Search

PURPOSES: To introduce a novel surface-based dose mapping method to improve quantitative bladder dosimetric assessment in prostate cancer (PC) radiotherapy. METHODS: Based on the planning and daily pre and post-fraction MRIs of 12 PC patients, bladder surface models (SM) were generated on manually delineated contours and regionally aligned via surface-based registration. Subsequently, bladder surface dose models (SDM) were created using face-wise dose sampling. To determine the bladder intrafractional and interfractional motion and dose variation, we performed a pose analysis between pre and post-fraction bladder SMs, as well as surface mapping for fractional SMs. Discrepancies between the received dose, accumulated from daily SDMs, and the planned dose were then assessed on the corresponding SDMs. Complementary to the surface dose mapping, dose-surface-histograms (DSHs) based comparisons were also performed. RESULTS: The intra-fraction pose analysis revealed a significant (p

Soft sensors have been extensively developed to estimate the difficult-to-measure quality variables for real-time process monitoring and control. Process nonlinearities and dynamics are two main challenges for accurate soft sensor modeling. To cope with these problems, two temporal convolutional network (TCN)-based soft sensor models are developed in this paper. With a hierarchy of temporal convolution kernels and large receptive fields, TCN is able to describe the nonlinearities and long dynamic dependence of process variables. Thus, a TCN model is first designed for nonlinear dynamic soft sensing. However, the original TCN model neglects the auto-correlations of the quality variable, as well as the cross-correlations between the quality variable and process variables. Hence, an autoregressive TCN (AR-TCN) is further constructed by integrating the lagged quality data in the past neighboring samples for quality prediction of the current sample. Thus, AR-TCN is able to capture the auto-correlations and cross-correlations between the quality variables and process variables, which are more beneficial for quality prediction. The effectiveness of two proposed models is validated on an industrial debutanizer column and an industrial hydrocracking process.

Recently, deep learning has attracted increasing attention for soft sensor applications in industrial processes. Hierarchical features can be learned from massive process data by deep learning, which is the key step for quality variable prediction. However, few deep learning algorithms consider the neighborhood structure of data samples for feature extraction in industrial processes. In this paper, a novel stacked neighborhood preserving autoencoder (S-NPAE) is proposed to extract hierarchical neighborhood-preserving features. As for each NPAE, a novel loss function is proposed to reconstruct the input data and preserve the neighborhood structure of the input data simultaneously. By minimizing this loss function, NPAE can efficiently extract the neighborhood-preserved features from its input data. Then, the deep S-NPAE network is constructed by stacking multiple NPAEs in a hierarchical way. Finally, the extracted features can be used for accurate quality prediction in soft sensor modeling. The experimental results on an industrial hydrocracking process demonstrate the effectiveness of the proposed method when compared with other commonly used methods.

A series of rare-earth transition metal GdFeCo films are fabricated by sputtering a composite target. The effects of sputtering parameters on perpendicular magnetic properties of GdFeCo films are systematically investigated. Typical perpendicular anisotropy of the order of 105 erg/cm3 is witnessed. Tunable changes in Gd-to-FeCo ratio in the films are demonstrated by manipulating sputtering parameters including sputtering power, working pressure, and gas flow rate. Magnetic compensation is shown to be achieved in the alloy films by changing sputtering parameters and film thickness, where the lowest net magnetization and the largest coercivity are simultaneously demonstrated. A wide range of variation of perpendicular magnetic properties can be easily realized by fabricating GdFeCo films with the composition across magnetic compensation point. Our results provide a low-cost and time-saving way for the fabrication of GdFeCo alloy films with desired properties.

The fractionation system of industrial hydrocracking aims to fractionate the feedstocks to highly-valued distillates according to different boiling ranges. Therefore, the operation optimization for the system can effectively increase output and profits of factory. However, there is a complicated series-parallel structure between the distillation columns, which makes it very difficult to optimize the system. In this study, a two-stage optimization method for the series-parallel fractionation system of industrial hydrocracking is proposed. Firstly, the fractionation system is divided into an optimal structure through a phased calculation method to cut down unnecessary computation. Then, an adaptive logic of computing resource and a new sampling and evaluation strategy are adopted in candidate adaptive state transition algorithm (CASTA) to solve the optimization problem. CASTA can balance the computation resource and speed up optimization speed. Benchmarks are provided to illustrate the effectiveness of the proposed CASTA. Moreover, the two-stage optimization method was successfully used in the operation optimization of the series-parallel fractionation system of industrial hydrocracking, and achieved more efficient results than other optimization algorithms.

Industrial process data are naturally complex time series with high nonlinearities and dynamics. To model nonlinear dynamic processes, a long short-term memory (LSTM) network is very suitable for soft sensor model development. However, the original LSTM does not consider variable and sample relevance for quality prediction. In order to overcome this problem, a spatiotemporal attention-based LSTM network is proposed for soft sensor modeling, which can, not only identify important input variables that are related to the quality variable at each time step, but also adaptively discover quality-related hidden states across all time steps. By taking the spatiotemporal quality-relevant interactions into consideration, the prediction performance can be improved for the soft sensor model. The effectiveness and flexibility of the proposed model is demonstrated on an industrial hydrocracking process to predict the initial boiling points of heavy naphtha and aviation kerosene.

Concurrent process-quality monitoring helps discover quality-relevant process anomalies and quality-irrelevant process anomalies. It especially works well in chemical plants with faults that cause quality problems. Traditional monitoring strategies are limitedly applied in chemical plants because quality targets in training data are insufficient. It is hard for inflexible models to fully capture the strongly nonlinear process-quality correlations. Also, deterministic models are mapped from process variables to qualities without any consideration of uncertainties. Simultaneously, a slow sampling rate for quality variables is ubiquitous in chemical plants since a product quality test is often time-consuming and expensive. Motivated by these limitations, this paper proposes a new concurrent process-quality monitoring scheme based on a probabilistic generative deep learning model developed from variational autoencoder. The supervised model is firstly developed and then the semi-supervised version is extended to solve the issue of missing targets. Especially, the semi-supervised learning algorithm is accomplished with an optimal parameter estimation in the light of maximum likelihood principle and no any hyperparameters are introduced. Two case studies validate that the proposed method effectively outperforms the other comparative methods in concurrent process-quality monitoring.

In modern industrial processes, soft sensors have played increasingly important roles for effective process monitoring, control and optimization. Deep learning has shown excellent ability for hierarchical nonlinear feature representation in soft sensors. However, the existing deep learning based soft sensors are mostly trained offline and applied online without updating mechanism. This may cause their performance degradation in time-varying processes. To deal with this problem, an adaptive updating framework is proposed for deep learning, which is based on just-in-time fine-tuning of stacked autoencoder (JIT-SAE). In JIT-SAE, an offline SAE model is first trained with layer-wise unsupervised pre-training and supervised fine tuning. For online prediction, the network is dynamically fine-tuned upon the query sample. For each query sample, the most relevant labeled samples are selected to form a fine-tuning dataset from the historical labeled database, which is regularly augmented once new labeled samples are available from laboratory analysis. Moreover, each relevant sample is assigned with a weight according to its similarity with the query sample. Then, the deep network is fine-tuned with these relevant labeled samples by designing a weighted loss function. Thus, JIT-SAE is able to track the newest process running state timely and match the data pattern accurately. Case study on an industrial hydrocracking process is provided to demonstrate the effectiveness of the JIT-SAE framework.

Deep belief network (DBN) has recently been applied for soft sensor modeling with its excellent feature representation capacity. However, DBN cannot guarantee that the extracted features are quality-related and beneficial for further quality prediction. To solve this problem, a novel supervised DBN (SDBN) is proposed in this article by introducing the quality information into the training phase. SDBN consists of multiple supervised restricted Boltzmann machines (SRBMs) with a stacked structure. In each SRBM, the quality variables are added to the visible layer for network pretraining and feature learning. Thus, the pretrained weights can act as better initializations for the whole network for fine-tuning. Moreover, it can ensure that the learned features are largely quality-related for soft sensor. Finally, the SDBN-based soft sensor model is applied to two industrial plants of a debutanizer column and a hydrocracking process for quality prediction.

Complex industrial process data often exhibit nonlinear static and dynamic characteristics. Traditional deep learning methods such as stacked autoencoder (SAE) have excellent nonlinear static feature learning capabilities, but they ignore the dynamic correlation existing in process data. Feature learning based on manifold learning using neighborhood structure preserving has been widely used in industrial dynamic process monitoring. However, most of the manifold learning methods extract linear features, and complex nonlinearities in process data are ignored. Therefore, a novel spatiotemporal neighborhood preserving stack autoencoder (STNP-SAE) is proposed to simultaneously learn deep nonlinear static and dynamic features of process data in this article. By constructing the spatial and temporal adjacent graphs, STNP-SAE can capture the spatiotemporal neighborhood structure information of process data during the feature learning process. Then, STNP-SAE is used to construct a soft sensor framework for quality prediction. The prediction performance of the proposed method is validated on a practical industrial process.

Recently, deep-learning (DL)-based soft sensor has been widely applied to industrial processes, which plays a vital role for process monitoring, control, and optimization. However, most existing soft sensor models are established for only one quality variable or multiple quality variables with the same sampling rate. There are very few models focusing on prediction for multirate quality variables, especially with DL networks. To handle this problem, a novel DL strategy based on multirate stacked autoencoder (MR-SAE) is proposed. In MR-SAE, the network is composed of two parts: the cascade shared network for joint feature representations and the parallel quality-specific network for individual feature learning and quality prediction. The training procedure consists of three steps for MR-SAE. First, the available input data are used to pretrain the shared network. Then, all the multirate data are used to fine-tune the whole network. Finally, individual fine-tuning is further carried out for quality-specific subnetworks. The proposed MR-SAE method is used to build a unified soft sensor model for predicting both the 50% boiling point and cetane content of diesel oil. The results show that the performance of MR-SAE-based model is superior to SAE and deep belief networks. Moreover, the parameters and training time of MR-SAE model are less than the other two methods.

A variety of clean-energy heating systems were applied in China with multiple energy systems complementing each other according to local conditions. The purposes are pushing forward a supply-side structural reformation; coordinating the utilization of waste heat in high energy-consuming enterprises; realizing the energy cascade use; improving the ecological environment. This paper introduces a coupling energy station of 10 new clean-energy heating systems located in the central city district in Jinan. It could make full use of renewable energy and ensure high-quality heating in winter, increase energy utilization efficiency, replace fossil fuels, and reduce primary energy consumption. Furthermore, it can make up for the deficiency of a single system and reduce the emissions from fossil fuels. After comparing the heating parameters, the authors got the conclusions that the heating area of this new energy system is equal to the traditional coal system burning 9,508.032 tons of standard coal, which means in each heating season, about 7,233.33 × 107 m3 of waste gases, 45.64 tons of SO2 emissions, 69.586 tons of NOx emissions, and 139.52 tons of soot emissions will be avoided. So, this new energy heating system is safe, stable, energy-saving, environmentally friendly, and effective, which can completely replace the heating system of traditional coal-fired boilers.

Electric machines with high specific power are crucial for the more electric aircraft propulsion. However, the harsh operating environment such as low air pressure and high temperature will challenge the motor’s insulation system and cause partial discharge (PD), which threatens the reliability of the machine. This article proposes a PD-free stator winding design regarding an air-core high-frequency permanent magnet synchronous machine for aircraft propulsion. The experiment and simulation are performed under sinusoidal voltage due to the power quality requirement by the standard for aircraft. PD inception voltage (PDIV) with respect to air pressure and temperature was obtained experimentally for form-wound windings and was analyzed to derive the PD-free criteria. A 3-D stator model was built and used to simulate the electric field and temperature distributions with a finite-element method. In addition, the effect of voids and delamination on electric field distortion was also taken into consideration to allow for a safety margin and meet the PD-free requirement. To achieve high-specific-power requirement and PD-free requirement simultaneously, a multiobjective genetic algorithm was adopted to perform the optimization and obtain the Pareto fronts at various air pressures. Finally, the PDIV value of the designed winding was measured to verify the PD-free design. The main purpose is to combine the experimental results and the simulation tool to obtain an acceptable PD-free design in the meantime achieving high power density, which may provide helpful guidance on motor insulation design for the more electric aircraft propulsion.

Form‐wound windings in electric machines designed for electric aircraft propulsion face reliability challenges due to the severe operating environment, such as high temperature and low pressure. This study proposes a forewarning method for insulation condition monitoring of form‐wound windings based on partial discharge (PD) and deep learning neural network. Three PD features are extracted from the PD profile, which provides information about physics‐of‐failure and reflects the degree of insulation degradation. An algorithm fusion extracted from auto‐encoder and long short‐term recurrent neural network is proposed to synthesize one failure precursor from these three features and make multi‐time‐step prediction through historical data to provide forewarning. An electrical and thermal accelerated ageing test is performed on the form‐wound windings at 0.2 atm to simulate working environment of electric aircraft. The proposed method is validated on the accelerated ageing dataset and shows better prediction accuracy than some existing time‐series prediction methods, indicating the advantages of the proposed method. Moreover, an on‐line hardware setup using a deep learning processor is recommended to implement the forewarning method. The proposed approach has the potential to be widely applied to other insulation systems and contribute to work on condition monitoring.

Data-driven soft sensors have been widely used in industrial processes. Traditional soft sensors are mostly shallow networks, which cannot easily describe the complicated process data patterns. In this article, a new deep learning approach is proposed for soft sensors, which is based on the stacked enhanced auto-encoder (SEAE). The original stacked auto-encoder (SAE) learns the hierarchical features of the raw observed input data with unsupervised layerwise pretraining. In each layer, the features are learned from its low-level ones with an AE by minimizing the reconstruction error for them. This usually causes accumulation of information loss from the lowest layer to the highest layer. Hence, the learned features may not well represent the raw input data patterns. The new SEAE network is designed by adding the constraint of additional reconstruction for the raw input data at each layer. In this way, the learned features at each layer are a good representation for the raw input data. The effectiveness of the proposed SEAE-based soft sensor method is validated on an industrial sulfur recovery unit.

Stacked auto-encoder (SAE)-based deep learning has been introduced for fault classification in recent years, which has the potential to extract deep abstract features from the raw input data. However, SAE cannot ensure the relevance of deep features with the fault types due to its unsupervised self-reconstruction in the pretraining stage. To overcome this problem, a stacked supervised auto-encoder is proposed to pretrain the deep network and obtain deep fault-relevant features from raw input data. In each supervised auto-encoder, informative features are learned from the input data with the goal that they can largely distinguish different fault types. By stacking multiple supervised auto-encoders hierarchically, high-level fault-relevant features are gradually learned from raw input data, which can improve the classification accuracy of the classifiers. The proposed SSAE is tested on the Tennessee–Eastman (TE) benchmark process and a real industrial hydrocracking process. The results show the effectiveness and flexibility of SSAE.

BACKGROUND: Disease progression prediction based on neuroimaging biomarkers is vital in Alzheimer’s disease (AD) research. Convolutional neural networks (CNN) have been proved to be powerful for various computer vision research by refining reliable and high-level feature maps from image patches. However, a key challenge in applying CNN to neuroimaging research is the limited labeled samples with high dimensional features. Another challenge regards how to improve the prediction accuracy by joint analysis of multiple data sources (i.e., multiple time points or multiple biomarkers). OBJECTIVE: To address these two challenges, we propose a novel multi-task learning framework based on CNN. METHODS: First, we pre-trained CNN on the ImageNet dataset and transferred the knowledge from the pre-trained model to neuroimaging representation. We used this deep model as feature extractor to generate high-level feature maps of different tasks. Then a novel unsupervised learning method, termed Multi-task Stochastic Coordinate Coding (MSCC), was proposed for learning sparse features of multi-task feature maps by using shared and individual dictionaries. Finally, Lasso regression was performed on these multi-task sparse features to predict AD progression measured by the Mini Mental State Examination (MMSE) and the Alzheimer's Disease Assessment Scale cognitive subscale (ADAS-Cog). RESULTS: We applied this novel CNN-MSCC system on the Alzheimer’s disease Neuroimaging Initiative (ADNI) dataset to predict future MMSE/ADAS-Cog scales. We found our method achieved superior performances compared with seven other methods. CONCLUSION: Our work may add new insights into data augmentation and multi-task deep model research and facilitate the adoption of deep models in neuroimaging research.

Deep learning is a recently developed feature representation technique for data with complicated structures, which has great potential for soft sensing of industrial processes. However, most deep networks mainly focus on hierarchical feature learning for the raw observed input data. For soft sensor applications, it is important to reduce irrelevant information and extract quality-relevant features from the raw input data for quality prediction. To deal with this problem, a novel deep learning network is proposed for quality-relevant feature representation in this article, which is based on stacked quality-driven autoencoder (SQAE). First, a quality-driven autoencoder (QAE) is designed by exploiting the quality data to guide feature extraction with the constraint that the potential features should largely reconstruct the input layer data and the quality data at the output layer. In this way, quality-relevant features can be captured by QAE. Then, by stacking multiple QAEs to construct the deep SQAE network, SQAE can gradually reduce irrelevant features and learn hierarchical quality-relevant features. Finally, the high-level quality-relevant features can be directly applied for soft sensing of the quality variables. The effectiveness and flexibility of the proposed deep learning model are validated on an industrial debutanizer column process.

The number of stages is a basic structural parameter for the heat integrated air separation column (HIASC) to improve its energy-saving potential. This paper, therefore, aims at the optimization of...

The selection of an appropriate descriptive system and modeling framework to capture system dynamics and support process control applications is a fundamental problem in the operation of industrial processes. In this study, to account for the highly complex dynamics of industrial process and additional requirements imposed by smart and optimal manufacturing systems, an extended state space descriptive system, named comprehensive state space, is first designed. Then, based on the descriptive system, a hybrid first principles/machine learning modeling framework is proposed. The hybrid model is formulated as a combination of a nominal term and a deviation term. The nominal term covers the underlying physicochemical principles. The deviation term handles the effects of high-dimensional influence factors using regression of low-dimensional deep process features. To handle the multimodal and time-varying properties of process dynamics, the comprehensive state space is divided into subspaces indicating different operating conditions. The model parameters are identified and trained for each operating condition to form the sub-models. Then the system dynamics are formulated as a weighted sum of sub-models, with the weights being the probabilities that the current operating point belongs to different operating conditions. The weights update with the movement of the operating point in the comprehensive state space. Moreover, the descriptive system provides a platform for visualization, and can act as a digital twin of the physical process. A case study illustrates the feasibility and performance of the proposed descriptive system.

A distillation curve is an essential property for petroleum. Its features are beneficial for the modeling and optimization of oil-refining processes. To capture these features with a small number o...

In industrial sintering processes, it is very important to monitor and control key quality indicators, which are often difficult to measure online. Soft sensor technology is a good solution for online prediction of quality indicators. Nowadays, deep learning is widely used in soft sensors due to its powerful ability in processing nonlinear data. In this paper, a supervised deep belief network (SDBN) is proposed by introducing quality variable into the input variables at each restricted Boltzmann machine to extract quality-related features for soft sensor. With case study on an actual industrial sintering process, SDBN shows much better prediction performance than the original deep belief network and stacked autoencoder.

Process complexities are characterized by strong nonlinearities, dynamics and uncertainties. Modeling such a complex process requires a flexible model with deep layers describing the corresponding strong nonlinear dynamic behavior. The proposed model is constructed by deep neural networks to represent the process of state transition and observation generation, both of which together constitute a stochastic nonlinear state space model. This model is evolved from the variational auto-encoder learned by the stochastic expectation-maximization algorithm. To solve the complexity of posteriors for dynamic processes, the posterior distributions with respect to state variables are constructed by a forward-backward recurrent neural network. One example is given to validate that the proposed method outperforms the comparative methods in modeling complex nonlinearities.

Amyloid-β (Aβ) plaques and tau protein tangles in the brain are now widely recognized as the defining hallmarks of Alzheimer’s disease (AD), followed by structural atrophy detectable on brain magnetic resonance imaging (MRI) scans. One of the particular neurodegenerative regions is the hippocampus to which the influence of Aβ/tau on has been one of the research focuses in the AD pathophysiological progress. This work proposes a novel framework, Federated Morphometry Feature Selection (FMFS) model, to examine subtle aspects of hippocampal morphometry that are associated with Aβ/tau burden in the brain, measured using positron emission tomography (PET). FMFS is comprised of hippocampal surface-based feature calculation, patch-based feature selection, federated group LASSO regression, federated screening rule-based stability selection, and region of interest (ROI) identification. FMFS was tested on two Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohorts to understand hippocampal alterations that relate to Aβ/tau depositions. Each cohort included pairs of MRI and PET for AD, mild cognitive impairment (MCI), and cognitively unimpaired (CU) subjects. Experimental results demonstrated that FMFS achieves an 89× speedup compared to other published state-of-the-art methods under five independent hypothetical institutions. In addition, the subiculum and cornu ammonis 1 (CA1 subfield) were identified as hippocampal subregions where atrophy is strongly associated with abnormal Aβ/tau. As potential biomarkers for Aβ/tau pathology, the features from the identified ROIs had greater power for predicting cognitive assessment and for survival analysis than five other imaging biomarkers. All the results indicate that FMFS is an efficient and effective tool to reveal associations between Aβ/tau burden and hippocampal morphometry.

Background Whether adjuvant chemotherapy (AC) after concurrent chemoradiotherapy (CCRT) could provide benefit to esophageal squamous cell carcinoma (ESCC) patients is controversial. Therefore, we decided to investigate the potential benefit of AC after CCRT for ESCC and to identify biomarkers predictive of a clinical benefit. Methods We retrospectively analysed the clinical data of ESCC patients with clinical stage II–IVa who underwent CCRT. Then, we compared patients who received CCRT and AC (CCRT + AC group) with those who received CCRT alone (CCRT group). Propensity score analysis, subgroup analysis and an additional Cox regression model were conducted to analyse the predictive factors. The overall survival (OS) and progression-free survival (PFS) rates were taken as the endpoints. Results From January 2013 to December 2017, 244 patients were recruited (n = 131 for CCRT + AC; n = 113 for CCRT alone) for the analysis. After propensity score matching was performed (1:1 and 99 patients for each group) with consideration of the basic clinical characteristics, no significant differences were found in OS (HR = 1.024; 95% CI 0.737–1.423; P = 0.886) or PFS (HR = 0.809; 95% CI 0.582–1.126; P = 0.197) between the two groups. The good short-term response subgroup showed a better PFS and favoured CCRT + AC treatment (HR = 0.542; 95% CI 0.336–0.876; P = 0.008), the independent predictive role of which was confirmed in additional multivariate Cox regression analysis. Conclusions Although AC did not significantly improve PFS and OS for all ESCC patients after CCRT, the short-term response to CCRT might help identify a subgroup that will benefit, which needs further prospective research to confirm.

Biomarker assisted preclinical/early detection and intervention in Alzheimer’s disease (AD) may be the key to therapeutic breakthroughs. One of the presymptomatic hallmarks of AD is the accumulation of beta-amyloid (Aβ) plaques in the human brain. However, current methods to detect Aβ pathology are either invasive (lumbar puncture) or quite costly and not widely available (amyloid PET). Our prior studies show that magnetic resonance imaging (MRI)-based hippocampal multivariate morphometry statistics (MMS) are an effective neurodegenerative biomarker for preclinical AD. Here we attempt to use MRI-MMS to make inferences regarding brain Aβ burden at the individual subject level. As MMS data has a larger dimension than the sample size, we propose a sparse coding algorithm, Patch Analysis-based Surface Correntropy-induced Sparse-coding and Max-Pooling (PASCS-MP), to generate a low-dimensional representation of hippocampal morphometry for each individual subject. Then we apply these individual representations and a binary random forest classifier to predict brain Aβ positivity for each person. We test our method in two independent cohorts, 841 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and 260 subjects from the Open Access Series of Imaging Studies (OASIS). Experimental results suggest that our proposed PASCS-MP method and MMS can discriminate Aβ positivity in people with mild cognitive impairment (MCI) [Accuracy (ACC) = 0.89 (ADNI)] and in cognitively unimpaired (CU) individuals [ACC = 0.79 (ADNI) and ACC = 0.81 (OASIS)]. These results compare favorably relative to measures derived from traditional algorithms, including hippocampal volume and surface area, shape measures based on spherical harmonics (SPHARM) and our prior Patch Analysis-based Surface Sparse-coding and Max-Pooling (PASS-MP) methods.

ObjectiveThis study aimed to evaluate the feasibility and efficacy of ultrasound-guided percutaneous microwave ablation (MWA) in the treatment of low-risk papillary thyroid microcarcinoma (PTMC), and to observe the histopathological changes after MWA.MethodsMWA was performed under ultrasound guidance for 73 unifocal PTMC patients without clinically cervical or distant metastasis. The target ablation zone exceeded the tumor edge (judged by contrast-enhanced US) to avoid marginal residue and recurrence. Ultrasound evaluation was performed at 1 day, 1, 3, 6, 12 and 24 months after treatment, and thyroid function evaluation at the first 6 months. Repeated fine needle aspiration cytology or core needle biopsy pathology was performed at 3 or 6 months after MWA to evaluate residual tumors. Any adverse event associated with MWA was evaluated.ResultsThe follow-up after MWA lasted 6 (6, 12) months. Tumor volume decreased significantly from 0.06 mm3 (0.04, 0.11 mm3) to 0.03 mm3 (0.00, 0.06 mm3) at 12 months after MWA (P< 0.001), with a median volume reduction ratio of 80.28% (-7.43, 100%) and 16 cases (21.92%) presenting complete remission. The largest diameter, volume and ablation energy were found to be different in patients with and without complete remission 12 months after MWA. On histopathological examinations, no atypical or malignant follicular cells were identified after thermal ablation. The most common pathological characteristics were fibroblastic proliferation (34/39, 87.18%) and chronic inflammation (32/39, 82.05%), followed by infarction (21/39, 53.85%). Five patients were transferred to thyroidectomy and 4 of them were confirmed with local recurrence and/or lymph node metastasis. Serum thyrotropin decreased transiently after MWA (P< 0.01) but normalized thereafter. No serious and permanent complications were reported.ConclusionsMWA is a safe and effective treatment for low-risk PTMC. Fibroblastic proliferation and chronic inflammation are the most common pathological changes after MWA of PTMC.

The anatomical landmarking on statistical shape models is widely used in structural and morphometric analyses. The current study focuses on leveraging geometric features to realize an automatic and reliable landmarking. The existing implementations usually rely on classical geometric features and data-driven learning methods. However, such designs often have limitations to specific shape types. Additionally, calculating the features as a standalone step increases the computational cost. In this paper, we propose a convolutional Bayesian model for anatomical landmarking on multi-dimensional shapes. The main idea is to embed the convolutional filtering in a stationary kernel so that the geometric features are efficiently captured and implicitly encoded into the prior knowledge of a Gaussian process. In this way, the posterior inference is geometrically meaningful without entangling with extra features. By using a Gaussian process regression framework and the active learning strategy, our method is flexible and efficient in extracting arbitrary numbers of landmarks. We demonstrate extensive applications on various publicly available datasets, including one brain imaging cohort and three skeletal anatomy datasets. Both the visual and numerical evaluations verify the effectiveness of our method in extracting significant landmarks.

With the improvement of computer technology and the widespread popularization of the Internet in China, e-commerce has greatly changed people's way of life, become a major promoting industry for China's economic development, and also increased the income of rural and urban residents in China to a certain extent. Since 2013, China's e-commerce development has reached a certain scale, so this article based on the system of literature, using data from 2013–2019, analyzes the electronic commerce to the influence of the Chinese people's income, found that in general the development of e-commerce of our country resident income and great role in promoting, and the impact on rural income is greater than the effect on the income in cities and towns. Finally, combined with the results of empirical analysis, put forward the corresponding countermeasures and suggestions.

Three-level T-type neutral-point-clamped converters (3L-TNPC) are widely used in motor drive and PV applications because of the higher efficiency, improved output THD, and lower common-mode noise. But such topology suffers from high stray inductance because of the complicated power loop structure. This paper proposes a SiC-MOSFET based 3L-TNPC power module with a hybrid packaging method. The module has a printed circuit board (PCB) directly soldered on top of direct bounding copper (DBC), with bare dies soldered on DBC and connecting to PCB by bonding wires. Such a structure allows the power loop on both DBC and PCB. This reduces stray inductance through enhanced mutual-inductance cancellation. The introduced PCB can provide extra flexibility in gate-loop optimization, enabling synchronous gate drive between parallel dies. The heat from dies can also be directly dissipated through DBC and the proposed direct-soldered coldplate. A 1.2 kV, 400 A SiC 3L-TNPC module is fabricated with measured loop inductance of 2.5 nH. Faster driving speed by using 0.33Ω gate resistance is enabled to reduce switching loss. Eventually, double-pulse tests and continuous tests are carried out to evaluate the module.

Partial discharge (PD) issue of high voltage power module packaging insulation is increasingly critical due to the development of high voltage wide bandgap power modules towards high power density, which leads to smaller packages and higher operating voltages. However, existing standards for high voltage power modules focus on PD measurement under the sinusoidal waveform rather than actual DC or square waveform that poses actual electric stress on packaging insulation, and thus cannot fully represent the PD phenomenon in power electronics systems. To bridge this gap, this paper introduces a PD testing platform trimmed for high voltage power electronics systems, including a high-voltage (10 kV) square wave generator using supercascode structure with off-shelf SiC devices, and a contactless PD monitor using Super High Frequency (SHF) signatures from the PD event. The challenges in the design of this platform include the development of a PD-free high voltage square wave generator, and SHF PD signature separation from EMI noises induced by high dV/dt switching actions. Experiment validation is included in the last section of this paper.

Structural and anatomical analyses of magnetic resonance imaging (MRI) data often require a reconstruction of the three-dimensional anatomy to a statistical shape model. Our prior work demonstrated the usefulness of tetrahedral spectral features for grey matter morphometry. However, most of the current methods provide a large number of descriptive shape features, but lack an unsupervised scheme to automatically extract a concise set of features with clear biological interpretations and that also carries strong statistical power. Here we introduce a new tetrahedral spectral feature-based Bayesian manifold learning framework for effective statistical analysis of grey matter morphology. We start by solving the technical issue of generating tetrahedral meshes which preserve the details of the grey matter geometry. We then derive explicit weak-form tetrahedral discretizations of the Hamiltonian operator (HO) and the Laplace-Beltrami operator (LBO). Next, the Schrodinger's equation is solved for constructing the scale-invariant wave kernel signature (SIWKS) as the shape descriptor. By solving the heat equation and utilizing the SIWKS, we design a morphometric Gaussian process (M-GP) regression framework and an active learning strategy to select landmarks as concrete shape descriptors. We evaluate the proposed system on publicly available data from the Alzheimers Disease Neuroimaging Initiative (ADNI), using subjects structural MRI covering the range from cognitively unimpaired (CU) to full blown Alzheimer's disease (AD). Our analyses suggest that the SIWKS and M-GP compare favorably with seven other baseline algorithms to obtain grey matter morphometry-based diagnoses. Our work may inspire more tetrahedral spectral feature-based Bayesian learning research in medical image analysis.

Retinotopic map, the mapping between visual inputs on the retina and neuronal responses on cortical surface, is one of the central topics in vision science. Typically, human retinotopic maps are constructed by analyzing functional magnetic resonance responses to designed visual stimuli on cortical surface. Although it is widely used in visual neuroscience, retinotopic maps are limited by measurement noise and resolution. One promising approach to improve the quality of retinotopic maps is to register individual subject’s retinotopic maps to a retinotopic template or atlas. However, none of the existing retinotopic registration methods has explicitly quantified the diffeomorphic condition, that is, retinotopic maps can be aligned by stretching/compressing but without tearing up. Here, we developed Diffeomorphic Registration for Retinotopic Maps (DRRM) to simultaneously align retinotopic maps in multiple visual regions under the diffeomorphic condition. Specifically, we used the Beltrami coefficient to model the diffeomorphic condition and performed surface registration based on retinotopic coordinates. The overall framework is simple and elegant and preserves topological condition defined in the atlas. We further developed a unique performance evaluation protocol and compared the performance of the new method with several existing image intensity-based registration methods on both synthetic and real datasets. The results showed that DRRM is superior to the existing methods in achieving diffeomorphic mappings in synthetic and empirical data from 3T and 7T magnets. DRRM may improve the interpretation of low-quality retinotopic maps and facilitate adoption of retinotopic maps in clinical settings.

Effective systemic therapy is highly desired for the treatment of hepatocellular carcinoma (HCC). In this study, a combination of nanoparticles of poly(L-glutamic acid)-graft-methoxy poly(ethylene glycol)/combretastatin A4 sodium salt (CA4-NPs) plus sorafenib is developed for the cooperative systemic treatment of HCC. The CA4-NPs leads to the disruption of established tumor blood vessels and extensive tumor necrosis, however, inducing increased expression of VEGF-A and angiogenesis. Sorafenib reduces the VEGF-A induced angiogenesis and further inhibits tumor proliferation, cooperating with the CA4-NPs. A significant decrease in tumor volume and prolonged survival time are observed in the combination group of CA4-NPs plus sorafenib compared with CA4-NPs or sorafenib monotherapy in subcutaneous and orthotopic H22 hepatic tumor models. Seventy-one percent of the mice are alive without residual tumor at 96 days post tumor inoculation for the subcutaneous models treated with CA4-NPs 30 or 35 mg·kg−1 plus sorafenib 30 mg·kg−1. Our findings suggest that co-administration of sorafenib and CA4-NPs possesses significant antitumor efficacy for HCC treatment. Statement of Significance Effective systemic therapy is highly desired for the treatment of hepatocellular carcinoma (HCC). Herein, we demonstrate that a combination of nanoparticles of poly(L-glutamic acid)-graft-methoxy poly(ethylene glycol)/combretastatin A4 sodium salt (CA4-NPs) plus sorafenib is a promising synergistic approach for systemic treatment of HCC. The CA4-NPs leads to the disruption of established tumor blood vessels and extensive tumor necrosis, however, inducing increased expression of VEGF-A and angiogenesis. Sorafenib reduces the VEGF-A induced angiogenesis and further inhibits tumor proliferation, cooperating with the CA4-NPs.

Iron ore sintering is a major source of gaseous and particulate pollutants emission in iron smelt plant. The aim of present study is to characterize the volatile organic compounds (VOCs) emission profiles from iron ore sintering process. Both sinter pot test and sinter simulation experiment were conducted and compared. Out results showed that sinter process produced large quantity of VOCs together with NOx and SO2. VOCs and NO were produced simultaneously in sinter pot test from 3 to 24 min after ignition, flowed by SO2 production from 15 min to the end of sintering. Total VOCs (TVOC) concentration in sinter flue gas was affected by the coal and coke ratio in sinter raw material. The maximum TVOC concentration was 34.5 ppm when using 100% coal as fuel. Sinter simulation experiments found that the number of VOCs species and their concentrations were found by sinter temperature. The largest VOCs species varieties were obtained at 500 °C. Benzene, toluene, xylene and ethylbenzene were major VOCs in sinter flue gas based on the results from both simulation test and sinter pot. It thus demonstrated that in addition to NOx, SO2 and metal oxide particles, sinter flue gas also contained significant amount of VOCs whose environmental impact cannot be ignored. Based on our work, it is timely needed to establish a new VOC emission standard for sinter flue gas and develop advanced techniques to simultaneously eliminate multi-pollutants in iron ore sinter process.

Background and purpose: Adaptive radiation planning for pancreatic adenocarcinoma (PA) relies on accurate treatment response assessment, while traditional response evaluation criteria inefficiently characterize tumors with complex morphological features or intrinsically low metabolism. To better assess treatment response of PA, we quantify and compare regional morphological and metabolic features of the 3D pre- and post-radiation therapy (RT) tumor models. Materials and methods: Thirty-one PA patients with pre and post-RT Positron emission tomography/computed tomography (PET/CT) scans were evaluated. 3D meshes of pre- and post-RT tumors were generated and registered to establish vertex-wise correspondence. To assess tumor response, Mahalanobis distances (Mdist|Fusion) between pre- and post-RT tumor surfaces with anatomic and metabolic fused vectors were calculated for each patient. Mdist|Fusion was evaluated by overall survival (OS) prediction and survival risk classification. As a comparison, the same analyses were conducted on traditional imaging/physiological predictors, and distances measurements based on metabolic and morphological features only. Results: Among all the imaging/physiological parameters, Mdist|Fusion was shown to be the best predictor of OS (HR = 0.52, p = 0.008), while other parameters failed to reach significance. Moreover, Mdist|Fusion outperformed traditional morphologic and metabolic measurements in patient risk stratification, either alone (HR = 11.51, p

In recent years, deep learning like stacked autoencoder (SAE) has been widely applied to soft sensor modeling for industrial process quality prediction. However, the original SAE based models cannot extract the quality-relevant features due to its unsupervised feature learning, which may even cause negative impact on the final prediction. To cope with this problem, a novel stacked attention-based autoencoder (SAAE) is proposed. By introducing quality information and attention mechanism to the pre-training process, the quality-relevant features can be obtained and simultaneously their importance can be distinguished. Furthermore, to fully utilize these features, feature fusion strategy is integrated with the SAAE method. Finally, a hydrocracking industrial process is used to validate the effectiveness of the proposed method.

An effective presymptomatic diagnosis and treatment of Alzheimer’s disease (AD) would have enormous public health benefits. Sparse coding (SC) has shown strong potential for longitudinal brain image analysis in preclinical AD research. However, the traditional SC computation is time-consuming and does not explore the feature correlations that are consistent over the time. In addition, longitudinal brain image cohorts usually contain incomplete image data and clinical labels. To address these challenges, we propose a novel two-stage Multi-Resemblance Multi-Target Low-Rank Coding (MMLC) method, which encourages that sparse codes of neighboring longitudinal time points are resemblant to each other, favors sparse code low-rankness to reduce the computational cost and is resilient to both source and target data incompleteness. In stage one, we propose an online multi-resemblant low-rank SC method to utilize the common and task-specific dictionaries in different time points to immune to incomplete source data and capture the longitudinal correlation. In stage two, supported by a rigorous theoretical analysis, we develop a multi-target learning method to address the missing clinical label issue. To solve such a multi-task low-rank sparse optimization problem, we propose multi-task stochastic coordinate coding with a sequence of closed-form update steps which reduces the computational costs guaranteed by a theoretical convergence proof. We apply MMLC on a publicly available neuroimaging cohort to predict two clinical measures and compare it with six other methods. Our experimental results show our proposed method achieves superior results on both computational efficiency and predictive accuracy and has great potential to assist the AD prevention.

Uncovering the complex network of the brain is of great interest to the field of neuroimaging. Mining from these rich datasets, scientists try to unveil the fundamental biological mechanisms in the human brain. However, neuroimaging data collected for constructing brain networks is generally costly, and thus extracting useful information from a limited sample size of brain networks is demanding. Currently, there are two common trends in neuroimaging data collection that could be exploited to gain more information: 1) multimodal data, and 2) longitudinal data. It has been shown that these two types of data provide complementary information. Nonetheless, it is challenging to learn brain network representations that can simultaneously capture network properties from multimodal as well as longitudinal datasets. Here we propose a general fusion framework for multi-source learning of brain networks - multimodal brain network fusion with longitudinal coupling (MMLC). In our framework, three layers of information are considered, including cross-sectional similarity, multimodal coupling, and longitudinal consistency. Specifically, we jointly factorize multimodal networks and construct a rotation-based constraint to couple network variance across time. We also adopt the consensus factorization as the group consistent pattern. Using two publicly available brain imaging datasets, we demonstrate that MMLC may better predict psychometric scores than some other state-of-the-art brain network representation learning algorithms. Additionally, the discovered significant brain regions are synergistic with previous literature. Our new approach may boost statistical power and sheds new light on neuroimaging network biomarkers for future psychometric prediction research by integrating longitudinal and multimodal neuroimaging data.

Purpose/objective(s) Cranio-spinal irradiation (CSI) using advanced modality (IMRT, VMAT, protons) requires contouring of the thecal sac, a tedious and time-consuming task. Additionally, sparing of the vertebrae to improve growth and bone marrow reserve necessitates vertebrae contours. The COG and other investigators have great interest in studying growth effects from vertebrae-sparing technique; this effort requires reporting of dose to the thecal sac and the vertebrae. The presence of bone-soft tissue interface and time required for manual contours make this an ideal use of deep-learning auto-contouring. Materials/methods Two convolutional neural network (CNN) models were trained and validated, one with 65 patients to auto contour thecal sac, kidneys and lungs, and the other with 55 to draw vertebrae. The training set consisted of mostly pediatric and young adult patients. The models were tested on ten new patients at the age of 5, 6, 8, 9, 11, 13, 16, 20, 31 and 47. Quantitatively, auto contours were evaluated against clinical manual contours using dice similarity coefficient (DSC) and mean distance to agreement (MDA). A qualitative scoring system was used by a dosimetrist experienced in CSI contouring: 0 (accepted upright), 1 (accepted, minor editing in 1 min), 2 (accepted, moderate editing in 3 min), 3 (accepted, significant editing with efficiency gain) and 4 (rejected, gross error or no efficiency gain). The time for editing each auto contour for each patient was recorded and compared to the time required for manual contouring measured using time/slice and the number of slices for thecal sac, vertebrae and each kidney, and as whole organ for each lung. Results As shown in Table 1, on average, the DSC was above 0.90 and MDA below 1.2 mm, both well within the tolerance recommended by AAPM TG 132. The mean score was below 1.0 for all volumes. Editing thecal sac and vertebrae auto contours only required 1 min or less for all patients. The models reduced contouring time by 84 and 40 min for thecal sac and vertebrae, respectively. Conclusion The deep-learning models provided high-quality auto contours, directly acceptable for lungs and requiring only minor editing in less than one min for thecal sac, vertebrae and kidneys. Given the labor-intensive nature of manually contouring thecal sac and vertebrae, the high accuracy of the auto contours resulted in substantial reduction of contouring time and thus faster turnaround for CSI treatment. The models can provide a uniform starting point for CSI contouring in our field, which is particularly beneficial to clinics with less experiences.

PURPOSE/OBJECTIVE(S) To evaluate a convolutional neural network auto-contouring (AC) model created for treatment planning in patients receiving radiotherapy for localized prostate cancer after insertion of a radiopaque rectal spacer hydrogel. MATERIALS/METHODS The deep learning model, trained by 125 patients, auto contours target volumes (prostate and proximal seminal vesicles), OARs (bladder, rectum, femoral heads and penile bulb) and radiopaque rectal spacers. ACs were evaluated against MD manual contours (MCs) submitted for treatment planning. ACs were not available for MD review while creating MCs. Individual volumes as well as composite volumes (overall performance) were qualitatively evaluated by a radiation oncologist using a 1 (minor discrepancy, little to no dose-volume impact), 2 (moderate discrepancy or editable with substantial efficiency gain), 3 (significant discrepancy or editable with meaningful efficiency gain), and 4 (rejected due to gross error or editable without efficiency gain) scoring scale. Quantitative evaluation of geometric differences was performed with mean distance to agreement (MDA) and Dice Similarity Coefficient (DSC) for each volume. RESULTS A total of 68 cases were evaluated by quantitative and qualitative metrics (Table 1). MCs were performed by 4 experienced MDs. Average composite score of 2.22 (SD 0.73) indicates substantial to meaningful efficiency gain in contouring process. Composite scores for 100% of cases evaluated indicate a meaningful efficiency gain in the contouring process, with 60% of composite scores indicating only minor to moderate discrepancy between contours. Prostate and SV AC average scores were above 2. Mean prostate MDA and DSC were 1.71 and 0.85, respectively, both within tolerance range recommended by AAPM TG 132. CONCLUSION The model can accurately auto contour target volumes, OARs and spacer with meaningful to substantial efficiency gain, with a majority of target volumes qualitatively evaluated by a physician as having only mild to moderate discrepancies. Quantitative analysis shows contours for most OARs have only minor geometric differences unlikely to have significant dose-volume impacts. Further analysis with comparison of treatment plans generated from ACs and MCs is warranted.