deep convolutional neural network, and 3 D panicle segmentation with ray tracing and supervoxel clustering. A multiview imaging system was built to acquire image sequences of rice canopies with an efficiency of approximately 4 min per rice plant. The execution time of panicle modeling per rice plant using 90 images was approximately 26 min. The outputs of the algorithm for a single rice plant are a shoot rice model, surface shoot rice model, panicle model, and surface panicle model, all represented by a list of spatial coordinates. The efficiency and performance were evaluated and compared with the classical structure-from-motion algorithm. The results demonstrated that the proposed method is well qualified to recover the 3 D shapes of rice panicles from multiview images and is readily adaptable to rice plants of diverse accessions and growth stages. The proposed algorithm is superior to the structure-from-motion method in terms of texture preservation and computational efficiency. The sample images and implementation of the algorithm are available online. This automatic, cost-efficient, and nondestructive method of 3 D panicle modeling may be applied to high-throughput 3 D phenotyping of large rice populations.展开更多
Compared with event-related potential(ERP)which is widely used in psychology research,functional near-infrared imaging(fNIRI)is a new technique providing hemodynamic information related to brain activity,except for el...Compared with event-related potential(ERP)which is widely used in psychology research,functional near-infrared imaging(fNIRI)is a new technique providing hemodynamic information related to brain activity,except for electrophysiological signals.Here,we use both these techniques to study ocular attention.We conducted a series of experiments with a classic paradigm of ocular nonselective attention,and monitored responses with fNIRI and ERP respectively.The results showed that fNIRI measured brain activations in the left prefrontal lobe,while ERPs showed activation in frontal lobe.More importantly,only with the combination measurements of fNIRI and ERP,we were then able to find the pinpoint source of ocular nonselective attention,which is in the left and upper corner in Brodmann area 10.These results demonstrated that fNIRI is a reliable technique in psychology,and the combination of fNIRI and ERP can be promising to reveal more information in the research of brain mechanism.展开更多
Vascular segmentation is a crucial task in biomedical image processing,which is significant for analyzing and modeling vascular networks under physiological and pathological states.With advances in fluorescent labelin...Vascular segmentation is a crucial task in biomedical image processing,which is significant for analyzing and modeling vascular networks under physiological and pathological states.With advances in fluorescent labeling and mesoscopic optical techniques,it has become possible to map the whole-mouse-brain vascular networks at capillary resolution.However,segmenting vessels from mesoscopic optical images is a challenging task.The problems,such as vascular signal discontinuities,vessel lumens,and background fluorescence signals in mesoscopic optical images,belong to global semantic information during vascular segmentation.Traditional vascular segmentation methods based on convolutional neural networks(CNNs)have been limited by their insufficient receptive fields,making it challenging to capture global semantic information of vessels and resulting in inaccurate segmentation results.Here,we propose SegVesseler,a vascular segmentation method based on Swin Transformer.SegVesseler adopts 3D Swin Transformer blocks to extract global contextual information in 3D images.This approach is able to maintain the connectivity and topology of blood vessels during segmentation.We evaluated the performance of our method on mouse cerebrovascular datasets generated from three different labeling and imaging modalities.The experimental results demonstrate that the segmentation effect of our method is significantly better than traditional CNNs and achieves state-of-the-art performance.展开更多
The caudal forelimb area(CFA)of the mouse cortex is essential in many forelimb movements,and diverse types of GABAergic interneuron in the CFA are distinct in the mediation of cortical inhibition in motor information ...The caudal forelimb area(CFA)of the mouse cortex is essential in many forelimb movements,and diverse types of GABAergic interneuron in the CFA are distinct in the mediation of cortical inhibition in motor information processing.However,their long-range inputs remain unclear.In the present study,we combined the monosynaptic rabies virus system with Cre driver mouse lines to generate a whole-brain map of the inputs to three major inhibitory interneuron types in the CFA.We discovered that each type was innervated by the same upstream areas,but there were quantitative differences in the inputs from the cortex,thalamus,and pallidum.Comparing the locations of the interneurons in two subregions of the CFA,we discovered that their long-range inputs were remarkably different in distribution and proportion.This whole-brain mapping indicates the existence of parallel pathway organization in the forelimb subnetwork and provides insight into the inhibitory processes in forelimb movement to reveal the structural architecture underlying the functions of the CFA.展开更多
Diabetes mellitus(DM)is a kind of metabolic disorder characterized by chronic hyperglycemia and glucose intolerance due to absolute or relative lack of insulin,leading to chronic damage of vasculature within various o...Diabetes mellitus(DM)is a kind of metabolic disorder characterized by chronic hyperglycemia and glucose intolerance due to absolute or relative lack of insulin,leading to chronic damage of vasculature within various organ systems.These detrimental e®ects on the vascular networks will result in the development of various diseases associated with microvascular injury.Modern optical imaging techniques provide essential tools for accurate evaluation of the structural and functional changes of blood vessels down to capillaries level,which can o®er valuable insight on understanding the development of DM-associated complications and design of targeted therapy.This review will brie°y introduce the DM-induced structural and functional alterations of vasculature within di®erent organs such as skin,cerebrum and kidneys,as well as how novel optical imaging techniques facilitate the studies focusing on exploration of these pathological changes of vasculature caused by DM both in-vivo and ex-vivo.展开更多
Total green leaf area(GLA)is an important trait for agronomic studies.However,existing methods for estimating the GLA of individual rice plants are destructive and labor-intensive.A nondestructive method for estimatin...Total green leaf area(GLA)is an important trait for agronomic studies.However,existing methods for estimating the GLA of individual rice plants are destructive and labor-intensive.A nondestructive method for estimating the total GLA of individual rice plants based on multi-angle color images is presented.Using projected areas of the plant in images,linear,quadratic,exponential and power regression models for estimating total GLA were evaluated.Tests demonstrated that the side-view projected area had a stronger relationship with the actual total leaf area than the top-projected area.And power models fit better than other models.In addition,the use of multiple side-view images was an efficient method for reducing the estimation error.The inclusion of the top-view projected area as a seoond predictor provided only a slight improvement of the total leaf area est imation.When the projected areas from multi angle images were used,the estimated leaf area(ELA)using the power model and the actual leaf area had a high correlation cofficient(R2>0.98),and the mean absolute percentage error(MAPE)was about 6%.The method was capable of estimating the total leaf area in a nondestructive,accurate and eficient manner,and it may be used for monitoring rice plant growth.展开更多
Manipulating and real-time monitoring of neuronal activities with cell-type specificity and precise spatiotemporal resolution during animal behavior are fundamental technologies for exploring the functional connectivi...Manipulating and real-time monitoring of neuronal activities with cell-type specificity and precise spatiotemporal resolution during animal behavior are fundamental technologies for exploring the functional connectivity, information transmission, and physiological functions of neural circuits in vivo. However, current techniques for optogenetic stimulation and neuronal activity recording mostly operate independently. Here, we report an all-fiber-transmission photometry system for simultaneous optogenetic manipulation and multi-color recording of neuronal activities and the neurotransmitter release in a freely moving animal. We have designed and manufactured a wavelength-independent multi-branch fiber bundle to enable simultaneous optogenetic manipulation and multi-color recording at different wavelengths. Further, we combine a laser of narrow linewidth with the lock-in amplification method to suppress the optogenetic stimulation-induced artifacts and channel crosstalk. We show that the collection efficiency of our system outperforms a traditional epi-fluorescence system. Further, we demonstrate successful recording of dynamic dopamine(DA) responses to unexpected rewards in the nucleus accumbens(NAc) in a freely moving mouse. We also show simultaneous dual-color recording of neuronal Ca2+ signals and DA dynamics in the NAc upon delivering an unexpected reward and the simultaneous optogenetic activating at dopaminergic terminals in the same location. Thus, our multi-function fiber photometry system provides a compatible, efficient, and flexible solution for neuroscientists to study neural circuits and neurological diseases.展开更多
2-A minoethyldiphenyl borate(2-APB)is the most commonly used pharmacological agent in the study of calcium release-activated channels(CRACa);however,its inhibitory mechanism to CRACs remains unclear.To address this is...2-A minoethyldiphenyl borate(2-APB)is the most commonly used pharmacological agent in the study of calcium release-activated channels(CRACa);however,its inhibitory mechanism to CRACs remains unclear.To address this issue,we systematically employed confocal imaging,dual-wavelength excitation photometry and FRET to examine the effects of 2-APB on the dynamic activities and function of STIM1 and Orail,two key components of CRACs.Imaging results support that there are two signaling pathways(Orail-independent and Orail-dependent)for the formation of STM1 puncta.2 APB could dose dependently block Orail-independent but not Oril-dependent STIM1 puncta formation,despite its obvious inhibition effect on store-opented Ca^(2+)entry(SOCE).In addition,we found that although 2-APB could not visibly alter near plasma membrane CAD-eYFP localization,it could completely block CAD-YFP-induced constitutive Ca^(2+)entry and promnote the interaction between Orail and CAD by FRET mea-surements.Therefore,we proposed that inhibitory action of 2-APB on SOCE might attribute to its direct inhbitory effects on Orail channel itself,but not the interference on puncta formation between STIM1 and Orail.展开更多
Three-dimensional image reconstruction with Feldkamp,Davis,and Kress(FDK)algorithm is the most time consuming part in Micro-CT.The parallel algorithm based on the computer cluster is capable of accelerating image reco...Three-dimensional image reconstruction with Feldkamp,Davis,and Kress(FDK)algorithm is the most time consuming part in Micro-CT.The parallel algorithm based on the computer cluster is capable of accelerating image reconstruction speed;however,the hardware is very expensive.In this paper,using the most current graphics processing units(GPU),we present a method based on common unified device architecture(CUDA)for speeding up the Micro-CT image reconstruction process.The most time consuming filtering and back-projection parts of the FDK algorithm are parallelized for the CUDA architecture.The CUDA-based reconstruction speed and image qualities are compared with CPU results for the projecting data of the Micro-CT system.The results show that the 3D image reconstruction speed based on CUDA is ten times faster than the speed with CPU.In conclusion the FDK algorithm based on CUDA for Micro-CT can reconstruct the 3D image right after the end of data acquisition.展开更多
Background and aims:The spectral properties of enhanced greenfluorescent protein(EGFP)used in current visualizable animal models for nasopharyngeal carcinoma(NPC)result in a limited imaging depth.Far-redfluorescent pr...Background and aims:The spectral properties of enhanced greenfluorescent protein(EGFP)used in current visualizable animal models for nasopharyngeal carcinoma(NPC)result in a limited imaging depth.Far-redfluorescent proteins have optimal spectral wavelengths that allow deep tissue penetration,thus are well-suited for the imaging of tumor growth and metastases in live animals.This study aims to establish an imageable animal model of NPC using far-redfluorescent proteins.Methods:Eukaryotic expression vectors of far-redfluorescent proteins,mLumin and Katushka S158A,were separately transfected into 5-8F NPC cells,and cell lines stably expressing the far-redfluorescent proteins were obtained.These cells were intraperitoneally or intravenously injected into mice,and their tumorigenic and metastatic potential were examined throughfluorescence imaging.Finally,factors affecting their tumorigenic ability were further assessed through testing side population(SP)cells proportion byflow cytometry.Results:NPC cell line with high tumorigenicity and metastasis(5-8F-mL2)was screened out,which stably expressed far-redfluorescent protein.Intraperitoneal and intravenous injection of 5-8F-mL2 cells resulted in an abdomen metastasis model and a lung metastasis model.In addition,NPC cell line without tumorigenicity(5-8F-Katushka S158A)was screened out.The percentage of SP cells between 5-8F-mL2 and 5-8F-Katushka S158A was found different,suggesting that the SP cell proportion may play a key role in the determination of cell tumorigenic ability.Conclusion:We successfully established animal models for NPC with high tumorigenicity and metastasis using a super-bright far-redfluorescent protein.Owing to the super-brightness and excellent wavelength parameters,these models may be applied as useful tools for intuitive and efficient monitoring of tumor growth and metastasis,as well as assessing the efficacy of nasopharyngeal cancer drugs.展开更多
基金supported by the Open-Fund of WNLO (Grant No.2018WNLOKF027)the Hubei Key Laboratory of Intelligent Robot in Wuhan Institute of Technology (Grant No.HBIRL 202003).
文摘Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems,yet still faces many challenges.Building on convolutional neural networks(CNNs),medical image segmentation has achieved tremendous progress.However,owing to the locality of convolution operations,CNNs have the inherent limitation in learning global context.To address the limitation in building global context relationship from CNNs,we propose LGNet,a semantic segmentation network aiming to learn local and global features for fast and accurate medical image segmentation in this paper.Specifically,we employ a two-branch architecture consisting of convolution layers in one branch to learn local features and transformer layers in the other branch to learn global features.LGNet has two key insights:(1)We bridge two-branch to learn local and global features in an interactive way;(2)we present a novel multi-feature fusion model(MSFFM)to leverage the global contexture information from transformer and the local representational features from convolutions.Our method achieves state-of-the-art trade-off in terms of accuracy and efficiency on several medical image segmentation benchmarks including Synapse,ACDC and MOST.Specifically,LGNet achieves the state-of-the-art performance with Dice's indexes of 80.15%on Synapse,of 91.70%on ACDC,and of 95.56%on MOST.Meanwhile,the inference speed attains at 172 frames per second with 224-224 input resolution.The extensive experiments demonstrate the effectiveness of the proposed LGNet for fast and accurate for medical image segmentation.
文摘The 9th Chinese–Russian Workshop on Biophotonics and Biomedical Optics was held online on 28–30 September 2020.The bilateral workshop brought together both Russian and Chinese scientists,engineers,and clinical researchers from a variety of disciplines engaged in applying optical science,photonics,and imaging technologies to problems in biology and medicine.During the workshop,2 plenary lectures,35 invited presentations,5 oral presentations,and 8 internet reports were presented.This special issue selects some papers from the attendees and includes both research and review articles.
文摘The 9th Chinese-Russian Workshop on Biopho-tonics and Biomedical Optics was held online on 28-30 September 2020.The bilateral workshop brought together both Russian and Chinese scientists,engineers,and clinical researchers from a variety of disciplines engaged in applying optical science,photonics,and imaging technologies to problems in biology and medicine.During the workshop,2 ple-nary lectures,35 invited presentations,5 oral pre-sentations,and 8 internet reports were presented.This special issue selects some papers from the attendees and includes both research and review articles.
基金supported in part by the National Natural Science Foundation of China[62301374]Hubei Provincial Natural Science Foundation of China[2022CFB804]+2 种基金Hubei Provincial Education Research Project[B2022057]the Youths Science Foundation of Wuhan Institute of Technology[K202240]the 15th Graduate Education Innovation Fund of Wuhan Institute of Technology[CX2023295].
文摘This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement.However,fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging.To deal with this problem,we propose a 3D multi-scale attention VoxelMorph(MAVoxelMorph)registration network.To alleviate the large deformation problem,a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction,and position-aware axial attention for long-distance dependencies between pixels capture.To further improve the large deformation and fine structure registration results,a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers.Our method was evaluated on four public lung datasets(DIR-Lab dataset,Creatis dataset,Learn2Reg dataset,OASIS dataset)and a local dataset.Results proved that the proposed method achieved better registration performance than current state-of-the-art methods,especially in handling the registration of large deformations and fine structures.It also proved to be fast in 3D image registration,using about 1.5 s,and faster than most methods.Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.
基金supported by the STI2030-Major-Projects(No.2021ZD0200104)the National Natural Science Foundations of China under Grant 61771437.
文摘Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a challenge for accurate segmentation.In this paper,we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue.Using an encoding-decoding structure,we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module(MSAW)after each encoding block.The MSAW module can not only emphasize the fine structures via an upsampling strategy,but also provide pixel-wise weights to measure the importance of the multi-scale features.Additionally,a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions.The proposed MSAW-based semantic segmentation network(MSAW-Net)was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain,demonstrating the efficiency of the proposed method.It achieved an F1 score of 91.8%on Fezf2-2A-CreER dataset,97.1%on LSL-H2B-GFP dataset,82.8%on Thy1-EGFP-Mline dataset,and 86.9%on macaque dataset,achieving improvements over the 3D U-Net model by 3.1%,3.3%,3.9%,and 2.3%,respectively.
基金supported by the Fund of Key Laboratory of Biomedical Engineering of Hainan Province(No.BME20240001)the STI2030-Major Projects(No.2021ZD0200104)the National Natural Science Foundations of China under Grant 61771437.
文摘Deep learning networks are increasingly exploited in the field of neuronal soma segmentation.However,annotating dataset is also an expensive and time-consuming task.Unsupervised domain adaptation is an effective method to mitigate the problem,which is able to learn an adaptive segmentation model by transferring knowledge from a rich-labeled source domain.In this paper,we propose a multi-level distribution alignment-based unsupervised domain adaptation network(MDA-Net)for segmentation of 3D neuronal soma images.Distribution alignment is performed in both feature space and output space.In the feature space,features from different scales are adaptively fused to enhance the feature extraction capability for small target somata and con-strained to be domain invariant by adversarial adaptation strategy.In the output space,local discrepancy maps that can reveal the spatial structures of somata are constructed on the predicted segmentation results.Then thedistribution alignment is performed on the local discrepancies maps across domains to obtain a superior discrepancy map in the target domain,achieving refined segmentation performance of neuronal somata.Additionally,after a period of distribution align-ment procedure,a portion of target samples with high confident pseudo-labels are selected as training data,which assist in learning a more adaptive segmentation network.We verified the superiority of the proposed algorithm by comparing several domain adaptation networks on two 3D mouse brain neuronal somata datasets and one macaque brain neuronal soma dataset.
基金Supported by National Natural Science Foundation of China(Grant No.10875047 and Grant No.30700214)Program for New Century Excellent Talents in University(Grant No.NCET- 10-0386)
文摘The internal radiation dose calculations based on Chinese models is important in nuclear medicine.Most of the existing models are based on the physical and anatomical data of Caucasian,whose anatomical structure and physiological parameters are quite different from the Chinese,may lead significant effect on internal radiation. Therefore,it is necessary to establish the model based on the Chinese ethnic characteristics,and applied to radiation dosimetry calculation.In this study,a voxel model was established based on the high resolution Visible Chinese Human(VCH).The transport procedure of photon and electron was simulated using the MCNPX Monte Carlo code. Absorbed fraction(AF)and specific absorbed fraction(SAF)were calculated and S-factors and mean absorbed doses for organs with ^(99m)Tc located in liver were also obtained.In comparison with those of VIP-Man and MIRD models, discrepancies were found to be correlated with the racial and anatomical differences in organ mass and inter-organ distance.The internal dosimetry data based on other models that were used to apply to Chinese adult population are replaced with Chinese specific data.The obtained results provide a reference for nuclear medicine,such as dose verification after surgery and potential radiation evaluation for radionuclides in preclinical research,etc.
基金supported by the National Key Technology R&D Program of China (2011BAI12B06)National Natural Science Foundation of China (61205197 and 61178077)
文摘Since its inception, endoscopy has aimed to establish an immediate diagnosis that is virtually consistent with a histologic diagnosis. In the past decade, confocal laser scanning microscopy has been brought into endoscopy, thus enabling in vivo microscopic tissue visualization with a magnification and resolution comparable to that obtained with the ex vivo microscopy of histological specimens. The major challenge in the development of instrumentation lies in the miniaturization of a fiber-optic probe for microscopic imaging with micron-scale resolution. Here, we present the design and construction of a confocal endoscope based on a fiber bundle with 1.4-μm lateral resolution and 8-frames per second(fps) imaging speed. The fiber-optic probe has a diameter of 2.6 mm that is compatible with the biopsy channel of a conventional endoscope. The prototype of a confocal endoscope has been used to observe epithelial cells of the gastrointestinal tracts of mice and will be further demonstrated in clinical trials. In addition, the confocal endoscope can be used for translational studies of epithelial function in order to monitor how molecules work and how cells interact in their natural environment.
基金supported by grants from the National Program on High Technology Development(2013AA102403)the Program for New Century Excellent Talents in University(NCET-10-0386)+1 种基金the National Natural Science Foundation of China(30921091,31200274)the Fundamental Research Funds for the Central Universities(2013PY034).
文摘For many tller crops,the plant archit ecture(PA),including the plant fresh weight,plant height,number of tllrs,tller angle and stem diameter,sigificantly afects the grain yield.In this study,we propose a method based on volumetric reconstruction for high-throughput three-dimensional(3D)wheat PA studies.The proposed methodology involves plant volumetric reconst ruction from multiple images,plant model processing and phenotypic parameter estimation and analysis.This study was performed on 80 Triticum aestium plants,and the results were analyzed.Comparing the automated measurements with manual measurements,the mean absolute per-centage error(MAPE)in the plant height and the plant fresh weight was 2.71%(1.08cm with an average plant height of 40.07cm)and 10.06%(1.41g with an average plant fresh weight of 14.06 g),respectively.The root mean square error(RMSE)was 137 cm and 1.79g for the plant height and plant fresh weight,respectively.The correlation cofficients were 0.95 and 0.96 for the plant height and plant fresh weight,respectively.Additionally,the proposed methodology,in-cluding plant reconstruction,model processing and trait ext raction,required only approximately 20s on average per plant using parallel computing on a graphics processing unit(GPU),dem-onstrating that the methodology would be valuable for a high-throughput phenotyping platform.
基金supported by the Shenzhen Science and Technology Innovation Commission(No.JCYJ20170818104421564)the Hong Kong Innovation and Technology Commission(No.ITS/022/18)+1 种基金the Hong Kong Research Grant Council(No.25204416)the National Natural Science Foundation of China(Nos.81671726 and 81627805).
文摘Wavefront shaping(WFS)techniques have been used as a powerful tool to control light propagation in complex media,including multimode fibers.In this paper,we propose a new application of WFS for multimode fber-based sensors.The use of a single multimode fiber alone,without any special fabrication,as a sensor based on the light intensity variations is not an easy task.The twist effect on multimode fiber is used as an example herein.Experimental results show that light intensity through the multimode fiber shows no direct relationship with the twist angle,but the correlation coefficient(CC)of speckle patterns does.Moreover,if WFS is applied to transform the spatially seemingly random light pattern at the exit of the multimode fiber into an optical focus.The focal pattern correlation and intensity both can serve to gauge the twist angle,with doubled measurement range and allowance of using a fast point detector to provide the feedback.With further development,WFS may find potentials to facilitate the development of multimode fber-based sensors in a variety of scenarios.
基金supported by the National Natural Science Foundation of China (U21A20205)Key Projects of Natural Science Foundation of Hubei Province (2021CFA059)+1 种基金Fundamental Research Funds for the Central Universities (2021ZKPY006)cooperative funding between Huazhong Agricultural University and Shenzhen Institute of Agricultural Genomics (SZYJY2021005,SZYJY2021007)。
文摘Self-occlusions are common in rice canopy images and strongly influence the calculation accuracies of panicle traits. Such interference can be largely eliminated if panicles are phenotyped at the 3 D level.Research on 3 D panicle phenotyping has been limited. Given that existing 3 D modeling techniques do not focus on specified parts of a target object, an efficient method for panicle modeling of large numbers of rice plants is lacking. This paper presents an automatic and nondestructive method for 3 D panicle modeling. The proposed method integrates shoot rice reconstruction with shape from silhouette, 2 D panicle segmentation with a deep convolutional neural network, and 3 D panicle segmentation with ray tracing and supervoxel clustering. A multiview imaging system was built to acquire image sequences of rice canopies with an efficiency of approximately 4 min per rice plant. The execution time of panicle modeling per rice plant using 90 images was approximately 26 min. The outputs of the algorithm for a single rice plant are a shoot rice model, surface shoot rice model, panicle model, and surface panicle model, all represented by a list of spatial coordinates. The efficiency and performance were evaluated and compared with the classical structure-from-motion algorithm. The results demonstrated that the proposed method is well qualified to recover the 3 D shapes of rice panicles from multiview images and is readily adaptable to rice plants of diverse accessions and growth stages. The proposed algorithm is superior to the structure-from-motion method in terms of texture preservation and computational efficiency. The sample images and implementation of the algorithm are available online. This automatic, cost-efficient, and nondestructive method of 3 D panicle modeling may be applied to high-throughput 3 D phenotyping of large rice populations.
基金supported by the National Nature Science Foundation of China(grant No.30070261,60025514).
文摘Compared with event-related potential(ERP)which is widely used in psychology research,functional near-infrared imaging(fNIRI)is a new technique providing hemodynamic information related to brain activity,except for electrophysiological signals.Here,we use both these techniques to study ocular attention.We conducted a series of experiments with a classic paradigm of ocular nonselective attention,and monitored responses with fNIRI and ERP respectively.The results showed that fNIRI measured brain activations in the left prefrontal lobe,while ERPs showed activation in frontal lobe.More importantly,only with the combination measurements of fNIRI and ERP,we were then able to find the pinpoint source of ocular nonselective attention,which is in the left and upper corner in Brodmann area 10.These results demonstrated that fNIRI is a reliable technique in psychology,and the combination of fNIRI and ERP can be promising to reveal more information in the research of brain mechanism.
基金supported by the STI2030-Major Projects (2021ZD0201002)the National Natural Science Foundation of China (82102137,T2122015)+2 种基金Natural Science Foundation of Shaanxi Provincial Department of Education (21JK0796)the Open Project Program of Wuhan National Laboratory for Optoelectronics (2021WNL OKF006)the Natural Science Foundation of Sichuan Province (2022NSFSC0964).
文摘Vascular segmentation is a crucial task in biomedical image processing,which is significant for analyzing and modeling vascular networks under physiological and pathological states.With advances in fluorescent labeling and mesoscopic optical techniques,it has become possible to map the whole-mouse-brain vascular networks at capillary resolution.However,segmenting vessels from mesoscopic optical images is a challenging task.The problems,such as vascular signal discontinuities,vessel lumens,and background fluorescence signals in mesoscopic optical images,belong to global semantic information during vascular segmentation.Traditional vascular segmentation methods based on convolutional neural networks(CNNs)have been limited by their insufficient receptive fields,making it challenging to capture global semantic information of vessels and resulting in inaccurate segmentation results.Here,we propose SegVesseler,a vascular segmentation method based on Swin Transformer.SegVesseler adopts 3D Swin Transformer blocks to extract global contextual information in 3D images.This approach is able to maintain the connectivity and topology of blood vessels during segmentation.We evaluated the performance of our method on mouse cerebrovascular datasets generated from three different labeling and imaging modalities.The experimental results demonstrate that the segmentation effect of our method is significantly better than traditional CNNs and achieves state-of-the-art performance.
基金supported by the National Natural Science Foundation of China(61721092,91749209,and 31871088)the Director Fund of Wuhan National Laboratory for Optoelectronics。
文摘The caudal forelimb area(CFA)of the mouse cortex is essential in many forelimb movements,and diverse types of GABAergic interneuron in the CFA are distinct in the mediation of cortical inhibition in motor information processing.However,their long-range inputs remain unclear.In the present study,we combined the monosynaptic rabies virus system with Cre driver mouse lines to generate a whole-brain map of the inputs to three major inhibitory interneuron types in the CFA.We discovered that each type was innervated by the same upstream areas,but there were quantitative differences in the inputs from the cortex,thalamus,and pallidum.Comparing the locations of the interneurons in two subregions of the CFA,we discovered that their long-range inputs were remarkably different in distribution and proportion.This whole-brain mapping indicates the existence of parallel pathway organization in the forelimb subnetwork and provides insight into the inhibitory processes in forelimb movement to reveal the structural architecture underlying the functions of the CFA.
基金the National Key Research and Development Program of China(Grant No.2017YFA0700501)the National Natu-ral Science Foundation of China(Grant Nos.61860206009,81870934,62105113 and 81961138015)+1 种基金China Postdoctoral Science Foundation Funded Project(Nos.BX20200138,BX20190131,2021M691145 and 2019M662633)the Innovation Fund of WNLO.
文摘Diabetes mellitus(DM)is a kind of metabolic disorder characterized by chronic hyperglycemia and glucose intolerance due to absolute or relative lack of insulin,leading to chronic damage of vasculature within various organ systems.These detrimental e®ects on the vascular networks will result in the development of various diseases associated with microvascular injury.Modern optical imaging techniques provide essential tools for accurate evaluation of the structural and functional changes of blood vessels down to capillaries level,which can o®er valuable insight on understanding the development of DM-associated complications and design of targeted therapy.This review will brie°y introduce the DM-induced structural and functional alterations of vasculature within di®erent organs such as skin,cerebrum and kidneys,as well as how novel optical imaging techniques facilitate the studies focusing on exploration of these pathological changes of vasculature caused by DM both in-vivo and ex-vivo.
基金supported by grants from the National Program on High Technology Development (2013AA102403)the National Program for Basic Research of China (2012CB114305)+2 种基金the National Natural Science Foundation of China (30921091,31200274)the Program for New Century Excellent Talents in University (No.NCET-10-0386)the Fundamental Research Funds for the Central Universities (No.2013PY034).
文摘Total green leaf area(GLA)is an important trait for agronomic studies.However,existing methods for estimating the GLA of individual rice plants are destructive and labor-intensive.A nondestructive method for estimating the total GLA of individual rice plants based on multi-angle color images is presented.Using projected areas of the plant in images,linear,quadratic,exponential and power regression models for estimating total GLA were evaluated.Tests demonstrated that the side-view projected area had a stronger relationship with the actual total leaf area than the top-projected area.And power models fit better than other models.In addition,the use of multiple side-view images was an efficient method for reducing the estimation error.The inclusion of the top-view projected area as a seoond predictor provided only a slight improvement of the total leaf area est imation.When the projected areas from multi angle images were used,the estimated leaf area(ELA)using the power model and the actual leaf area had a high correlation cofficient(R2>0.98),and the mean absolute percentage error(MAPE)was about 6%.The method was capable of estimating the total leaf area in a nondestructive,accurate and eficient manner,and it may be used for monitoring rice plant growth.
基金supported by Beijing Municipal Governmentsupported by the National Natural Science Foundation of China(Grant Nos.61890952)the Director Fund of WNLO。
文摘Manipulating and real-time monitoring of neuronal activities with cell-type specificity and precise spatiotemporal resolution during animal behavior are fundamental technologies for exploring the functional connectivity, information transmission, and physiological functions of neural circuits in vivo. However, current techniques for optogenetic stimulation and neuronal activity recording mostly operate independently. Here, we report an all-fiber-transmission photometry system for simultaneous optogenetic manipulation and multi-color recording of neuronal activities and the neurotransmitter release in a freely moving animal. We have designed and manufactured a wavelength-independent multi-branch fiber bundle to enable simultaneous optogenetic manipulation and multi-color recording at different wavelengths. Further, we combine a laser of narrow linewidth with the lock-in amplification method to suppress the optogenetic stimulation-induced artifacts and channel crosstalk. We show that the collection efficiency of our system outperforms a traditional epi-fluorescence system. Further, we demonstrate successful recording of dynamic dopamine(DA) responses to unexpected rewards in the nucleus accumbens(NAc) in a freely moving mouse. We also show simultaneous dual-color recording of neuronal Ca2+ signals and DA dynamics in the NAc upon delivering an unexpected reward and the simultaneous optogenetic activating at dopaminergic terminals in the same location. Thus, our multi-function fiber photometry system provides a compatible, efficient, and flexible solution for neuroscientists to study neural circuits and neurological diseases.
基金supported by the National Natural Sciences Foundation of China(Grant Nos.31371217 and 30871311).
文摘2-A minoethyldiphenyl borate(2-APB)is the most commonly used pharmacological agent in the study of calcium release-activated channels(CRACa);however,its inhibitory mechanism to CRACs remains unclear.To address this issue,we systematically employed confocal imaging,dual-wavelength excitation photometry and FRET to examine the effects of 2-APB on the dynamic activities and function of STIM1 and Orail,two key components of CRACs.Imaging results support that there are two signaling pathways(Orail-independent and Orail-dependent)for the formation of STM1 puncta.2 APB could dose dependently block Orail-independent but not Oril-dependent STIM1 puncta formation,despite its obvious inhibition effect on store-opented Ca^(2+)entry(SOCE).In addition,we found that although 2-APB could not visibly alter near plasma membrane CAD-eYFP localization,it could completely block CAD-YFP-induced constitutive Ca^(2+)entry and promnote the interaction between Orail and CAD by FRET mea-surements.Therefore,we proposed that inhibitory action of 2-APB on SOCE might attribute to its direct inhbitory effects on Orail channel itself,but not the interference on puncta formation between STIM1 and Orail.
基金the financial support by National Nature Science Foundation of China(Grant No.30070261).
文摘Three-dimensional image reconstruction with Feldkamp,Davis,and Kress(FDK)algorithm is the most time consuming part in Micro-CT.The parallel algorithm based on the computer cluster is capable of accelerating image reconstruction speed;however,the hardware is very expensive.In this paper,using the most current graphics processing units(GPU),we present a method based on common unified device architecture(CUDA)for speeding up the Micro-CT image reconstruction process.The most time consuming filtering and back-projection parts of the FDK algorithm are parallelized for the CUDA architecture.The CUDA-based reconstruction speed and image qualities are compared with CPU results for the projecting data of the Micro-CT system.The results show that the 3D image reconstruction speed based on CUDA is ten times faster than the speed with CPU.In conclusion the FDK algorithm based on CUDA for Micro-CT can reconstruct the 3D image right after the end of data acquisition.
基金The authors thank Prof.Yi-Xin Zeng and Prof.Mu-Sheng Zeng(Sun Yat-sen University Cancer Center,Guangzhou,China)for providing the 5-8F cell line.This work was supported by National Natural Science Foundation of China(Grant No.81172153)National Science and Technology Support Program of China(Grant No.2012BAI23B02).
文摘Background and aims:The spectral properties of enhanced greenfluorescent protein(EGFP)used in current visualizable animal models for nasopharyngeal carcinoma(NPC)result in a limited imaging depth.Far-redfluorescent proteins have optimal spectral wavelengths that allow deep tissue penetration,thus are well-suited for the imaging of tumor growth and metastases in live animals.This study aims to establish an imageable animal model of NPC using far-redfluorescent proteins.Methods:Eukaryotic expression vectors of far-redfluorescent proteins,mLumin and Katushka S158A,were separately transfected into 5-8F NPC cells,and cell lines stably expressing the far-redfluorescent proteins were obtained.These cells were intraperitoneally or intravenously injected into mice,and their tumorigenic and metastatic potential were examined throughfluorescence imaging.Finally,factors affecting their tumorigenic ability were further assessed through testing side population(SP)cells proportion byflow cytometry.Results:NPC cell line with high tumorigenicity and metastasis(5-8F-mL2)was screened out,which stably expressed far-redfluorescent protein.Intraperitoneal and intravenous injection of 5-8F-mL2 cells resulted in an abdomen metastasis model and a lung metastasis model.In addition,NPC cell line without tumorigenicity(5-8F-Katushka S158A)was screened out.The percentage of SP cells between 5-8F-mL2 and 5-8F-Katushka S158A was found different,suggesting that the SP cell proportion may play a key role in the determination of cell tumorigenic ability.Conclusion:We successfully established animal models for NPC with high tumorigenicity and metastasis using a super-bright far-redfluorescent protein.Owing to the super-brightness and excellent wavelength parameters,these models may be applied as useful tools for intuitive and efficient monitoring of tumor growth and metastasis,as well as assessing the efficacy of nasopharyngeal cancer drugs.
