Search

Objective: Proton therapy offers an advantageous dose distribution compared to the photon therapy, since it deposits most of the energy at the end of range, namely the Bragg peak (BP). Protoacoustic technique was developed to in vivo determine the BP locations. However, it requires large dose delivery to the tissue to obtain an averaged acoustic signal with a sufficient signal to noise ratio (SNR), which is not suitable in clinics. We propose a deep learning-based technique to acquire denoised acoustic signals and reduce BP range uncertainty with much lower doses. Approach: Three accelerometers were placed on the distal surface of a cylindrical polyethylene (PE) phantom to collect protoacoustic signals. In total 512 raw signals were collected at each device. Device-specific stack autoencoder (SAE) denoising models were trained to denoise the input signals, which were generated by averaging 1, 2, 4, 8, 16, or 32 raw signals. Both supervised and unsupervised learning training strategies were tested for comparison. Mean squared error (MSE), signal-to-noise ratio (SNR) and the Bragg peak (BP) range uncertainty were used for model evaluation. Main results: After SAE denoising, the MSE was substantially reduced, and the SNR was enhanced. Overall, the supervised SAEs outperformed the unsupervised SAEs in BP range verification. For the high accuracy detector, it achieved a BP range uncertainty of 0.20 +/- 3.44 mm by averaging over 8 raw signals, while for the other two low accuracy detectors, they achieved the BP uncertainty of 1.44 +/- 6.45 mm and -0.23 +/- 4.88 mm by averaging 16 raw signals, respectively. Significance: We have proposed a deep learning based denoising method to enhance the SNR of protoacoustic measurements and improve the accuracy in BP range verification, which greatly reduces the dose and time for potential clinical applications., Comment: 15 pages, 5 figures

Objective. Artificial intelligence (AI) methods have gained popularity in medical imaging research. The size and scope of the training image datasets needed for successful AI model deployment does not always have the desired scale. In this paper, we introduce a medical image synthesis framework aimed at addressing the challenge of limited training datasets for AI models. Approach. The proposed 2D image synthesis framework is based on a diffusion model using a Swin-transformer-based network. This model consists of a forward Gaussian noise process and a reverse process using the transformer-based diffusion model for denoising. Training data includes four image datasets: chest x-rays, heart MRI, pelvic CT, and abdomen CT. We evaluated the authenticity, quality, and diversity of the synthetic images using visual Turing assessments conducted by three medical physicists, and four quantitative evaluations: the Inception score (IS), Fréchet Inception Distance score (FID), feature similarity and diversity score (DS, indicating diversity similarity) between the synthetic and true images. To leverage the framework value for training AI models, we conducted COVID-19 classification tasks using real images, synthetic images, and mixtures of both images. Main results. Visual Turing assessments showed an average accuracy of 0.64 (accuracy converging to 50 % indicates a better realistic visual appearance of the synthetic images), sensitivity of 0.79, and specificity of 0.50. Average quantitative accuracy obtained from all datasets were IS = 2.28, FID = 37.27, FDS = 0.20, and DS = 0.86. For the COVID-19 classification task, the baseline network obtained an accuracy of 0.88 using a pure real dataset, 0.89 using a pure synthetic dataset, and 0.93 using a dataset mixed of real and synthetic data. Significance. A image synthesis framework was demonstrated for medical image synthesis, which can generate high-quality medical images of different imaging modalities with the purpose of supplementing existing training sets for AI model deployment. This method has potential applications in many data-driven medical imaging research.

This paper explores the possibility of applying neural network architecture search to monocular 3D target detection tasks, so as to solve the problem that researchers need a lot of time and prior knowledge to manually design network structures. The algorithm applies the differentiable neural network architecture search technology to the backbones search of the 3D monocular target detection network FCOS3D. At the same time, to improve the accuracy of the algorithm for 3D target detection tasks and reduce the computational complexity of the algorithm, we add deformable convolution and depth separable convolution to the network searchable space. Finally, our algorithm is superior to the original FCOS3D algorithm in the KITTI3D monocular target detection dataset.

Objective. CBCTs in image-guided radiotherapy provide crucial anatomy information for patient setup and plan evaluation. Longitudinal CBCT image registration could quantify the inter-fractional anatomic changes, e.g. tumor shrinkage, and daily OAR variation throughout the course of treatment. The purpose of this study is to propose an unsupervised deep learning-based CBCT-CBCT deformable image registration which enables quantitative anatomic variation analysis. Approach. The proposed deformable registration workflow consists of training and inference stages that share the same feed-forward path through a spatial transformation-based network (STN). The STN consists of a global generative adversarial network (GlobalGAN) and a local GAN (LocalGAN) to predict the coarse- and fine-scale motions, respectively. The network was trained by minimizing the image similarity loss and the deformable vector field (DVF) regularization loss without the supervision of ground truth DVFs. During the inference stage, patches of local DVF were predicted by the trained LocalGAN and fused to form a whole-image DVF. The local whole-image DVF was subsequently combined with the GlobalGAN generated DVF to obtain the final DVF. The proposed method was evaluated using 100 fractional CBCTs from 20 abdominal cancer patients in the experiments and 105 fractional CBCTs from a cohort of 21 different abdominal cancer patients in a holdout test. Main Results. Qualitatively, the registration results show good alignment between the deformed CBCT images and the target CBCT image. Quantitatively, the average target registration error calculated on the fiducial markers and manually identified landmarks was 1.91 ± 1.18 mm. The average mean absolute error, normalized cross correlation between the deformed CBCT and target CBCT were 33.42 ± 7.48 HU, 0.94 ± 0.04, respectively. Significance. In summary, an unsupervised deep learning-based CBCT-CBCT registration method is proposed and its feasibility and performance in fractionated image-guided radiotherapy is investigated. This promising registration method could provide fast and accurate longitudinal CBCT alignment to facilitate inter-fractional anatomic changes analysis and prediction.

Under long cruise conditions, the forebody and inlet of a hypersonic vehicle were sensitive to deform under the aerodynamic thermal load, which hurt aerodynamic characteristics, intake performance, and aircraft safety. A pre-deformation method was proposed by using the numerical method of fluid-thermal-structural coupling based on the in-house software, aiming at the static deformation of a typical forebody-inlet model. Three pre-deformed models were combined by the forebody/inlet deformation and then compared the intake performance between the pre-deformed models and the ideal rigid model. The results show that the pre-deformation of the forebody can not only ensure the intake performance partly but also avoid the risk of excessive pressure on the intake wall. The pre-deformation of the inlet can reduce the flow separation near the ramp wall and reduce the pressure of intake, but it also reduces the total pressure recovery coefficient and affects intake performance negatively.

Objective. Computed tomography (CT) to material property conversion dominates proton range uncertainty, impacting the quality of proton treatment planning. Physics-based and machine learning-based methods have been investigated to leverage dual-energy CT (DECT) to predict proton ranges. Recent development includes physics-informed deep learning (DL) for material property inference. This paper aims to develop a framework to validate Monte Carlo dose calculation (MCDC) using CT-based material characterization models. Approach. The proposed framework includes two experiments to validate in vivo dose and water equivalent thickness (WET) distributions using anthropomorphic and porcine phantoms. Phantoms were irradiated using anteroposterior proton beams, and the exit doses and residual ranges were measured by MatriXX PT and a multi-layer strip ionization chamber. Two pre-trained conventional and physics-informed residual networks (RN/PRN) were used for mass density inference from DECT. Additional two heuristic material conversion models using single-energy CT (SECT) and DECT were implemented for comparisons. The gamma index was used for dose comparisons with criteria of 3%/3 mm (10% dose threshold). Main results. The phantom study showed that MCDC with PRN achieved mean gamma passing rates of 95.9% and 97.8% for the anthropomorphic and porcine phantoms. The rates were 86.0% and 79.7% for MCDC with the empirical DECT model. WET analyses indicated that the mean WET variations between measurement and simulation were −1.66 mm, −2.48 mm, and −0.06 mm for MCDC using a Hounsfield look-up table with SECT and empirical and PRN models with DECT. Validation experiments indicated that MCDC with PRN achieved consistent dose and WET distributions with measurement. Significance. The proposed framework can be used to identify the optimal CT-based material characterization model for MCDC to improve proton range uncertainty. The framework can systematically verify the accuracy of proton treatment planning, and it can potentially be implemented in the treatment room to be instrumental in online adaptive treatment planning.

A novel high-performance readout circuit with multi-Junction field-effect transistors (Multi-JFETs) and high-speed integrated operational amplifiers as the core was designed. By coupling the circuit to a large area passivated ion injection planar silicon (PIPS) detector, excellent α and β energy spectrum can be obtained. Therefore, it can be applied to radon monitor, radioactive aerosol monitor and other equipment to achieve high precision information on radioactive substances in the environment. The input stage of the circuit adopts the parallel structure of JFETs to achieve the matching between the amplifier circuit with the detector junction capacitance. The bias circuit adopts JFETs to form a constant current source while using the gate-source self-biased parallel structure to obtain large transconductance gain and good amplification linearity. The main amplifier circuit adopts a high-speed and low-noise operational amplifier with the advantages of high open-loop gain and stable quiescent point. The performance of this readout circuit was tested, in which the rise time of the signal is 35 ns, the sensitivity of charge-voltage conversion is better than 0.979 mV/fC when the input capacitance is less than 100 pF, and the equivalent noise charge is 1.55 fC, noise slope is 0.00366 fC/pF. By coupling the readout circuit to a PIPS detector with a sensitive area of 450 mm2, the signal-to-noise ratio of the output signal is 116:1 for the 241Am alpha source and 20:1 for the 90Sr-90Y beta particles. The energy spectrum measurement of the 238Pu source was performed, providing FWHM is 16.90 keV@5499 keV.

Piezoelectric actuators (PEAs) are widely used in ultra-precision detection platforms where nano-precision and non-magnetic features are required. With the development of the semiconductor industry, actuators develop toward the tendency of smaller size, higher precision, and longer travel. However, these demands are difficult to meet merely by virtue of a single piezoelectric actuator or a simple structured inchworm piezo motor, which makes it necessary to develop a new drive mode following a different drive principle. In this paper, a novel inchworm piezo motor with bender-type PEAs in phalanx distribution was proposed, which facilitates in reducing the dimension of the motor and enhances the performance and stability of the piezo motor. For the purpose of accommodating the bender-type PEAs and providing the preloads to the bender-type PEAs, a flexible mechanism housing was designed and the modal analysis was finished, avoid resonance and reduce structural vibration. Experimental results show that the resolution of the developed motor is 2 nm or less under the laser interference with an adoption rate of 10 MHz and a resolution of 0.1 nm, while the maximum stroke is over 19 mm at the constant speed of 2.3 mm s−1, and the maximum output force is 41.6 N.

Delamination is one of the common damages affecting the safety of composite structures. In this paper, a Lamb wavefield-based monogenic signal processing algorithm is proposed to quantify the delamination parameters in composite laminates, including location, size, shape, and depth. A quality-guided fast phase unwrapping algorithm is developed to solve the problem of phase wrapping after Riesz transform-based monogenic signal processing. Then, space distribution of the phase of Lamb wavefield can be extracted for calculating wavenumber distribution, which is related to the structural thickness or delamination depth and can be used for delamination imaging. Simulated Lamb wavefield signals calculated by finite element simulation are employed to evaluate the parameters of delamination in composite laminates. Compared with other traditional methods, the damage identification algorithm based on Riesz transform has excellent identification effect and shorter calculation time. The results show that the algorithm can be used not only for single delamination recognition but also for multi-delamination recognition with good accuracy. In particular, the interaction between incident waves along different ply directions and delamination is explored, and its influence on delamination quantification is studied, whose results are worthy of attention in engineering application. Finally, a completely non-contact laser ultrasonic system is established to obtain the Lamb wavefield with delamination. Experiments show that the algorithm can accurately quantify the location, size, shape, and depth of delamination.

With the continuous expansion of the power grid, the number of alarm information collected by the dispatching center is also increasing. How to filter out key information from massive alarm information, delete irrelevant data, classify the importance of alarm information, and make preparations for power grid fault diagnosis based on alarm information has become an urgent problem to be solved in online fault diagnosis. Based on this, this paper proposes an importance classification method of alarm information based on bigru attention to screen the data with the strongest correlation with fault, avoid the impact of irrelevant data on fault diagnosis, and meet the needs of intelligence. Firstly, the two-way gated loop (bigru) neural network layer is used to preprocess the alarm information text and extract the features of the deep-seated information; Secondly, the attention mechanism layer is used to assign corresponding weights to the extracted text deep-seated information; Finally, the text feature information of alarm information with different weights is put into the softmax function layer for alarm information text classification. Finally, an actual fault case in an area is tested to verify the effectiveness and practicability of the proposed method.

Accurate segmentation of glioma and its subregions plays an important role in radiotherapy treatment planning. Due to a very populated multiparameter magnetic resonance imaging image, manual segmentation tasks can be very time-consuming, meticulous, and prone to subjective errors. Here, we propose a novel deep learning framework based on mutual enhancing networks to automatically segment brain tumor subregions. The proposed framework is suitable for the segmentation of brain tumor subregions owing to the contribution of Retina U-Net followed by the implementation of a mutual enhancing strategy between the classification localization map (CLM) module and segmentation module. Retina U-Net is trained to accurately identify view-of-interest and feature maps of the whole tumor (WT), which are then transferred to the CLM module and segmentation module. Subsequently, CLM generated by the CLM module is integrated with the segmentation module to bring forth a mutual enhancing strategy. In this way, our proposed framework first focuses on WT through Retina U-Net, and since WT consists of subregions, a mutual enhancing strategy then further aims to classify and segment subregions embedded within WT. We implemented and evaluated our proposed framework on the BraTS 2020 dataset consisting of 369 cases. We performed a 5-fold cross-validation on 200 datasets and a hold-out test on the remaining 169 cases. To demonstrate the effectiveness of our network design, we compared our method against the networks without Retina U-Net, mutual enhancing strategy, and a recently published Cascaded U-Net architecture. Results of all four methods were compared to the ground truth for segmentation and localization accuracies. Our method yielded significantly (P

Focal boost to dominant intraprostatic lesions (DILs) has recently been proposed for prostate radiation therapy. Accurate and fast delineation of the prostate and DILs is thus required during treatment planning. In this paper, we develop a learning-based method using positron emission tomography (PET)/computed tomography (CT) images to automatically segment the prostate and its DILs. To enable end-to-end segmentation, a deep learning-based method, called cascaded regional-Net, is utilized. The first network, referred to as dual attention network, is used to segment the prostate via extracting comprehensive features from both PET and CT images. A second network, referred to as mask scoring regional convolutional neural network (MSR-CNN), is used to segment the DILs from the PET and CT within the prostate region. Scoring strategy is used to diminish the misclassification of the DILs. For DIL segmentation, the proposed cascaded regional-Net uses two steps to remove normal tissue regions, with the first step cropping images based on prostate segmentation and the second step using MSR-CNN to further locate the DILs. The binary masks of DILs and prostates of testing patients are generated on the PET/CT images by the trained model. For evaluation, we retrospectively investigated 49 prostate cancer patients with PET/CT images acquired. The prostate and DILs of each patient were contoured by radiation oncologists and set as the ground truths and targets. We used five-fold cross-validation and a hold-out test to train and evaluate our method. The mean surface distance and DSC values were 0.666 ± 0.696 mm and 0.932 ± 0.059 for the prostate and 0.814 ± 1.002 mm and 0.801 ± 0.178 for the DILs among all 49 patients. The proposed method has shown promise for facilitating prostate and DIL delineation for DIL focal boost prostate radiation therapy.

The delineation of the prostate and organs-at-risk (OARs) is fundamental to prostate radiation treatment planning, but is currently labor-intensive and observer-dependent. We aimed to develop an automated computed tomography (CT)-based multi-organ (bladder, prostate, rectum, left and right femoral heads (RFHs)) segmentation method for prostate radiation therapy treatment planning. The proposed method uses synthetic MRIs (sMRIs) to offer superior soft-tissue information for male pelvic CT images. Cycle-consistent adversarial networks (CycleGAN) were used to generate CT-based sMRIs. Dual pyramid networks (DPNs) extracted features from both CTs and sMRIs. A deep attention strategy was integrated into the DPNs to select the most relevant features from both CTs and sMRIs to identify organ boundaries. The CT-based sMRI generated from our previously trained CycleGAN and its corresponding CT images were inputted to the proposed DPNs to provide complementary information for pelvic multi-organ segmentation. The proposed method was trained and evaluated using datasets from 140 patients with prostate cancer, and were then compared against state-of-art methods. The Dice similarity coefficients and mean surface distances between our results and ground truth were 0.95 ± 0.05, 1.16 ± 0.70 mm; 0.88 ± 0.08, 1.64 ± 1.26 mm; 0.90 ± 0.04, 1.27 ± 0.48 mm; 0.95 ± 0.04, 1.08 ± 1.29 mm; and 0.95 ± 0.04, 1.11 ± 1.49 mm for bladder, prostate, rectum, left and RFHs, respectively. Mean center of mass distances was within 3 mm for all organs. Our results performed significantly better than those of competing methods in most evaluation metrics. We demonstrated the feasibility of sMRI-aided DPNs for multi-organ segmentation on pelvic CT images, and its superiority over other networks. The proposed method could be used in routine prostate cancer radiotherapy treatment planning to rapidly segment the prostate and standard OARs.

In this paper, raw materials such as fly ash-S95 grade mineral powder composite mineral admixture, barite coarse and fine aggregate and polycarboxylic acid water reducer are selected to successfully formulate C30 self-compacting barite concrete with good performance. It also analyzes the development law of early drying shrinkage of self-compacting barite concrete with age, and discusses the effects of fly ash and slag mixing ratio and water reducer content on concrete work, compressive strength and early drying shrinkage. The results show that: the drying shrinkage of self-compacting barite concrete develops rapidly within 1 day of age, and the drying shrinkage rate at 1 day accounts for 50% to 66% of that at 7 days of age, and then the growth trend is slow; as the proportion of fly ash increases, the performance of the concrete mixture improves. While the strength of concrete decreases in the early and mid-term, the strength gradually increases in the later period, and the shrinkage strain of concrete in the early stage also increases significantly; the proper amount of water-reducing agent can greatly improve the fluidity of the mixture. In the dosage range of 1.2%~1.4%, with the increase of the amount of water reducing agent, the compressive strength first increases and then decreases, reaching the maximum when the dosage is 1.3%, and the early drying shrinkage strain of concrete increases slightly.

The purpose of this study is to develop a deep learning method for thyroid delineation with high accuracy, efficiency, and robustness in noncontrast-enhanced head and neck CTs. The cross-sectional analysis consisted of six tests, including randomized cross-validation and hold-out experiments, tests of prediction accuracy between cancer and benign and cross-gender analysis were performed to evaluate the proposed deep-learning-based performance method. CT images of 1977 patients with suspected thyroid carcinoma were retrospectively investigated. The automatically segmented thyroid gland volume was compared against physician-approved clinical contours using metrics, the Pearson correlation and Bland–Altman analysis. Quantitative metrics included: the Dice similarity coefficient (DSC), sensitivity, specificity, Jaccard index (JAC), Hausdorff distance (HD), mean surface distance (MSD), residual mean square distance (RMSD) and the center of mass distance (CMD). The robustness of the proposed method was further tested using the nonparametric Kruskal–Wallis test to assess the equality of distribution of DSC values. The proposed method’s accuracy remained high through all the tests, with the median DSC, JAC, sensitivity and specificity higher than 0.913, 0.839, 0.856 and 0.979, respectively. The proposed method also resulted in median MSD, RMSD, HD and CMD, of less than 0.31 mm, 0.48 mm, 2.06 mm and 0.50 mm, respectively. The MSD and RMSD were 0.40 ± 0.29 mm and 0.70 ± 0.46 mm, respectively. Concurrent testing of the proposed method with 3D U-Net and V-Net showed that the proposed method had significantly improved performance. The proposed deep-learning method achieved accurate and robust performance through six cross-sectional analysis tests.

Jet impingement is widely used in anti-icing and de-icing of aeroengines. To improve the efficiency of anti-icing and de-icing, heat transfer in jet impingement flow should be further enhanced. By using large eddy simulation (LES) method, jet impingement flow was analysed in time-domain and spatial-domain. It was revealed that the jet flow has quasi-periodic characteristics in time, and heat transfer is dominated by vortex structure. In addition, the impinged surface curvature has a certain influence on the spatial distribution of Nusselt number. Better heat transfer effect can be achieved by inducing more vortices and applying proper surface curvature.

Hypersonic aircrafts and aero-engine combustion chambers both generate non-equilibrium and high-enthalpy flows and bring complex material-relied heat convection performance. The convective heating prediction is difficult due to unknown surface thermal state, leading to poor usability of wall temperature correlation method (WTCM). This paper aims at improving WTCM for convective heating prediction in chemically reacting flows through coupling computation of catalysis on thermal protection materials. Modified WTCM for chemically reacting flows accounts for two distinct physical events driven by temperature gradient and species reaction, which follow the Fourier’s and Fick’s laws, respectively. Preliminary validation testing demonstrates the feasibility of the modified WTCM to rapidly evaluate aerodynamic heating with limited deviation. The current research provides essential technical support for the evaluation and design of hypersonic aircrafts and aero-engine combustion chambers.

Due to the inter- and intra- variation of respiratory motion, it is highly desired to provide real-time volumetric images during the treatment delivery of lung stereotactic body radiation therapy (SBRT) for accurate and active motion management. In this proof-of-concept study, we propose a novel generative adversarial network integrated with perceptual supervision to derive instantaneous volumetric images from a single 2D projection. Our proposed network, named TransNet, consists of three modules, i.e. encoding, transformation and decoding modules. Rather than only using image distance loss between the generated 3D images and the ground truth 3D CT images to supervise the network, perceptual loss in feature space is integrated into loss function to force the TransNet to yield accurate lung boundary. Adversarial supervision is also used to improve the realism of generated 3D images. We conducted a simulation study on 20 patient cases, who had received lung SBRT treatments in our institution and undergone 4D-CT simulation, and evaluated the efficacy and robustness of our method for four different projection angles, i.e. 0°, 30°, 60° and 90°. For each 3D CT image set of a breathing phase, we simulated its 2D projections at these angles. For each projection angle, a patient’s 3D CT images of 9 phases and the corresponding 2D projection data were used to train our network for that specific patient, with the remaining phase used for testing. The mean absolute error of the 3D images obtained by our method are 99.3 ± 14.1 HU. The peak signal-to-noise ratio and structural similarity index metric within the tumor region of interest are 15.4 ± 2.5 dB and 0.839 ± 0.090, respectively. The center of mass distance between the manual tumor contours on the 3D images obtained by our method and the manual tumor contours on the corresponding 3D phase CT images are within 2.6 mm, with a mean value of 1.26 mm averaged over all the cases. Our method has also been validated in a simulated challenging scenario with increased respiratory motion amplitude and tumor shrinkage, and achieved acceptable results. Our experimental results demonstrate the feasibility and efficacy of our 2D-to-3D method for lung cancer patients, which provides a potential solution for in-treatment real-time on-board volumetric imaging for tumor tracking and dose delivery verification to ensure the effectiveness of lung SBRT treatment.

Correcting or reducing the effects of voxel intensity non-uniformity (INU) within a given tissue type is a crucial issue for quantitative magnetic resonance (MR) image analysis in daily clinical practice. Although having no severe impact on visual diagnosis, the INU can highly degrade the performance of automatic quantitative analysis such as segmentation, registration, feature extraction and radiomics. In this study, we present an advanced deep learning based INU correction algorithm called residual cycle generative adversarial network (res-cycle GAN), which integrates the residual block concept into a cycle-consistent GAN (cycle-GAN). In cycle-GAN, an inverse transformation was implemented between the INU uncorrected and corrected magnetic resonance imaging (MRI) images to constrain the model through forcing the calculation of both an INU corrected MRI and a synthetic corrected MRI. A fully convolution neural network integrating residual blocks was applied in the generator of cycle-GAN to enhance end-to-end raw MRI to INU corrected MRI transformation. A cohort of 55 abdominal patients with T1-weighted MR INU images and their corrections with a clinically established and commonly used method, namely, N4ITK were used as a pair to evaluate the proposed res-cycle GAN based INU correction algorithm. Quantitatively comparisons of normalized mean absolute error (NMAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC) indices, and spatial non-uniformity (SNU) were made among the proposed method and other approaches. Our res-cycle GAN based method achieved an NMAE of 0.011 ± 0.002, a PSNR of 28.0 ± 1.9 dB, an NCC of 0.970 ± 0.017, and a SNU of 0.298 ± 0.085. Our proposed method has significant improvements (p < 0.05) in NMAE, PSNR, NCC and SNU over other algorithms including conventional GAN and U-net. Once the model is well trained, our approach can automatically generate the corrected MR images in a few minutes, eliminating the need for manual setting of parameters.

This paper presents a review of deep learning (DL) based medical image registration methods. We summarized the latest developments and applications of DL-based registration methods in the medical field. These methods were classified into seven categories according to their methods, functions and popularity. A detailed review of each category was presented, highlighting important contributions and identifying specific challenges. A short assessment was presented following the detailed review of each category to summarize its achievements and future potentials. We provided a comprehensive comparison among DL-based methods for lung and brain deformable registration using benchmark datasets. Lastly, we analyzed the statistics of all the cited works from various aspects, revealing the popularity and future trend of development in medical image registration using deep learning., 32 pages, 4 figures, 9 tables

The segmentation of neoplasms is an important part of radiotherapy treatment planning, monitoring disease progression, and predicting patient outcome. In the brain, functional magnetic resonance imaging (MRI) like dynamic susceptibility contrast enhanced (DSCE) or T1-weighted dynamic contrast enhanced (DCE) perfusion MRI are important tools for diagnosis. They play a crucial role in providing pre-operative assessment of tumor histology, grading, and biopsy guidance. However, the manual contouring of these neoplasms is tedious, expensive, time-consuming, and vulnerable to inter-observer variability. In this work, we propose a 3D mask region-based convolutional neural network (R-CNN) method to automatically segment brain tumors in DSCE MRI perfusion images. As our goal is to simultaneously localize and segment the tumor, our training process contained both a region-of-interest (ROI) localization and regression with voxel-wise segmentation. The combination of classification loss, ROI location and size regression loss, and segmentation loss were used to supervise the proposed network. We retrospectively investigated 21 patients' perfusion images, with between 50 and 70 perfusion time point volumes, a total of 1260 3D volumes. Tumor contours were automatically segmented by our proposed method and compared against other state-of-the-art methods and those delineated by physicians as the ground truth. The results of our method demonstrated good agreement with the ground truth contours. The average DSC, precision, recall, Hausdorff distance, mean surface distance (MSD), root MSD, and center of mass distance were 0.90 ± 0.04, 0.91 ± 0.04, 0.90 ± 0.06, 7.16 ± 5.78 mm, 0.45 ± 0.34 mm, 1.03 ± 0.72 mm, and 0.86 ± 0.91 mm, respectively. These results support the feasibility of our method in accurately localizing and segmenting brain tumors in DSCE perfusion MRI. Our 3D Mask R-CNN segmentation method in DSCE perfusion imaging has great promise for future clinical use.

Deformable image registration (DIR) of 4D-CT images is important in multiple radiation therapy applications including motion tracking of soft tissue or fiducial markers, target definition, image fusion, dose accumulation and treatment response evaluations. It is very challenging to accurately and quickly register 4D-CT abdominal images due to its large appearance variances and bulky sizes. In this study, we proposed an accurate and fast multi-scale DIR network (MS-DIRNet) for abdominal 4D-CT registration. MS-DIRNet consists of a global network (GlobalNet) and local network (LocalNet). GlobalNet was trained using down-sampled whole image volumes while LocalNet was trained using sampled image patches. MS-DIRNet consists of a generator and a discriminator. The generator was trained to directly predict a deformation vector field (DVF) based on the moving and target images. The generator was implemented using convolutional neural networks with multiple attention gates. The discriminator was trained to differentiate the deformed images from the target images to provide additional DVF regularization. The loss function of MS-DIRNet includes three parts which are image similarity loss, adversarial loss and DVF regularization loss. The MS-DIRNet was trained in a completely unsupervised manner meaning that ground truth DVFs are not needed. Different from traditional DIRs that calculate DVF iteratively, MS-DIRNet is able to calculate the final DVF in a single forward prediction which could significantly expedite the DIR process. The MS-DIRNet was trained and tested on 25 patients' 4D-CT datasets using five-fold cross validation. For registration accuracy evaluation, target registration errors (TREs) of MS-DIRNet were compared to clinically used software. Our results showed that the MS-DIRNet with an average TRE of 1.2 ± 0.8 mm outperformed the commercial software with an average TRE of 2.5 ± 0.8 mm in 4D-CT abdominal DIR, demonstrating the superior performance of our method in fiducial marker tracking and overall soft tissue alignment.

PURPOSE: To develop an automated cone-beam computed tomography (CBCT) multi-organ segmentation method for potential CBCT-guided adaptive radiation therapy workflow. METHODS AND MATERIALS: The proposed method combines the deep leaning-based image synthesis method, which generates magnetic resonance images (MRIs) with superior soft-tissue contrast from on-board setup CBCT images to aid CBCT segmentation, with a deep attention strategy, which focuses on learning discriminative features for differentiating organ margins. The whole segmentation method consists of 3 major steps. First, a cycle-consistent adversarial network (CycleGAN) was used to estimate a synthetic MRI (sMRI) from CBCT images. Second, a deep attention network was trained based on sMRI and its corresponding manual contours. Third, the segmented contours for a query patient was obtained by feeding the patient’s CBCT images into the trained sMRI estimation and segmentation model. In our retrospective study, we included 100 prostate cancer patients, each of whom has CBCT acquired with prostate, bladder and rectum contoured by physicians with MRI guidance as ground truth. We trained and tested our model with separate datasets among these patients. The resulting segmentations were compared with physicians’ manual contours. RESULTS: The Dice similarity coefficient and mean surface distance indices between our segmented and physicians’ manual contours (bladder, prostate, and rectum) were 0.95±0.02, 0.44±0.22 mm, 0.86±0.06, 0.73±0.37 mm, and 0.91±0.04, 0.72±0.65 mm, respectively. CONCLUSION: We have proposed a novel CBCT-only pelvic multi-organ segmentation strategy using CBCT-based sMRI and validated its accuracy against manual contours. This technique could provide accurate organ volume for treatment planning without requiring MR images acquisition, greatly facilitating routine clinical workflow.

In order to study the waste cathode ray tube glass fine aggregate (particle size < 0.15 mm) on barite concreteperformance, the waste cathode ray tube glass fine aggregate is blended into the barite concrete at 0%, 4%, 6%, 8%, and 10% of the mass of the barite sand quality to conduct experimental research on the related properties of barite concrete. The experimental results show that when the amount of waste cathode ray tube glass fine aggregate is 6%∼8%, the prepared barite concrete has the best working performance. The slump expansion reaches 560mm×560mm, and the 28d cube compressive strength is 37.76MPa. The V-box passage time is 17.56s, and barite concrete has the best shielding gamma ray capability.

Sapphire crystals are used in the manufacture of LEDs, optical window materials, etc. The presence of air bubbles in the crystal affects the optical properties of the material. If the position of the bubble is determined, the bubble can be bypassed for subsequent slicing to obtain pure high quality sapphire crystal. At present, the detection and identification of bubbles in sapphire crystals still rely on human eye observation and empirical judgment, which is inefficient and easily harmful to the human eye. It is necessary to propose a highly efficient method of machine vision detection instead of human eye detection. Based on the machine vision detection technology, this paper uses the laser as the light source to enter the ingot from the bottom, which produces the laser scattering effect. With this effect as the imaging principle, the image is collected by CCD . The image is analyzed by image processing means to achieve detection of bubbles, this paper proposes and elaborates the following two steps: edge detection based on edge pixels and locking the target area based on the calibration connected domain. The experimental results show that compared with the human eye detection method, the detection method greatly improves the detection rate and accuracy of identifying the bubble in the sapphire and determining its position.

The drone radiation measurement system has a sharp increase in demand for environmental radiation monitoring due to its easy manoeuvrability, low aerial survey cost, high patrol efficiency and no personnel safety concerns. In order to adapt to the fast, flexible and efficient characteristics of the drone, the detector for the Unmanned Aerial Vehicle (UAV) radiation measurement generally consists of a plurality of small-sized detectors to form a detection array. Due to the low single crystal count rate and the influence of electronic system temperature and packaging, many detectors of the UAV radiation measurement system will have “spectral drifts”, which affects the quality of the synthesized spectrum and affects the accuracy of the whole system. Aiming at the characteristics of UAV spectrum measurement system, a three feature point recognition model is established, and the K-40 feature peak (1.46MeV) is approximated by reverse piecewise summation. Thus, the backscattering peaks and characteristic peaks K-40 of the measured spectrum can be obtained to achieve fast spectrum stabilization. Finally, through Gaussian Mixture Model (GMM) peak estimation and experimental validation, the method can obtain the characteristic peak position accurately, and the time between instrument start-up and automatic spectrum stabilization is less than 2 minutes.

Attenuation correction (AC) of PET/MRI faces challenges including inter-scan motion, image artifacts such as truncation and distortion, and erroneous transformation of structural voxel-intensities to PET mu-map values. We propose a deep-learning-based method to derive synthetic CT (sCT) images from non-attenuation corrected PET (NAC PET) images for AC on whole-body PET/MRI imaging. A 3D cycle-consistent generative adversarial networks (CycleGAN) framework was employed to synthesize CT images from NAC PET. The method learns a transformation that minimizes the difference between sCT, generated from NAC PET, and true CT. It also learns an inverse transformation such that cycle NAC PET image generated from the sCT is close to true NAC PET image. A self-attention strategy was also utilized to identify the most informative component and mitigate the disturbance of noise. We conducted a retrospective study on a total of 119 sets of whole-body PET/CT, with 80 sets for training and 39 sets for testing and evaluation. The whole-body sCT images generated with proposed method demonstrate great resemblance to true CT images, and show good contrast on soft tissue, lung and bony tissues. The mean absolute error (MAE) of sCT over true CT is less than 110 HU. Using sCT for whole-body PET AC, the mean error of PET quantification is less than 1% and normalized mean square error (NMSE) is less than 1.4%. Average normalized cross correlation on whole body is close to one, and PSNR is larger than 42 dB. We proposed a deep learning-based approach to generate sCT from whole-body NAC PET for PET AC. sCT generated with proposed method shows great similarity to true CT images both qualitatively and quantitatively, and demonstrates great potential for whole-body PET AC in the absence of structural information.

Magnetic resonance imaging (MRI) has been widely used in combination with computed tomography (CT) radiation therapy because MRI improves the accuracy and reliability of target delineation due to its superior soft tissue contrast over CT. The MRI-only treatment process is currently an active field of research since it could eliminate systematic MR-CT co-registration errors, reduce medical cost, avoid diagnostic radiation exposure, and simplify clinical workflow. The purpose of this work is to validate the application of a deep learning-based method for abdominal synthetic CT (sCT) generation by image evaluation and dosimetric assessment in a commercial proton pencil beam treatment planning system (TPS). This study proposes to integrate dense block into a 3D cycle-consistent generative adversarial networks (cycle GAN) framework in an effort to effectively learn the nonlinear mapping between MRI and CT pairs. A cohort of 21 patients with co-registered CT and MR pairs were used to test the deep learning-based sCT image quality by leave-one-out cross validation. The CT image quality, dosimetric accuracy and the distal range fidelity were rigorously checked, using side-by-side comparison against the corresponding original CT images. The average mean absolute error (MAE) was 72.87 ± 18.16 HU. The relative differences of the statistics of the PTV dose volume histogram (DVH) metrics between sCT and CT were generally less than 1%. Mean 3D gamma analysis passing rate of 1 mm/1%, 2 mm/2%, 3 mm/3% criteria with 10% dose threshold were 90.76% ± 5.94%, 96.98% ± 2.93% and 99.37% ± 0.99%, respectively. The median, mean and standard deviation of absolute maximum range differences were 0.170 cm, 0.186 cm and 0.155 cm. The image similarity, dosimetric and distal range agreement between sCT and original CT suggests the feasibility of further development of an MRI-only workflow for liver proton radiotherapy.

Deriving accurate attenuation maps for PET/MRI remains a challenging problem because MRI voxel intensities are not related to properties of photon attenuation and bone/air interfaces have similarly low signal. This work presents a learning-based method to derive patient-specific computed tomography (CT) maps from routine T1-weighted MRI in their native space for attenuation correction of brain PET. We developed a machine-learning-based method using a sequence of alternating random forests under the framework of an iterative refinement model. Anatomical feature selection is included in both training and predication stages to achieve optimal performance. To evaluate its accuracy, we retrospectively investigated 17 patients, each of which has been scanned by PET/CT and MR for brain. The PET images were corrected for attenuation on CT images as ground truth, as well as on pseudo CT (PCT) images generated from MR images. The PCT images showed mean average error of 66.1 ± 8.5 HU, average correlation coefficient of 0.974 ± 0.018 and average Dice similarity coefficient (DSC) larger than 0.85 for air, bone and soft tissue. The side-by-side image comparisons and joint histograms demonstrated very good agreement of PET images after correction by PCT and CT. The mean differences of voxel values in selected VOIs were less than 4%, the mean absolute difference of all active area is around 2.5%, and the mean linear correlation coefficient is 0.989 ± 0.017 between PET images corrected by CT and PCT. This work demonstrates a novel learning-based approach to automatically generate CT images from routine T1-weighted MR images based on a random forest regression with patch-based anatomical signatures to effectively capture the relationship between the CT and MR images. Reconstructed PET images using the PCT exhibit errors well below accepted test/retest reliability of PET/CT indicating high quantitative equivalence.

Due to their structural simplicity, linear motors are increasingly receiving attention for use in high velocity and high precision applications. The force ripple, as a space-periodic disturbance, however, would deteriorate the achievable dynamic performance. Conventional force ripple measurement approaches are time-consuming and have high requirements on the experimental conditions. In this paper, a novel learning identification algorithm is proposed for force ripple intelligent measurement and compensation. Existing identification schemes always use all the error signals to update the parameters in the force ripple. However, the error induced by noise is non-effective for force ripple identification, and even deteriorates the identification process. In this paper only the most pertinent information in the error signal is utilized for force ripple identification. Firstly, the effective error signals caused by the reference trajectory and the force ripple are extracted by projecting the overall error signals onto a subspace spanned by the physical model of the linear motor as well as the sinusoidal model of the force ripple. The time delay in the linear motor is compensated in the basis functions. Then, a data-driven approach is proposed to design the learning gain. It balances the trade-off between convergence speed and robustness against noise. Simulation and experimental results validate the proposed method and confirm its effectiveness and superiority.

Five heavy metal contents from five sediments and seven sediment profiles in an upstream reach of Zhedong river in Laowangzhai gold deposit were investigated in this research, along with analysis of the horizontal distribution, the surface distribution, the vertical distribution and the interlayer distribution of five heavy metal contents: arsenic (As), mercury (Hg), copper (Cu), lead (Pb) and zinc (Zn). The potential ecological risk of five heavy metals was evaluated to help understanding pollution control of Laowangzhai deposit.

We present a computational modeling approach to study the fusion of multicellular aggregate systems in a novel scaffold-less biofabrication process, known as 'bioprinting'. In this novel technology, live multicellular aggregates are used as fundamental building blocks to make tissues or organs (collectively known as the bio-constructs,) via the layer-by-layer deposition technique or other methods; the printed bio-constructs embedded in maturogens, consisting of nutrient-rich bio-compatible hydrogels, are then placed in bioreactors to undergo the cellular aggregate fusion process to form the desired functional bio-structures. Our approach reported here is an agent-based modeling method, which uses the kinetic Monte Carlo (KMC) algorithm to evolve the cellular system on a lattice. In this method, the cells and the hydrogel media, in which cells are embedded, are coarse-grained to material's points on a three-dimensional (3D) lattice, where the cell-cell and cell-medium interactions are quantified by adhesion and cohesion energies. In a multicellular aggregate system with a fixed number of cells and fixed amount of hydrogel media, where the effect of cell differentiation, proliferation and death are tactically neglected, the interaction energy is primarily dictated by the interfacial energy between cell and cell as well as between cell and medium particles on the lattice, respectively, based on the differential adhesion hypothesis. By using the transition state theory to track the time evolution of the multicellular system while minimizing the interfacial energy, KMC is shown to be an efficient time-dependent simulation tool to study the evolution of the multicellular aggregate system. In this study, numerical experiments are presented to simulate fusion and cell sorting during the biofabrication process of vascular networks, in which the bio-constructs are fabricated via engineering designs. The results predict the feasibility of fabricating the vascular structures via the bioprinting technology and demonstrate the morphological development process during cellular aggregate fusion in various engineering designed structures. The study also reveals that cell sorting will perhaps not significantly impact the final fabricated products, should the maturation process be well-controlled in bioprinting.

The SC Project is an alliance of 10 colleges and universities working together to achieve the goal of engineering a functional, 3D, bioengineered construct. Scientific progress includes computational modeling of vascular trees and experimental testing of natural and engineered constructs. Future directions of the science focus on overcoming challenges such as scalability, sustainability of biofabricated constructs, and identification of chemical or physiological factors that can accelerate the differentiation and maturation of biofabricated vascular tissues (maturogens). Studies include those of hemodynamic forces or growth factors that can promote expression and assembly of collagen and elastin fibers.