2D Projection Research Articles

Reconstructing a 3D Medical Image from a Few 2D Projections Using a B-Spline-Based Deformable Transformation

(1) Background: There was a need for 3D image reconstruction from a series of 2D projections in medical applications. However, additional exposure to X-ray projections may harm human health. To alleviate it, minimizing the projection numbers is a solution to reduce X-ray exposure, but this would cause significant image noise and artifacts. (2) Purpose: In this study, a method was proposed for the reconstruction of a 3D image from a minimal set of 2D X-ray projections using a B-spline-based deformable transformation. (3) Methods: The inputs of this method were a 3D image which was acquired in previous treatment and used as a prior image and a minimal set of 2D projections which were acquired during the current treatment. The goal was to reconstruct a new 3D image in current treatment from the two inputs. The new 3D image was deformed from the prior image via the displacement matrixes that were interpolated by the B-spline coefficients. The B-spline coefficients were solved with the objective function, which was defined as the mean square error between the reconstructed and the ground-truth projections. In the optimization process the gradient of the objective function was calculated, and the B-spline coefficients were then updated. For the acceleration purpose, the computation of the 2D and 3D image reconstructions and B-spline interpolation were implemented on a graphics processing unit (GPU). (4) Results: When the scan angles were more than 60°, the image quality was significantly improved, and the reconstructed image was comparable to that of the ground-truth image. As the scan angles were less than 30°, the image quality was significantly degraded. The influence of the scan orientation on the image quality was minor. With the application of GPU acceleration, the reconstruction efficiency was improved by hundred times compared to that of the conventional CPU. (5) Conclusions: The proposed method was able to generate a high-quality 3D image using a few 2D projections, which amount to ~ 20% of the total projections required for a standard image. The introduction of the B-spline-interpolated displacement matrix was effective in the suppressing noise in the reconstructed image. This method could significantly reduce the imaging time and the radiation exposure of patients under treatment.

Symmetries and Dynamics of Generalized Biquaternionic Julia Sets Defined by Various Polynomials

Higher-dimensional hypercomplex fractal sets are getting more and more attention because of the discovery of more and more interesting properties and visual aesthetics. In this study, the attention was focused on generalized biquaternionic Julia sets and a generalization of classical Julia sets, defined by power and monic higher-order polynomials. Despite complex and quaternionic Julia sets, their biquaternionic analogues are still not well investigated. The performed morphological analysis of 3D projections of these sets allowed for definition of symmetries, limit shapes, and similarities with other fractal sets of this class. Visual observations were confirmed by stability analysis for initial cycles, which confirm similarities with the complex, bicomplex, and quaternionic Julia sets, as well as manifested differences between the considered formulations of representing polynomials.

Appropriate Technology-Based Project-Based Learning: 3D Printing Utilization for Learning Media Class

Indonesia has more than ten thousand islands, making it an archipelago; therefore, the distribution of learning media and facilities over the country is difficult. This condition may affect the learning material delivery and the student's understanding. Thus, the do-it-yourself (DIY) learning media helps to overcome this condition. We investigated the usefulness of applying appropriate technology through the 3D printing technique to fabricate learning media inspired by local wisdom to meet the needs of learning media. The students were guided by a project-based learning (PjBL) method to find the problem, design their learning media, fabricate it using a 3D printer, and expected to be applied to deliver the teaching material in the class. This study aimed to observe the implementation of the designing process based on theory to reality. A qualitative method and questionnaire were used to measure student's responses. We found that PjBL learning based on a 3D printing project increases students' motivation through creativity and effectiveness in delivering the teaching materials. Furthermore, 3D printed-based learning media can easily be fabricated, producing less waste.

Lyapunov Exponents and Kaplan-yorke Dimension for Fivedimensional System

This work introduces a new high-dimensional five-dimensional system with chaotic and periodic solutions. For special values of parameters, we calculate the Kaplan- Yorke dimension and we show the dynamics of Lyapunov exponents. Some definitions and propositions are given. The main intent is to use the 2D and 3D projections of the 5D trajectories on different subspaces, to construct the graphs of solutions for understanding and managing the system. Visualizations where possible, are provided.

Structural, elastic, optical and thermodynamic properties of CdGeP2 from theoretical prediction

We have studied the structural, elastic anisotropy, electronic properties and optical properties of CdGeP2 crystal by using first-principles method. The effect of pressure on the elastic constants of CdGeP2 was explored. The changes in the anisotropy of the CdGeP2 at different pressures were analyzed in detail by three-dimensional (3D) and 2D projections of bulk modulus and Young's modulus. The electronic properties, including the band structure, density of states, were investigated, and it turned out that CdGeP2 is a semiconductor. Furthermore, we analyze the dielectric function, refractive index, extinction coefficient, absorption coefficient, and electrical conductivity of CdGeP2 in detail. Finally, the influence laws of temperature and pressure on Debye temperature, heat capacity and Grüneisen parameter are obtained, respectively

Three-dimensional rotational angiography improves mechanical thrombectomy recanalization rate for acute ischaemic stroke due to middle cerebral artery M2 segment occlusions.

Occlusions of the middle cerebral artery (MCA) M2 segments can be difficult to address with mechanical thrombectomy (MTB) using standard projections and this can affect the final recanalization. Three-dimensional rotational angiography (3D-RA) allows to obtain a 3D model of cerebral vessels in a few seconds and to determine the best two-dimensional (2D) projections to be selected to evaluate and treat cerebrovascular diseases, such as aneurysms or vascular malformations. We aimed to determine if 3D-RA could be applied also in MTB. A retrospective review of two patient cohorts treated during two time periods of 12 months before and after the introduction of 3D-RA use at our institution for MTB in M2 occlusions. Analyses were conducted to compare the two groups for procedural characteristics, such as timing, recanalization rate and complications and clinical outcome. One hundred acute ischaemic stroke (AIS) patients (3D-RA group = 57; controls = 43) underwent MTB for an M2 occlusion during the two study periods. Recanalization rates were significantly higher in cases treated with 3D-RA. The mean 3D technique thrombectomy time was compared to that of non-3D cases (47 vs. 49 min, respectively). Our findings showed that 3D-RA is a useful tool to select specific working projections to AIS patients presenting an M2 occlusion by improving final recanalization compared to standard projections, without increasing the overall procedural time.

An AI-Based Colonic Polyp Classifier for Colorectal Cancer Screening Using Low-Dose Abdominal CT.

Among the non-invasive Colorectal cancer (CRC) screening approaches, Computed Tomography Colonography (CTC) and Virtual Colonoscopy (VC), are much more accurate. This work proposes an AI-based polyp detection framework for virtual colonoscopy (VC). Two main steps are addressed in this work: automatic segmentation to isolate the colon region from its background, and automatic polyp detection. Moreover, we evaluate the performance of the proposed framework on low-dose Computed Tomography (CT) scans. We build on our visualization approach, Fly-In (FI), which provides "filet"-like projections of the internal surface of the colon. The performance of the Fly-In approach confirms its ability with helping gastroenterologists, and it holds a great promise for combating CRC. In this work, these 2D projections of FI are fused with the 3D colon representation to generate new synthetic images. The synthetic images are used to train a RetinaNet model to detect polyps. The trained model has a 94% f1-score and 97% sensitivity. Furthermore, we study the effect of dose variation in CT scans on the performance of the the FI approach in polyp visualization. A simulation platform is developed for CTC visualization using FI, for regular CTC and low-dose CTC. This is accomplished using a novel AI restoration algorithm that enhances the Low-Dose CT images so that a 3D colon can be successfully reconstructed and visualized using the FI approach. Three senior board-certified radiologists evaluated the framework for the peak voltages of 30 KV, and the average relative sensitivities of the platform were 92%, whereas the 60 KV peak voltage produced average relative sensitivities of 99.5%.

From 2D projections to the 3D rotation matrix: an attempt for finding a machine learning approach for the efficient evaluation of mechanical joining elements in X-ray computed tomography volume data

Destructive and predominantly manual procedures are commonly used in the automotive industry for the testing of mechanical joints, such as rivets or screws. Combining X-ray computed tomography (CT) and machine learning (ML) bears the potential of a non-destructive and largely automated methodology. Assuming the desired result is a comprehensible and documentable evaluation, three basic steps need to be automatized: First, a joint must be detected and identified as such in a CT scan of the joined parts. Second, the detected region containing the joint is rotated to a predefined orientation. Third, key measures in cross-sections from the newly oriented joint are dimensioned and documented. This work deals only with the second step, the rotation. On the one hand, we present a methodology for creating a well-curated data set for the contextual machine learning application. On the other, we evaluate its performance on the well-known ResNet50. More concretely, we investigate if it is possible for a deep convolutional neural network (CNN) to learn the respective rotation matrix from three volume projections that are perpendicular to each other. Two scenarios are investigated: In one scenario we assume that future data that is presented to the network has similar rivet demographics to historic data. We therefore do not employ hold-out sets for the network evaluation. In the other scenario we assume the opposite and therefore evaluating the networks performance with hold-out sets. We show that from a machine learning point of view, a CNN like ResNet50 is well able to learn this relationship with acceptable accuracy. In most cases the validation loss dropped below 0.1 after only a couple of epochs. In one particular case, we even reached both mean and median errors lower than 0.2 for approximately 80% of the entire test set of 1600 examples using our methodology. From an application point of view, however, these low test set errors should be treated with caution since small deviations from the intended rotation matrix can cause volume warping and translation. In another case, in which we used a hold-out set, only a fraction of the median errors were below 0.2.

Exploring 2D projection and 3D spatial information for aircraft 6D pose

The 6D pose estimation is important for the safe take-off and landing of the aircraft using a single RGB image. Due to the large scene and large depth, the exiting pose estimation methods have unstratified performance on the accuracy. To achieve precise 6D pose estimation of the aircraft, an end-to-end method using an RGB image is proposed. In the proposed method, the 2D and 3D information of the keypoints of the aircraft is used as the intermediate supervision, and 6D pose information of the aircraft in this intermediate information will be explored. Specifically, an off-the-shelf object detector is utilized to detect the Region of the Interest (RoI) of the aircraft to eliminate background distractions. The 2D projection and 3D spatial information of the pre-designed keypoints of the aircraft is predicted by the keypoint coordinate estimator (KpNet). The proposed method is trained in an end-to-end fashion. In addition, to deal with the lack of the related datasets, this paper builds the Aircraft 6D Pose dataset to train and test, which captures the take-off and landing process of three types of aircraft from 11 views. Compared with the latest Wide-Depth-Range method on this dataset, our proposed method improves the average 3D distance of model points metric (ADD) and 5° and 5 m metric by 86.8% and 30.1%, respectively. Furthermore, the proposed method gets 9.30 ms, 61.0% faster than YOLO6D with 23.86 ms.

Accuracy of five different 3D printing workflows for dental models comparing industrial and dental desktop printers

ObjectivesThe aim of this study was to evaluate the accuracy, in terms of trueness and precision, of printed models using five different industrial and dental desktop 3D printers.Materials and methodsFull-arch digital models with scanbodies of 15 patients were printed with five different 3D printers. The industrial printers were 3D system Project MJP2500 (3DS) and Objet30 OrthoDesk (Obj). The dental desktop printers were NextDent 5100 (ND), Formlabs Form 2 (FL) and Rapidshape D30 (RS). A total of 225 printed models were analysed. The printed models were digitized and compared with the reference cast model using the Control X software (Geomagic). The descriptive statistics and one-way ANOVA with the post hoc Tukey test were performed (α = 0.05).ResultsThe one-way ANOVA for the trueness and precision of the printed model presented the best results for the 3DS, followed by ND, Obj, FL and RS (P < 0.01). In the scanbody zone, the best results were for the 3DS group, followed by Obj, ND, FL and RS (P < 0.01). Comparing the technologies, the Multijet technology used in industrial printers presented better results than the DLP and SLA technologies used in dental desktop printers (P > 0.01).ConclusionsThere were statistically significant differences in terms of the accuracy of the printed models, with better results for the industrial than the dental desktop 3D printers.Clinical relevanceThe industrial 3D printers used in dental laboratories presented better accuracy than the in-office dental desktop 3D printers, and this should be considered when the best accuracy is needed to perform final prosthetic restorations.

Image-to-Graph Convolutional Network for 2D/3D Deformable Model Registration of Low-Contrast Organs.

Organ shape reconstruction based on a single-projection image during treatment has wide clinical scope, e.g., in image-guided radiotherapy and surgical guidance. We propose an image-to-graph convolutional network that achieves deformable registration of a three-dimensional (3D) organ mesh for a low-contrast two-dimensional (2D) projection image. This framework enables simultaneous training of two types of transformation: from the 2D projection image to a displacement map, and from the sampled per-vertex feature to a 3D displacement that satisfies the geometrical constraint of the mesh structure. Assuming application to radiation therapy, the 2D/3D deformable registration performance is verified for multiple abdominal organs that have not been targeted to date, i.e., the liver, stomach, duodenum, and kidney, and for pancreatic cancer. The experimental results show shape prediction considering relationships among multiple organs can be used to predict respiratory motion and deformation from digitally reconstructed radiographs with clinically acceptable accuracy.

A Blockchain and modified Best-worst algorithm enhanced Blackboard model for stranger collaboration based social product design

Social product design (SPD) is a new kind of product design pattern emerged under the background of sharing economic trends and fast developing online communication technologies. Due to the participation of large numbers of socialized designers (SDs), SPD pattern has the capability of utilizing socialized design resources to increase the design power of design companies with relatively lower costs. However, the SDs have the characteristics of self-driven, self-organized, unaffiliated, distributed and discontinuous participation, and stranger collaboration. These characteristics lead to the problems of low collaboration efficiency, low controllability, and low mutual trust during SPD process. Consequently these problems could limit the industrial application of SPD in relatively complicated and intellectual property issues involved design projects. Addressing these problems, a Blockchain and modified Best-worst algorithm enhanced Blackboard model is established. The Blackboard model with customized components can improve the collaboration efficiency and orderliness among SDs. The Blockchain can realize reliable recording of the discontinuous participation of SDs in SPD process. On this basis, a modified Best-Worst algorithm is established to support the estimation of the contribution degrees of SDs during SPD process. Combined together, the model can improve the operability of SPD application. A 3D printer concept design project was used as case study, and the result shows that the model can improve the operability of SPD pattern in relatively complicated and intellectual property issues involved design projects compared with ordinary crowdsourcing or opensourcing design models, but its higher operating cost may limit its application only for high valued design projects.

Research on 3D Technology in The Field of Education

With the continuous development of science and technology, the continuous innovation of network and digital electronic culture has brought a huge impact on the traditional book culture, which makes the educational means and classroom mode in the field of education constantly change, and the future classroom will be more scientific and technological. Among them, 3D holographic projection is an embodiment. With the improvement of teaching requirements, 3D holographic projection technology will one day replace today's ordinary projector into the classroom and into the classroom, giving students strong visual shock and wonderful classroom effects.

Toward real-time automatic treatment planning (RTTP) with a one-step 3D fluence map prediction method and (nonorthogonal) convolution technique

PurposeTo establish and evaluate a (quasi) real-time automated treatment planning (RTTP) strategy utilizing a one-step full 3D fluence map prediction model based on a nonorthogonal convolution operation for rectal cancer radiotherapy. MethodsThe RTTP approach directly extracts 3D projections from volumetric CT and anatomical data according to the beam incident direction. A 3D deep learning model with a nonorthogonal convolution operation was established that takes projections in cone beam space as input, extracts the features along and around the ray-trace path, and outputs a predicted fluence map (PFM) for each beam. The PFM is then converted to the MLC sequence with deliverable MUs to generate the final treatment plan. A total of 314 rectal adenocarcinoma patients with 2198 projection data samples were used in model training and validation. An extra 20 patients were used to test the feasibility of the RTTP method by comparing the plan quality, efficiency, deliverability performance, and physician blinded review results with the manual plans. ResultsOverall, the RTTP plans met the clinical dose criteria for target coverage, conformity, homogeneity, and organ-at-risk dose sparing. Compared to manual plans, the RTTP plans showed increases in PTV D1% by only 2.33% (p < 0.001) and a decrease in PTV D99% by 0.45% (p < 0.05). The RTTP plans showed a dose increase in the bladder, with a V50 of 14.01 ± 11.75% vs. 10.74 ± 8.51%, respectively, and no significant increases in the femoral head with the mean dose. The planning efficiency was improved in RTTP planning, with 39 s vs. 944 s in fluence map generation; the deliverability performance was saved by 1.91% (p < 0.001) in total MU. According to the blinded plan review by our physician, 55% of RTTP plans can be directly used in clinical radiotherapy treatment. ConclusionThe quasi RTTP method improves the planning efficiency and deliverability performance while maintaining a plan quality close to that of the optimized manual plans in rectal radiotherapy.

Genetic Alterations in Members of the Proteasome 26S Subunit, AAA-Atpase (PSMC) Gene Family Transmit Proteasome Inhibitor Resistance in Multiple Myeloma By Impacting the ADP/ATP Binding Pocket

Background For the treatment of Multiple Myeloma (MM), proteasome inhibitors (PIs) are highly efficient and widely used, but resistance is a major obstacle to successful therapy. Several underlying mechanisms have been proposed but were only reported for a minority of resistant patients e.g., mutations in the PI binding domain PSMB5. However, the proteasome is a large and complex machinery and its mechanistic understanding remains incomplete. In this study, we unravel the role of the six ATPases of the 19S proteasome regulator (PSMC1-6) in MM resistance. Methods As an example of cancer evolution and the acquisition of PI resistance, we conducted an in-depth analysis of an index patient applying deep sequencing of immunoglobulin rearrangements (IG), FISH, and WES in serial samples at MGUS, newly diagnosed MM, and five later timepoints including two relapses. Then we expand towards generic meta-analyses of PSMC alterations in our own and other published datasets comprising 1,499 newly diagnosed (NDMM) and 447 PI pretreated MM (PMM) patients for WGS/WES. For RNA sequencing, we analyzed 16 normal plasma cells (NPCs) and 360 NDMM patients. Furthermore, we explore the structural impact of PSMC single nucleotide variants on the proteasome and their relevance for PI resistance using 3D modeling with Chimera and Pymol. Results The WES analysis of our index patient uncovers a PSMC2 Y429S mutation acquired at the second relapse after intensive bortezomib-containing therapy. We functionally confirm this variant to mediate PI resistance in the cell line RPMI 8226. In the meta-analysis, a total of 36 mutations were identified (mutation incidence: 1.85%). The PSMC5 G152S mutation was found in three independent patients, and a G152R in a fourth patient, thus, representing a potential hotspot. Based on 3D structural projections, the identified mutations can be clustered into three main groups: first, mutations affecting the intrinsic conformation of the respective subunit; second, mutations perturbing the PSMC protomer interactions and the proper folding/functioning of the whole complex; and third, mutations impairing the ADP/ATP binding pocket (Figure). Nine mutations belong to the third category interacting with the adenine ring or the phosphate groups of the ATP. The expression analysis reveals a significant upregulation of PSMC3 (P=0.003), PSMC4 (P≤0.001), and PSMC5 (P=0.043) from NPC to NDMM. Conclusions The 19S hexameric AAA ATPase ring of the 26S proteasome is responsible for unfolding and introducing the substrates into the catalytic 20S complex. PSMC gene alterations are structurally often located in regulatory sites for activity e.g. impacting the ADP/ATP binding pocket. The following activation of the 26S proteasome increases the PI tolerance and represents a new PI resistance mechanism. Moreover, changes in the expression level of the PSMC members seem to play a role not only in the development of PI resistance but already in the development of MM, probably by increasing the protein turnover capacity of myeloma cells. This study opens the way to explore new classes of proteasome inhibitors avoiding this kind of mechanism of resistance. Figure:Mutations in the'PSMC' subunits of the 19S Base AAA-ATPase hexamer accumulate in proximity to the ATP/ADP binding pockets. The backbone of the PSMC hexamer is displayed as a ribbon diagram, mutations (red) and ADP (blue) as 3D spheres (PDB 5GJR). The mutations cluster in areas responsible for proper folding and functioning of the complex or in proximity to the ADP/ATP binding pockets, exemplarily shown for PSMC1, PSMC2, and PSMC5. ADP is presented as a wireframe stick in light green (PDB 5GJQ). The PSMC2Y429S mutation of the index patient is indicated in yellow. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Between Material Permanence and Digital Evanescence. 3D Projection Mappings during the Festivals of Cluj

Between Material Permanence and Digital Evanescence. 3D Projection Mappings during the Festivals of Cluj

Agile Leader’s Emotional Resilience and Their Digital Innovations and Business Transformations in a Workplace in Msme Sector (New Normal) to Mitigate COVID-19 &amp; Its Successors

Theoretical Framework:During Covid-19, IndCzech invented numerous digital business transformations to maintain its sustainability in the global economy. IndCzech’s digital innovations (New normal) through online safety study workshops, business meetings, Factory Acceptance test, 3D Model review, Design review, Quality audit, project implementation and business transformation through work sharing possibilities in different places, integrated work culture, changes in native infrastructure have been analyzed the possibilities to implement. Based on the study, conceptual model proposes with Digital innovation, Business Transformation, Emotional Resilience, Challenges in workplace and Organizational growth. Design / Methodology/approach:129 agile leaders in the workplace in MSME sector data’s were collected through questionnaire and data analyzed with descriptive analysis, t-test, one way ANOVA, Correlation analysis, Regression analysis through SPSS and SEM model through SPSS AMOS. Findings:We found that Digital innovation and business transformation with emotional resilience are positively linked with organizational growth. Research, Practical &amp; Social Implications:The purpose and implication of this empirical study to highlight on one’s agile leader’s emotional resilience in workplace (IndCzech) and their performance through digital innovations and business transformations and recommend government and MSME to improve organizational growth. Originality / Value:Agile leaders in Engineering service industry and their emotional resilience during pandemic period were not evaluated and it added value in the engineering service sector to mitigate their issues to have better business sustainability and organizational growth.

Predictions of structural stability, elastic anisotropy, and thermodynamic properties of TM5Si3C (TM = Cr, Mo, and W)

In this work, we have systematically studied the structural stability, elastic anisotropy, and thermodynamic properties of TM5Si3C (TM = Cr, Mo, and W) using first-principles calculations. The optimized lattice constants and elastic constants of Mo5Si3C are in an excellent agreement with the available experimental data. The elastic anisotropy of TM5Si3C were illustrated by elastic anisotropy index, three-dimension (3D) surface structure and two-dimension (2D) projection, and the results show that the order of elastic anisotropy is W5Si3C > Mo5Si3C > Cr5Si3C. In addition, the hardness, Debye temperature, sound velocity and Grüneisen constant are determined by the elastic constants. The thermal conductivity of TM5Si3C (TM = Cr, Mo and W) was calculated in detail using the Clark model. Finally, the thermodynamic properties were discussed.

Assessment of imaging models for volumetric tomography of fluid flows

Volumetric tomography is a powerful tool that employs 2D projections to reconstruct unknown physical quantities in 3D fluid flows. A mathematical model of the imaging process is required to recover the desired volumetric fields. Errors in the imaging model can corrupt the reconstructions, so an accurate model is critical. This work reports the first systematic assessment of nine imaging models in terms of accuracy, computational cost, and range of applicability. A sample matrix method is developed to assess and improve the model’s accuracy. A flame chemiluminescence tomography experiment and a synthetic tomographic particle image velocimetry test were conducted. For both luminosity field reconstruction and velocity estimation, the Voxel Spread Function (VSF) model is the most accurate, but the computational cost is tens or hundreds of times higher than other models. The worst models are the Ray-length and VC Direct models. Their error increases when the spatial frequency or seeding density of the object field increases.

PGNet: Projection generative network for sparse-view reconstruction of projection-based magnetic particle imaging.

Magnetic particle imaging (MPI) is a novel tomographic imaging modality that scans the distribution of superparamagnetic iron oxide nanoparticles. However, it is time-consuming to scan multiview two-dimensional (2D) projections for three-dimensional (3D) reconstruction in projection MPI, such as computed tomography (CT). An intuitive idea is to use the sparse-view projections for reconstruction to improve the temporal resolution. Tremendous progress has been made toward addressing the sparse-view problem in CT, because of the availability of large data sets. For the novel tomography of MPI, to the best of our knowledge, studies on the sparse-view problem have not yet been reported. The acquisition of multiview projections for 3D MPI imaging is time-consuming. Our goal is to only acquire sparse-view projections for reconstruction to improve the 3D imaging temporal resolution of projection MPI. We propose to address the sparse-view problem in projection MPI by generating novel projections. The data set we constructed consists of three parts: simulation data set (including 3000 3D data), four phantoms data, and an in vivo mouse data. The simulation data set is used to train and validate the network, and the phantoms and in vivo mouse data are used to test the network. When the number of novel generated projections meets the requirements of filtered back projection, the streaking artifacts will be absent from MPI tomographic imaging. Specifically, we propose a projection generativenetwork (PGNet), that combines an attention mechanism, adversarial training strategy, and a fusion loss function and can generate novel projections based on sparse-view real projections. To the best of our knowledge, we are the first to propose a deep learning method to attempt to overcome the sparse-view problem in projection MPI. We compare our method with several sparse-view methods on phantoms and in vivo mouse data and validate the advantages and effectiveness of our proposed PGNet. Our proposed PGNet enables the 3D imaging temporal resolution of projection MPI to be improved by 6.6 times, while significantly suppressing the streaking artifacts. We proposed a deep learning method operated in projection domain to address the sparse-view reconstruction of MPI, and the data scarcity problem in projection MPI reconstruction is alleviated by constructing a sparse-dense simulated projection data set. By our proposed method, the number of acquisitions of real projections can be reduced. The advantage of our method is that it prevents the generation of streaking artifacts at the source. Our proposed sparse-view reconstruction method has great potential for application to time-sensitive in vivo 3D MPI imaging.