Single-channel Model Research Articles

Deep learning model for diagnosis of thyroid nodules with size less than 1 cm: A multicenter, retrospective study

ObjectiveTo develop a ultrasound images based dual-channel deep learning model to achieve accurate early diagnosis of thyroid nodules less than 1 cm. MethodsA dual-channel deep learning model called thyroid nodule transformer network (TNT-Net) was proposed. The model has two input channels for transverse and longitudinal ultrasound images of thyroid nodules, respectively. A total of 9649 nodules from 8455 patients across five hospitals were retrospectively collected. The data were divided into a training set (8453 nodules, 7369 patients), an internal test set (565 nodules, 512 patients), and an external test set (631 nodules, 574 patients). ResultsTNT-Net achieved an area under the curve (AUC) of 0.953 (95 % confidence interval (CI): 0.934, 0.969) on the internal test set and 0.941 (95 % CI: 0.921, 0.957) on the external test set, significantly outperforming traditional deep convolutional neural network models and single-channel swin transformer model, whose AUCs ranged from 0.800 (95 % CI: 0.759, 0.837) to 0.856 (95 % CI: 0.819, 0.881). Furthermore, feature heatmap visualization showed that TNT-Net could extract richer and more energetic malignant nodule patterns. ConclusionThe proposed TNT-Net model significantly improved the recognition capability for thyroid nodules with size less than 1 cm. This model has the potential to reduce overdiagnosis and overtreatment of such nodules, providing essential support for precise management of thyroid nodules while complementing fine-needle aspiration biopsy.

Seizure Onset Zone Detection Based on Convolutional Neural Networks and EEG Signals

Background: The localization of seizure onset zones (SOZs) is a critical step before the surgical treatment of epilepsy. Methods and Results: In this paper, we propose an SOZ detection method based on convolutional neural networks and EEG signals. This method aims to locate SOZs through the seizure status of each channel in multi-channel EEG signals. First, we preprocess the data with filtering, segmentation, resampling, and standardization to ensure their quality and consistency. Then, the single-channel UCI epilepsy seizure recognition dataset is used to train and test the convolutional neural network (CNN) model, achieving an accuracy of 98.70%, a sensitivity of 97.53%, and a specificity of 98.98%. Next, the multi-channel clinical EEG dataset collected by a hospital is divided into 21 single-channel site datasets and input into the model for detection, and then the seizure results of 21 sites per second are obtained. Finally, the seizure sites are visualized through the international 10–20 system electrode distribution map, diagrams of the change process of the seizure sites during seizures are drawn, and patients’ SOZs are located. Conclusions: Our proposed method well classifies seizure and non-seizure data and successfully locates SOZs by detecting the seizure results of 21 sites through a single-channel model. This study can effectively assist doctors in locating the SOZs of patients and provide help for the diagnosis and treatment of epilepsy.

Quantum Evolution and Resonance in a Simple Single-Channel Model

Quantum Evolution and Resonance in a Simple Single-Channel Model

Holistic Hydraulic Simulation for Pebble Bed Using Porous Media Approach

The porous media approach is regarded as an appropriate methodology for hydraulic simulations of complex pebble beds in fusion reactors. In order to determine the parameters (permeability α and inertial loss coefficient C) of the porous media accurately, two methods are proposed: the correction method and the fitting method. In this paper, a single-channel model with sequentially packed pebbles is constructed in order to obtain the pressure drop gradient against superficial velocities. Two methods, the correction method and fitting method, are employed to determine the permeability and inertial loss coefficient, and the results are evaluated with comparisons. Based on the results, both the correction method and fitting method are deemed feasible for the parameter determinations. In consideration of the consumption of resources and time for simulation, the fitting method is recommended during the preliminary design phase to shorten the duration of design, while the correction method is suggested to obtain precise results when the design is accomplished. Both of the methods would be evaluated with the data obtained from experiments in the future.

Solute transport along a single channel with radial diffusion in the porous rock matrix: A simple analytical solution and the implementation of time domain random walk algorithm

Numerous field observations show that the channels in one fracture are narrow and the solute penetration depth might be larger than the width. For this case, the diffusion from a channel into the matrix is more realistic to be modeled as radial diffusion than one-dimensional. In present work, the single channel model with radial diffusion is revisited and a simple and robust analytical solution is developed. This solution takes a convolution form of two functions, in which different transport mechanisms are accounted for. The statistical interpretations of the two functions and the analytical solution aid to develop a simple Time Domain Random Walk (TDRW) algorithm and an extension is made to improve its accuracy, efficiency and applicability. To demonstrate the accuracy and efficacy of the extended algorithm, three groups of simulations are performed and it is found that the results of all approaches are identical. The TDRW algorithm, having the same performance as that of inverse Laplace transform solution, is superior to Gaussian quadrature method in computational time. However, due to Monte Carlo nature of the algorithm, the computational burden of TDRW algorithm is dependent on the number of particles applied, which also influences the calculation accuracy. Therefore, a trade-off between computational burden and calculation accuracy should always be made, once the TDRW algorithm is used.

Deep-learning-based motion correction using multichannel MRI data: a study using simulated artifacts in the fastMRI dataset.

Deep learning presents a generalizable solution for motion correction requiring no pulse sequence modifications or additional hardware, but previous networks have all been applied to coil-combined data. Multichannel MRI data provide a degree of spatial encoding that may be useful for motion correction. We hypothesize that incorporating deep learning for motion correction prior to coil combination will improve results. A conditional generative adversarial network was trained using simulated rigid motion artifacts in brain images acquired at multiple sites with multiple contrasts (not limited to healthy subjects). We compared the performance of deep-learning-based motion correction on individual channel images (single-channel model) with that performed after coil combination (channel-combined model). We also investigate simultaneous motion correction of all channel data from an image volume (multichannel model). The single-channel model significantly (p < 0.0001) improved mean absolute error, with an average 50.9% improvement compared with the uncorrected images. This was significantly (p < 0.0001) better than the 36.3% improvement achieved by the channel-combined model (conventional approach). The multichannel model provided no significant improvement in quantitative measures of image quality compared with the uncorrected images. Results were independent of the presence of pathology, and generalizable to a new center unseen during training. Performing motion correction on single-channel images prior to coil combination provided an improvement in performance compared with conventional deep-learning-based motion correction. Improved deep learning methods for retrospective correction of motion-affected MR images could reduce the need for repeat scans if applied in a clinical setting.

Scientific equipment for the “Sun-Terahertz” space experiment: study of the temperature effect of the Golay cell

The future space experiment «Sun-Terahertz» is aimed at studying the Sun in the unexplored terahertz range, obtaining new data on the terahertz radiation of the Sun, solar active regions and solar flares. The scientific equipment being developed is a set of eight detectors sensitive to radiation of various frequencies in the range 0.4–12.0 THz. The Golay cell optoacoustic receiver was used as a sensitive element of the detectors. This paper examines the problem of changing the sensitivity of receivers of this type when their own temperature changes (temperature effect). Using a test bench based on a single-channel model of scientific equipment and a black body simulator, the linear section of the temperature dependence of the receiver was determined. A method for compensating the temperature effect of optoacoustic converters Golay cell has been developed and the results of its testing on a test bench are presented. This technique will be useful in the development of scientific equipment based on Golay cells, the operation of which is expected under conditions of temperature changes in the surrounding space.

Economic and environmental evaluation of different collection models for spent power batteries

The efficient collection of spent batteries is important for the recycling of power batteries and the sustainable advancement of electric vehicles (EVs). This study introduces and evaluates 14 power battery collection models, encompassing single-channel, dual-channel, and triple-channel scenarios, which encapsulate the dynamics of competition, cooperation, and co-opetition among key stakeholders. Employing Stackelberg game theory and simulation analysis, this study assesses the economic and environmental benefits of these models, with particular attention to the impact of consumer environmental awareness and competition intensity. Findings indicate that the collaborative collection model involving battery manufacturers, EV companies, and third-party recyclers (Model MVTP) delivers superior economic outcomes, with benefits ranging from 20.3 to 20.31 billion RMB. Meanwhile, the competitive three-party collection model (Model M-V-TP) stands out for its environmental benefits, achieving collection rates between 14.95 % and 25.65 %. Intriguingly, while competition intensity does not influence the outcomes of single-channel models, it diminishes both economic and environmental benefits in dual- and triple-channel scenarios. Moreover, enhanced consumer environmental awareness consistently elevates the economic benefits of all models, though its effect on environmental benefits varies—enhancing collection rates in single- and dual-channel models but detrimentally impacting the triple-channel model.

A GNU radio-based design for GPS software simulation source

GPS (Global Positioning System) plays a crucial role in modern navigation and positioning applications. Traditional research and development of GPS signals require expensive hardware equipment, limiting the flexibility and scalability of the system. In this paper, an open-source software called GNU Radio is utilized to construct a GPS signal simulation system, leveraging its rich modules and signal processing libraries. The GPS baseband signals are generated using signal processing modules provided by GNU Radio. Appropriate noise and interference are added to simulate the signal characteristics in real-world environments. Both single-channel and multi-channel signal models are presented. By comparing the generated GPS signals with those from hardware-based simulators, the results demonstrate that the GPS software simulator based on GNU Radio can produce high-quality GPS signals with excellent flexibility and adjustability.

A prior-information-based automatic segmentation method for the clinical target volume in adaptive radiotherapy of cervical cancer.

Adaptive planning to accommodate anatomic changes during treatment often requires repeated segmentation. In this study, prior patient-specific data was integrateda into a registration-guided multi-channel multi-path (Rg-MCMP) segmentation framework to improve the accuracy of repeated clinical target volume (CTV) segmentation. This study was based on CT image datasets for a total of 90 cervical cancer patients who received two courses of radiotherapy. A total of 15 patients were selected randomly as the test set. In the Rg-MCMP segmentation framework, the first-course CT images (CT1) were registered to second-course CT images (CT2) to yield aligned CT images (aCT1), and the CTV in the first course (CTV1) was propagated to yield aligned CTV contours (aCTV1). Then, aCT1, aCTV1, and CT2 were combined as the inputs for 3D U-Net consisting of a channel-based multi-path feature extraction network. The performance of the Rg-MCMP segmentation framework was evaluated and compared with the single-channel single-path model (SCSP), the standalone registration methods, and the registration-guided multi-channel single-path (Rg-MCSP) model. The Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and average surface distance (ASD) were used as the metrics. The average DSC of CTV for the deformable image DIR-MCMP model was found to be 0.892, greater than that of the standalone DIR (0.856), SCSP (0.837), and DIR-MCSP (0.877), which were improvements of 4.2%, 6.6%, and 1.7%, respectively. Similarly, the rigid body DIR-MCMP model yielded an average DSC of 0.875, which exceeded standalone RB (0.787), SCSP (0.837), and registration-guided multi-channel single-path (0.848), which were improvements of 11.2%, 4.5%, and 3.2%, respectively. These improvements in DSC were statistically significant (p<0.05). The proposed Rg-MCMP framework achieved excellent accuracy in CTV segmentation as part of the adaptive radiotherapy workflow.

Joint Brain Tumor Segmentation from Multi-magnetic Resonance Sequences through a Deep Convolutional Neural Network.

Brain tumor segmentation is highly contributive in diagnosing and treatment planning. Manual brain tumor delineation is a time-consuming and tedious task and varies depending on the radiologist's skill. Automated brain tumor segmentation is of high importance and does not depend on either inter- or intra-observation. The objective of this study is to automate the delineation of brain tumors from the Fluid-attenuated inversion recovery (FLAIR), T1-weighted (T1W), T2-weighted (T2W), and T1W contrast-enhanced (T1ce) magnetic resonance (MR) sequences through a deep learning approach, with a focus on determining which MR sequence alone or which combination thereof would lead to the highest accuracy therein. The BraTS-2020 challenge dataset, containing 370 subjects with four MR sequences and manually delineated tumor masks, is applied to train a residual neural network. This network is trained and assessed separately for each one of the MR sequences (single-channel input) and any combination thereof (dual- or multi-channel input). The quantitative assessment of the single-channel models reveals that the FLAIR sequence would yield higher segmentation accuracy compared to its counterparts with a 0.77 ± 0.10 Dice index. As to considering the dual-channel models, the model with FLAIR and T2W inputs yields a 0.80 ± 0.10 Dice index, exhibiting higher performance. The joint tumor segmentation on the entire four MR sequences yields the highest overall segmentation accuracy with a 0.82 ± 0.09 Dice index. The FLAIR MR sequence is considered the best choice for tumor segmentation on a single MR sequence, while the joint segmentation on the entire four MR sequences would yield higher tumor delineation accuracy.

Identifying galaxy cluster mergers with deep neural networks using idealized Compton-y and X-ray maps

ABSTRACT We present a novel approach to identify galaxy clusters that are undergoing a merger using a deep learning approach. This paper uses massive galaxy clusters spanning 0 ≤ z ≤ 2 from The Three Hundred project, a suite of hydrodynamic resimulations of 324 large galaxy clusters. Mock, idealized Compton-y and X-ray maps were constructed for the sample, capturing them out to a radius of 2R200. The idealized nature of these maps mean they do not consider observational effects such as foreground or background astrophysical objects, any spatial resolution limits or restriction on X-ray energy bands. Half of the maps belong to a merging population as defined by a mass increase ΔM/M ≥ 0.75, and the other half serves as a controlled, relaxed population. We employ a convolutional neural network architecture and train the model to classify clusters into one of the groups. A best-performing model was able to correctly distinguish between the two populations with a balanced accuracy (BA) and recall of 0.77, ROC-AUC of 0.85, PR-AUC of 0.55, and F1 score of 0.53. Using a multichannel model relative to a single-channel model, we obtain a 3 per cent improvement in BA score, and a 6 per cent improvement in F1 score. We use a saliency interpretation approach to discern the regions most important to each classification decision. By analysing radially binned saliency values we find a preference to utilize regions out to larger distances for mergers with respect to non-mergers, greater than ∼1.2R200 and ∼0.7R200 for SZ and X-ray, respectively.

Effects of Inhomogeneous Gas Diffusion Layer Properties on the Transportation Phenomenon and Performances of Proton-Exchange Membrane Fuel Cells.

The encapsulation of the proton-exchange membrane fuel cell (PEMFC) is an essential step of fuel cell stack assembly. The selection of the assembly pressure is very important to the stack performance and life. Based on that, this paper presents a method to describe the internal physical deformation of the gas diffusion layer (GDL) after inhomogeneous pressure by using user-defined functions (UDFs). The results show that the transmission difference caused by the nonuniform GDL deformation can be clearly seen by the UDFs method, and there is an obvious transition of GDL at the interface of the channel and the rib. A three-dimensional single-channel PEMFC model is established, and an optimal assembly pressure range is obtained, between 1.5 and 2.0 MPa. The maximum thermal stress inside the cell occurs in the middle of the membrane electrode assembly and decreases as the assembly pressure increases. Furthermore, the influence of rib-channel ratios is discussed. Compared to the fuel cells with ratios of 2:1, 2:3, and 1:2, the one with 1:1 exhibits the maximum current density and the highest power density.

Multi-Channel Graph Convolutional Networks for Graphs with Inconsistent Structures and Features

Graph convolutional networks (GCNs) have attracted increasing attention in various fields due to their significant capacity to process graph-structured data. Typically, the GCN model and its variants heavily rely on the transmission of node features across the graph structure, which implicitly assumes that the graph structure and node features are consistent, i.e., they carry related information. However, in many real-world networks, node features may unexpectedly mismatch with the structural information. Existing GCNs fail to generalize to inconsistent scenarios and are even outperformed by models that ignore the graph structure or node features. To address this problem, we investigate how to extract representations from both the graph structure and node features. Consequently, we propose the multi-channel graph convolutional network (MCGCN) for graphs with inconsistent structures and features. Specifically, the MCGCN encodes the graph structure and node features using two specific convolution channels to extract two separate specific representations. Additionally, two joint convolution channels are constructed to extract the common information shared by the graph structure and node features. Finally, an attention mechanism is utilized to adaptively learn the importance weights of these channels under the guidance of the node classification task. In this way, our model can handle both consistent and inconsistent scenarios. Extensive experiments on both synthetic and real-world datasets for node classification and recommendation tasks show that our methods, MCGCN-A and MCGCN-I, achieve the best performance on seven out of eight datasets and the second-best performance on the remaining dataset. For simpler graph structures or tasks where the overhead of multiple convolution channels is not justified, traditional single-channel GCN models might be more efficient.

Analisis Sistem Antrian Cafe Ipong Kane Menggunakan Metode Single Channel

The purpose of this research is to find out how long it takes customers when experiencing a queue for service at Cafe Ipong Kane using the single channel method. The research methodology used is descriptive observational research by directly observing the queuing system that occurs at the Ipong Kane Café. The results and discussion of this research are the arrival of data and services, and queue calculations. The conclusion of the research that has been carried out is: Single channel model queuing process, from the calculation that the average number of customers waiting in line is around 10 customers within 25 minutes in busy conditions. The average time spent by customers waiting in line is around 9.6 minutes at the longest and 3 minutes at the fastest

An Electrochemical Performance Model Considering of Non-Uniform Gas Distribution Based on Porous Media Method in PEMFC Stack

Proton exchange membrane fuel cell (PEMFC) is significant and favorable to the long-range and short refueling time in the vehicle industry. However, the non-uniform distribution of gas flow supply, particularly in the fuel cell stack is neglected in the electrochemical model for PEMFC performance optimization. The purpose of this study is to break through this limitation to establish an optimized electrochemical fuel cell performance model, with porous media methods considering the non-uniform gas flow distribution in fuel cell stack with different compression of the gas distribution layer (GDL). The numerical models are validated by experimentation of a practical fuel cell stack. For the established fuel cell model, there is a 5% difference between the maximum and minimum speeds of various flow channels in the anode flow field under 10% GDL compression. Furthermore, the single-channel electrochemical performance model is optimized by considering the non-uniform gas flow distribution of the fuel cell stack. The results of the optimized electrochemical fuel cell performance model demonstrate that the correlation coefficient between the experiment results and the simulation results is nearly 99.50%, which is higher than that of the original model under 20% GDL compression. This established model is effective in enhancing the prediction accuracy of the PEMFC performance.

Thermal-hydraulic analysis of lead-bismuth cooled reactors core based on single channel model

Thermal-hydraulic analysis of lead-bismuth cooled reactors core based on single channel model

Mass and efficiency optimization of a He-Xe Brayton cycle for mobile, land-based nuclear microreactors

Nuclear microreactors represent an option for distributed power generation, which can be deployed in military installations, remote off-grid villages, and locations recovering from natural disasters. Small system mass is necessary to achieve their mobility. In this study, a He-Xe gas-cooled reactor coupled with a Closed Brayton cycle (CBC) is optimized in terms of system mass and thermal efficiency. A single-channel model is developed for the reactor to predict the minimum mass according to the thermal–hydraulic limitations. Masses of the heat exchanger and turbo-machines are calculated based on empirical curves. The simple regenerative and intercooling cycles are considered. A parametric study is conducted to investigate the influence of key parameters on system mass and to explore the tradeoffs between different components that satisfy the power requirements of 8 MWe. Furthermore, the Non-dominated Sorting Genetic Algorithm (NSGA-II) is used to optimize the system mass and thermal efficiency. Optimal values of critical parameters on the Pareto Frontier are presented and discussed. Results show that the masses of the reactor core and coolers dominate the total mass in most cases, and increasing the height of the reactor core can help reduce the system mass by allowing a smaller core radius. The Pareto Frontier shows that at thermal efficiencies around 40% or below, the system mass is depressed by using large CPR, molecular weight of the mixture, height of the reactor core, and small recuperator effectiveness. Increasing the recuperator's effectiveness from about 90% is required for higher thermal efficiency design, but its mass increases dramatically and can surpass the mass share of other components. The current analysis can provide a guide for detailed system and component design.

Predicting tumor deposits in rectal cancer: a combined deep learning model using T2-MR imaging and clinical features

BackgroundTumor deposits (TDs) are associated with poor prognosis in rectal cancer (RC). This study aims to develop and validate a deep learning (DL) model incorporating T2-MR image and clinical factors for the preoperative prediction of TDs in RC patients.Methods and methodsA total of 327 RC patients with pathologically confirmed TDs status from January 2016 to December 2019 were retrospectively recruited, and the T2-MR images and clinical variables were collected. Patients were randomly split into a development dataset (n = 246) and an independent testing dataset (n = 81). A single-channel DL model, a multi-channel DL model, a hybrid DL model, and a clinical model were constructed. The performance of these predictive models was assessed by using receiver operating characteristics (ROC) analysis and decision curve analysis (DCA).ResultsThe areas under the curves (AUCs) of the clinical, single-DL, multi-DL, and hybrid-DL models were 0.734 (95% CI, 0.674–0.788), 0.710 (95% CI, 0.649–0.766), 0.767 (95% CI, 0.710–0.819), and 0.857 (95% CI, 0.807–0.898) in the development dataset. The AUC of the hybrid-DL model was significantly higher than the single-DL and multi-DL models (both p < 0.001) in the development dataset, and the single-DL model (p = 0.028) in the testing dataset. Decision curve analysis demonstrated the hybrid-DL model had higher net benefit than other models across the majority range of threshold probabilities.ConclusionsThe proposed hybrid-DL model achieved good predictive efficacy and could be used to predict tumor deposits in rectal cancer.Critical relevance statementThe proposed hybrid-DL model achieved good predictive efficacy and could be used to predict tumor deposits in rectal cancer.Key points• Preoperative non-invasive identification of TDs is of great clinical significance.• The combined hybrid-DL model achieved good predictive efficacy and could be used to predict tumor deposits in rectal cancer.• A preoperative nomogram provides gastroenterologist with an accurate and effective tool.Graphical

DTIN: Dual Transformer-based Imputation Nets for multivariate time series emitter missing data

As a kind of multivariate time series (MTS) data, emitter signals often exhibit missing or corrupt values, posing serious challenges to emitter data research such as specific emitter identification (SEI). Existing multivariate missing data imputation (MDI) methods are deficient in two aspects when applied to MTS emitter data: first, single-channel models cannot handle variables with varying numbers of complete samples; second, they cannot efficiently impute MTS data that are out of the model’s domain. To address these issues, a dual channel architecture tailored for MTS emitter data was devised in this study, which is called Dual Transformer-based Imputation Nets (DTIN). DTIN processes different types of variables through two parallel channels to extract different spatiotemporal features. Furthermore, drawing inspiration from image style manipulation, multivariate time series pivotal tuning inversion (MTSPTI) techniques are employed for better imputation performance, in which an in-domain pivotal code is created and input into the generator that is tuned for out-of-domain MTS emitter data. Extensive experiments on two real-world emitter datasets demonstrate that DTIN outperforms several existing MDI models.