Automatic Pipeline Research Articles

Twice matched fruit counting system: An automatic fruit counting pipeline in modern apple orchard using mutual and secondary matches

Twice matched fruit counting system: An automatic fruit counting pipeline in modern apple orchard using mutual and secondary matches

Automatic statistical diagnosis of COVID-19 based on multi-modal CT feature extraction

Background and purposeComputed tomography (CT) is highly sensitive to lung-related abnormalities as a non-invasive method and has become an essential tool for screening and diagnosing Coronavirus disease 2019 (COVID-19). To reduce the stress of work for physicians and speed up diagnosis, we propose a novel automatic diagnosis pipeline for COVID-19 based on high-dimensional radiomic features extracted from multimodal CT scans (multi-geometric and multiscale). Materials and methodsThere are 746 CT scans involved in this study, where 349 CT scans are COVID-19 positive and 397 CT scans are COVID-19 negative. All of them are from the public dataset. We first construct a transfer learning-based auto-segmentation model with a morphological post-processing block to improve the lung region segmentation. Then the radiomics feature extraction is guided by the proposed multi-modal CT scans strategy. In addition, our automatic diagnosis pipeline is driven by a well-designed loss function. We also explain the diagnosis capability from the related theory of linear subspace spanned by multi-modal radiomics features. ResultsUnder the 10-fold cross-validation strategy, our approach can achieve an improvement in diagnostic performance of 5. 77%, 7. 78%, 7. 74%, 7. 78%, 7. 45% compared to the radiomic features extracted from the original CT scans, and diagnosis performance is promoted to 91.53%, 86.46%, 86.47%, 86.46%, 86.95% in terms of AUC, Acc, F1, Recall and Precision in public datasets. ConclusionsWe demonstrate a statistically significant improvement of the proposed statistical learning method compared to the state-of-the-art machine learning-based diagnosis approaches. Thanks to theoretical support and excellent diagnostic performance, our method can be deployed in clinical auxiliary diagnosis, releasing the overstretched medical resources.

Knee Bone Models From Ultrasound.

The number of total knee arthroplasties performed worldwide is on the rise. Patient-specific planning and implants may improve surgical outcomes but require 3-D models of the bones involved. Ultrasound (US) may become a cheap and nonharmful imaging modality if the shortcomings of segmentation techniques in terms of automation, accuracy, and robustness are overcome; furthermore, any kind of US-based bone reconstruction must involve some kind of model completion to handle occluded areas, for example, the frontal femur. A fully automatic and robust processing pipeline is proposed, generating full bone models from 3-D freehand US scanning. A convolutional neural network (CNN) is combined with a statistical shape model (SSM) to segment and extrapolate the bone surface. We evaluate the method in vivo on ten subjects, comparing the US-based model to a magnetic resonance imaging (MRI) reference. The partial freehand 3-D record of the femur and tibia bones deviate by 0.7-0.8 mm from the MRI reference. After completion, the full bone model shows an average submillimetric error in the case of the femur and 1.24 mm in the case of the tibia. Processing of the images is performed in real time, and the final model fitting step is computed in less than a minute. It took an average of 22 min for a full record per subject.

Meta-data for absorption tomography measurements

The simplicity of nonclinical x-ray tomography data collection has caused some to overlook the importance of saving additional experimental information or experiment meta-data. Sample meta-data are often saved in the experimenter’s logbook while meta-data about the instrument and experimental conditions are saved by the instrument itself or by the instrument operator. The lack of standardization of this approach has limited the development of automatic tools for data analysis and experiment logging but has also hindered the ability to reproduce the data collection and data analysis under the same conditions. In this paper we introduce tomo-meta, a publicly available repository of laboratory and synchrotron based tomography instrument meta-data files with the aim of presenting how meta-data are currently collected and identify best practices that enable data collection and data analysis repeatability. Structured and machine readable meta-data files, such as HDF, CSV, JSON, XML, etc., are essential for creating automatic processing pipeline. When the tomography meta-data files are structured as machine readable, we also provide a simple python script to automatically load them into a python dictionary.

Autologous Transplantation Tooth Guide Design Based on Deep Learning

BackgroundAutogenous tooth transplantation requires precise surgical guide design, involving manual tracing of donor tooth contours based on patient CBCT scans. While manual corrections are time-consuming and prone to human errors, deep learning-based approaches show promise in reducing labor and time costs while minimizing errors. However, the application of deep learning techniques in this particular field is yet to be investigated. PurposeWe aimed to assess the feasibility of replacing the traditional design pipeline with a deep learning-enabled autologous tooth transplantation guide design pipeline. Study design, setting, sampleThis retrospective cross-sectional study used 79 CBCT images collected at the Guangzhou Medical University Hospital between October 2022 and March 2023. Following preprocessing, a total of 5,070 region of interest (ROI) images were extracted from 79 CBCT images. Predictor variableAutologous tooth transplantation guide design pipelines, either based on traditional manual design or deep learning-based Main outcome variableThe main outcome variable was the error between the reconstructed model and the gold standard benchmark. We used the third molar extracted clinically as the gold standard and leveraged it as the benchmark for evaluating our reconstructed models from different design pipelines. Both trueness and accuracy were used to evaluate this error. Trueness was assessed using the root mean square (RMS), and accuracy was measured using the standard deviation (SD). The secondary outcome variable was the pipeline efficiency, assessed based on the time cost. Time cost refers to the amount of time required to acquire the third molar model using the pipeline. AnalysesData were analyzed using the Kruskal–Wallis test. Statistical significance was set at p < 0.05. ResultsIn the surface matching comparison for different reconstructed models, the deep learning group achieved the lowest RMS value (0.335±0.066 mm). There were no significant differences in RMS values between manual design by a senior doctor and deep learning-based design (p=0.688), and the SD values did not differ among the three groups (p=0.103). The deep learning-based design pipeline (0.017±0.001 min) provided a faster assessment compared to the manual design pipeline by both senior (19.676±2.386 min) and junior doctors (30.613±6.571 min) (p<0.001). Conclusions and RelevanceThe deep learning-based automatic pipeline exhibited similar performance in surgical guide design for autogenous tooth transplantation compared to manual design by senior doctors, and it minimized time costs.

Constant false alarm rate detection of pipeline leakage based on acoustic sensors

During the transportation of oil and gas pipelines, there are many potential factors that can lead to pipeline leakage with serious consequences, making automatic and real-time pipeline leakage detection urgent. In response to the inconvenience of manual detection, constant false alarm rate (CFAR) detection technique in radar target detection theory is introduced for pipeline leakage detection based on acoustic signals. In this paper, an automatic pipeline leakage detection algorithm based on an improved CFAR detector is proposed. The improved CFAR detection is executed after pre-processing the acoustic signals so as to adaptively set the detection threshold to achieve the purpose of automatic detection of pipeline leakage incidents. A simulated leakage test of a real pipeline is used for validation, and the proposed method achieves detection accuracies of 84.6%, 97.7%, and 98% for different leakage diameter settings, i.e., 5 mm, 7 mm, and 10 mm leak hole diameters, respectively, with an overall accuracy of 94.1%, while the false alarm rates are 3.3%, 0.7%, and 0, respectively, as well as an overall of 1.2%. The results of experimental data based on real scenarios demonstrate the effectiveness of the proposed method.

Automatic Data Curation and Analysis Pipeline for Electrochemical Impedance Spectroscopy Measurements Conducted on Solid Oxide Cell Stacks

To better understand degradation in electrochemical converters and helping to correlate certain phenomena with specific operating conditions, machine learning (ML) methods are increasingly being applied. Success has already been achieved in the field of degradation analysis and prediction of capacity of lithium ion batteries1, for instance. In terms of Solid Oxide Cell (SOC) stacks ML methods have been applied mainly with the aim of identification of faulty operation modes and degradation related fault diagnosis2. ML approaches usually require a considerable amount of real training data, when used for forecasting models. A data consolidation and curation strategy was developed with the aim of processing the historic long-term test bench data of SOCs collected by Forschungszentrum Jülich over the past years. In comparison to other datasets developed in this field3, the one presented in this work contains SOC stack tests in fuel cell operation with significantly longer operating times under load. A compilation of the sample experiments and the consolidation into a hierarchical data format are presented. Further, an essential part of the strategy is the automatic curation and analysis of electrochemical impedance spectroscopy (EIS) measurements, using a specifically developed procedure in Python. The varying quality of measurements from past years, as well as recurring artefacts such as parasitic inductances, can be addressed in this way. Additional distribution of relaxation times (DRT) deconvolutions and equivalent circuit modelling (ECM) are performed, as part of the procedure to automatically retrieve feature values from measurements (cf. Fig. 1). The novel dataset, which to the authors’ knowledge includes some of the longest SOC stack tests available, serves as the basis for several evaluations. In addition to classification and clustering work to derive degradation patterns, in particular based on the EIS data, another focus is on the development of forecasting models. The current work is primarily concerned with long short-term memory (LSTM), as well as regression models that make use of both the time series data and the characterisation measurements, such as EIS. Acknowledgement The authors would like to thank their colleagues at Forschungszentrum Jülich GmbH for their great support and the Helmholtz Society as well as the German Federal Ministry of Education and Research for financing these activities as part of the WirLebenSOFC project (03SF0622B). References 1: Jones, P.K., Stimming, U. & Lee, A.A. Impedance-based forecasting of lithium-ion battery performance amid uneven usage. Nature Communications 13, 4806 (2022).2: B. Yang et al. Solid oxide fuel cell systems fault diagnosis: Critical summarization, classification, and perspectives. Journal of Energy Storage 34, 102153 (2021).3: A.K. Padinjarethil, S. Pollok & A. Hagen. Degradation studies using machine learning on novel solid oxide cell database. Fuel Cells 21, 566–576 (2021). Figure caption: Fig.1: Flow diagram of EIS data curation pipeline and curation results for example EIS measurement. Figure 1

ArrayNet: A Combined Seismic Phase Classification and Back-Azimuth Regression Neural Network for Array Processing Pipelines

ABSTRACT Array processing is an integral part of automatic seismic event detection pipelines for measuring apparent velocity and backazimuth of seismic arrivals. Both quantities are usually measured under the plane-wave assumption, and are essential to classify the phase type and to determine the direction toward the event epicenter. However, structural inhomogeneities can lead to deviations from the plane-wave model, which must be taken into account for phase classification and back-azimuth estimation. We suggest a combined classification and regression neural network, which we call ArrayNet, to determine the phase type and backazimuth directly from the arrival-time differences between all combinations of stations of a given seismic array without assuming a plane-wave model. ArrayNet is trained using regional P- and S-wave arrivals of over 30,000 seismic events from reviewed regional bulletins in northern Europe from the past three decades. ArrayNet models are generated and trained for each of the ARCES, FINES, and SPITS seismic arrays. We observe excellent performance for the seismic phase classification (up to 99% accuracy), and the derived back-azimuth residuals are significantly improved in comparison with traditional array processing results using the plane-wave assumption. The SPITS array in Svalbard exhibits particular issues when it comes to array processing in the form of high apparent seismic velocities and a multitude of frost quake signals inside the array, and we show how our new approach better handles these obstacles. Furthermore, we demonstrate the performance of ArrayNet on 20 months of continuous phase detections from the ARCES array and investigate the results for a selection of regional seismic events of interest. Our results demonstrate that automatic event detection at seismic arrays can be further enhanced using a machine learning approach that takes advantage of the unique array data recorded at these stations.

A fully automatic data processing and structure determination pipeline for X-ray biological macromolecule crystallography beamlines

A fully automatic data processing and structure determination pipeline for X-ray biological macromolecule crystallography beamlines

Automatic Deep Learning-Based Pipeline for Automatic Delineation and Measurement of Fetal Brain Structures in Routine Mid-Trimester Ultrasound Images

Introduction: The aim of this study was to develop a pipeline using state-of-the-art deep learning methods to automatically delineate and measure several of the most important brain structures in fetal brain ultrasound (US) images. Methods: The dataset was composed of 5,331 images of the fetal brain acquired during the routine mid-trimester US scan. Our proposed pipeline automatically performs the following three steps: brain plane classification (transventricular, transthalamic, or transcerebellar plane); brain structures delineation (9 different structures); and automatic measurement (from the structure delineations). The methods were trained on a subset of 4,331 images and each step was evaluated on the remaining 1,000 images. Results: Plane classification reached 98.6% average class accuracy. Brain structure delineation obtained an average pixel accuracy higher than 96% and a Jaccard index higher than 70%. Automatic measurements get an absolute error below 3.5% for the four standard head biometries (head circumference, biparietal diameter, occipitofrontal diameter, and cephalic index), 9% for transcerebellar diameter, 12% for cavum septi pellucidi ratio, and 26% for Sylvian fissure operculization degree. Conclusions: The proposed pipeline shows the potential of deep learning methods to delineate fetal head and brain structures and obtain automatic measures of each anatomical standard plane acquired during routine fetal US examination.

A Fully Automatic Method to Segment Choroid Plexuses in Multiple Sclerosis Using Conventional MRI Sequences.

Choroid plexus (CP) volume has been recently proposed as a proxy for brain neuroinflammation in multiple sclerosis (MS). To develop and validate a fast automatic method to segment CP using routinely acquired brain T1-weighted and FLAIR MRI. Retrospective. Fifty-five MS patients (33 relapsing-remitting, 22 progressive; mean age = 46.8 ± 10.2 years; 31 women) and 60 healthy controls (HC; mean age = 36.1 ± 12.6 years, 33 women). 3D T2-weighted FLAIR and 3D T1-weighted gradient echo sequences at 3.0 T. Brain tissues were segmented on T1-weighted sequences and a Gaussian Mixture Model (GMM) was fitted to FLAIR image intensities obtained from the ventricle masks of the SIENAX. A second GMM was then applied on the thresholded and filtered ventricle mask. CP volumes were automatically determined and compared with those from manual segmentation by two raters (with 3 and 10 years' experience; reference standard). CP volumes from previously published automatic segmentation methods (freely available Freesurfer [FS] and FS-GMM) were also compared with reference standard. Expanded Disability Status Scale (EDSS) score was assessed within 3 days of MRI. Computational time was assessed for each automatic technique and manual segmentation. Comparisons of CP volumes with reference standard were evaluated with Bland Altman analysis. Dice similarity coefficients (DSC) were computed to assess automatic CP segmentations. Volume differences between MS and HC for each method were assessed with t-tests and correlations of CP volumes with EDSS were assessed with Pearson's correlation coefficients (R). A P value <0.05 was considered statistically significant. Compared to manual segmentation, the proposed method had the highest segmentation accuracy (mean DSC = 0.65 ± 0.06) compared to FS (mean DSC = 0.37 ± 0.08) and FS-GMM (0.58 ± 0.06). The percentage CP volume differences relative to manual segmentation were -0.1% ± 0.23, 4.6% ± 2.5, and -0.48% ± 2 for the proposed method, FS, and FS-GMM, respectively. The Pearson's correlations between automatically obtained CP volumes and the manually obtained volumes were 0.70, 0.54, and 0.56 for the proposed method, FS, and FS-GMM, respectively. A significant correlation between CP volume and EDSS was found for the proposed automatic pipeline (R = 0.2), for FS-GMM (R = 0.3) and for manual segmentation (R = 0.4). Computational time for the proposed method (32 ± 2 minutes) was similar to the manual segmentation (20 ± 5 minutes) but <25% of the FS (120 ± 15 minutes) and FS-GMM (125 ± 15 minutes) methods. This study developed an accurate and easily implementable method for automatic CP segmentation in MS using T1-weighted and FLAIR MRI. 1 TECHNICAL EFFICACY: Stage 4.

Accurate measurement of magnetic resonance parkinsonism index by a fully automatic and deep learning quantification pipeline.

This study aims at a fully automatic pipeline for measuring the magnetic resonance parkinsonism index (MRPI) using deep learning methods. MRPI is defined as the product of the pons area to the midbrain area ratio and the middle cerebellar peduncle (MCP) width to the superior cerebellar peduncle (SCP) width ratio. In our proposed pipeline, we first used nnUNet to segment the brainstem and then employed HRNet to identify two key boundary points so as to sub-divide the whole brainstem into midbrain and pons. HRNet was also employed to predict the MCP endpoints for measuring the MCP width. Finally, we segmented the SCP on an oblique coronal plane and calculated its width. A total of 400 T1-weighted magnetic resonance images (MRIs) were used to train the nnUNet and HRNet models. Five-fold cross-validation was conducted to evaluate our proposed pipeline's performance on the training dataset. We also evaluated the performance of our proposed pipeline on three external datasets. Two of them had two raters manually measuring the MRPI values, providing insights into automatic accuracy versus inter-rater variability. We obtained average absolute percentage errors (APEs) of 17.21%, 18.17%, 20.83%, and 22.83% on the training dataset and the three external validation datasets, while the inter-rater average APE measured on the first two external validation datasets was 11.31%. Our proposed pipeline significantly improved the MRPI quantification accuracy over a representative state-of-the-art traditional approach (p < 0.001). The proposed automatic pipeline can accurately predict MRPI that is comparable with manual measurement. This study presents an automated magnetic resonance parkinsonism index measurement tool that can analyze large amounts of magnetic resonance images, enhance the efficiency of Parkinsonism-Plus syndrome diagnosis, reduce the workload of clinicians, and minimize the impact of human factors on diagnosis. • We propose an automatic pipeline for measuring the magnetic resonance parkinsonism index from magnetic resonance images. • The effectiveness of the proposed pipeline is successfully established on multiple datasets and comparisons with inter-rater measurements. • The proposed pipeline significantly outperforms a state-of-the-art quantification approach, being much closer to ground truth.

DETECTION OF KNEE OSTEOARTHRITIS BASED ON CENTER OF PRESSURE DATA AND THE BAT ALGORITHM

The high rate of knee osteoarthritis has raised the need for accurate diagnostic methods. In this study, we propose a precise detection method using the center of pressure data obtained from the patients. The introduced automatic detection pipeline is based on the two modern algorithms of grey wolf and BAT. The extracted statistical features and the obtained data from healthy individuals and patients are processed with the grey wolf binary algorithm. The results are fed into the binary bat algorithm to select important features and increase the pipeline accuracy. Then the groups are classified using a four-layer neural network. We show that the proposed method with a simple four-layer neural network offers fantastic accuracy in high-speed processing large data and classifies the high-dimensional knee osteoarthritis center of pressure data with appropriate precision, recall, specificity, and F1 values. The proposed method has direct applications in knee osteoarthritis diagnostics in clinics.

An automated nondestructive testing system for the surface of pressure pipeline welds

AbstractPressure pipelines are widely used in hydropower generation, oil and gas transmission, and other fields. After years of operation, a pressure pipeline needs regular maintenance to ensure its safety. At present, manual detection methods are unable to meet this demand. An automatic pressure pipeline detection technology is urgently needed to achieve improved efficiency and accuracy. On the basis of the above requirements, a wall‐climbing robot is designed for automatic pressure pipe inspection and maintenance tasks. Moreover, rapid nondestructive testing of welds on the inner surface of pressure pipelines was performed, and a weld tracking function was developed for wall‐climbing robots. We propose an algorithm framework for weld recognition and centerline extraction by combining computer vision technology with traditional image processing technology using visual images. The experimental verification of the wall‐climbing robot designed in this paper and the algorithm framework for weld recognition and centerline extraction were performed based on actual pressure pipelines. The results show that the algorithm framework developed based on the wall‐climbing robot equipped with an industrial camera for pressure pipeline weld detection can achieve greatly improved efficiency, and the actual weld identification accuracy can exceed 90%, which is very meaningful for practical applications.

VPBR: An Automatic and Low-Cost Vision-Based Biophysical Properties Recognition Pipeline for Pumpkin.

Pumpkins are a nutritious and globally enjoyed fruit for their rich and earthy flavor. The biophysical properties of pumpkins play an important role in determining their yield. However, manual in-field techniques for monitoring these properties can be time-consuming and labor-intensive. To address this, this research introduces a novel approach that feeds high-resolution pumpkin images to train a mathematical model to automate the measurement of each pumpkin's biophysical properties. Color correction was performed on the dataset using a color-checker panel to minimize the impact of varying light conditions on the RGB images. A segmentation model was then trained to effectively recognize two fundamental components of each pumpkin: the fruit and vine. Real-life measurements of various biophysical properties, including fruit length, fruit width, stem length, stem width and fruit peel color, were computed and compared with manual measurements. The experimental results on 10 different pumpkin samples revealed that the framework obtained a small average mean absolute percentage error (MAPE) of 2.5% compared to the manual method, highlighting the potential of this approach as a faster and more efficient alternative to conventional techniques for monitoring the biophysical properties of pumpkins.

Hemodynamics of thrombus formation in intracranial aneurysms: An in silico observational study.

How prevalent is spontaneous thrombosis in a population containing all sizes of intracranial aneurysms? How can we calibrate computational models of thrombosis based on published data? How does spontaneous thrombosis differ in normo- and hypertensive subjects? We address the first question through a thorough analysis of published datasets that provide spontaneous thrombosis rates across different aneurysm characteristics. This analysis provides data for a subgroup of the general population of aneurysms, namely, those of large and giant size (>10 mm). Based on these observed spontaneous thrombosis rates, our computational modeling platform enables the first in silico observational study of spontaneous thrombosis prevalence across a broader set of aneurysm phenotypes. We generate 109 virtual patients and use a novel approach to calibrate two trigger thresholds: residence time and shear rate, thus addressing the second question. We then address the third question by utilizing this calibrated model to provide new insight into the effects of hypertension on spontaneous thrombosis. We demonstrate how a mechanistic thrombosis model calibrated on an intracranial aneurysm cohort can help estimate spontaneous thrombosis prevalence in a broader aneurysm population. This study is enabled through a fully automatic multi-scale modeling pipeline. We use the clinical spontaneous thrombosis data as an indirect population-level validation of a complex computational modeling framework. Furthermore, our framework allows exploration of the influence of hypertension in spontaneous thrombosis. This lays the foundation for in silico clinical trials of cerebrovascular devices in high-risk populations, e.g., assessing the performance of flow diverters in aneurysms for hypertensive patients.

Harmonized multi-metric and multi-centric assessment of EEG source space connectivity for dementia characterization.

Harmonization protocols that address batch effects and cross-site methodological differences in multi-center studies are critical for strengthening electroencephalography (EEG) signatures of functional connectivity (FC) as potential dementia biomarkers. We implemented an automatic processing pipeline incorporating electrode layout integrations, patient-control normalizations, and multi-metric EEG source space connectomics analyses. Spline interpolations of EEG signals onto a head mesh model with 6067 virtual electrodes resulted in an effective method for integrating electrode layouts. Z-score transformations of EEG time series resulted in source space connectivity matrices with high bilateral symmetry, reinforced long-range connections, and diminished short-range functional interactions. A composite FC metric allowed for accurate multicentric classifications of Alzheimer's disease and behavioral variant frontotemporal dementia. Harmonized multi-metric analysis of EEG source space connectivity can address data heterogeneities in multi-centric studies, representing a powerful tool for accurately characterizing dementia.

Radiomics-Based Quality Control System for Automatic Cardiac Segmentation: A Feasibility Study.

In the past decade, there has been a rapid increase in the development of automatic cardiac segmentation methods. However, the automatic quality control (QC) of these segmentation methods has received less attention. This study aims to address this gap by developing an automatic pipeline that incorporates DL-based cardiac segmentation and radiomics-based quality control. In the DL-based localization and segmentation part, the entire heart was first located and cropped. Then, the cropped images were further utilized for the segmentation of the right ventricle cavity (RVC), myocardium (MYO), and left ventricle cavity (LVC). As for the radiomics-based QC part, a training radiomics dataset was created with segmentation tasks of various quality. This dataset was used for feature extraction, selection, and QC model development. The model performance was then evaluated using both internal and external testing datasets. In the internal testing dataset, the segmentation model demonstrated a great performance with a dice similarity coefficient (DSC) of 0.954 for whole heart segmentations. Images were then appropriately cropped to 160 × 160 pixels. The models also performed well for cardiac substructure segmentations. The DSC values were 0.863, 0.872, and 0.940 for RVC, MYO, and LVC for 2D masks and 0.928, 0.886, and 0.962 for RVC, MYO, and LVC for 3D masks with an attention-UNet. After feature selection with the radiomics dataset, we developed a series of models to predict the automatic segmentation quality and its DSC value for the RVC, MYO, and LVC structures. The mean absolute values for our best prediction models were 0.060, 0.032, and 0.021 for 2D segmentations and 0.027, 0.017, and 0.011 for 3D segmentations, respectively. Additionally, the radiomics-based classification models demonstrated a high negative detection rate of >0.85 in all 2D groups. In the external dataset, models showed similar results. We developed a pipeline including cardiac substructure segmentation and QC at both the slice (2D) and subject (3D) levels. Our results demonstrate that the radiomics method possesses great potential for the automatic QC of cardiac segmentation.

Automatic Pipeline Parallelism: A Parallel Inference Framework for Deep Learning Applications in 6G Mobile Communication Systems

With the rapid development of wireless communication, achieving the neXt generation Ultra-Reliable and Low-Latency Communications (xURLLC) in 6G mobile communication systems has become a critical problem. Among many applications in xURLLC, deep learning model inference requires improvement over its efficiency. Due to the heterogeneous hardware environment in 6G, parallel schedules from distributed machine learning and edge computing has been borrowed to tackle the efficiency problem. However, traditional parallel schedules suffer from high latency, low throughput, and low device utility. In this paper, we propose Automatic Pipeline Parallelism ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AP</i> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ), a parallel inference framework for deep learning applications in 6G mobile communication systems, to improve the model inference efficiency while maintaining reliability. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AP</i> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> contains three sub-modules. A task-device affinity predictor predicts a task’s expected execution time on a given device. The parallel inference arrangement optimizer finds the most suitable device for each task. The parallel inference scheduler converts the arrangement to a schedule that can be directly executed in the system. The experimental results show that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AP</i> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> can achieve better latency, throughput, reliability, and device utility than other parallel schedules. Also, the priority of the sub-module designs has been approved through the experiments.

Data-constrained Solar Modeling with GX Simulator

To facilitate the study of solar flares and active regions, we have created a modeling framework, the freely distributed GX Simulator IDL package, that combines 3D magnetic and plasma structures with thermal and nonthermal models of the chromosphere, transition region, and corona. Its object-based modular architecture, which runs on Windows, Mac, and Unix/Linux platforms, offers the ability to either import 3D density and temperature distribution models, or to assign numerically defined coronal or chromospheric temperatures and densities, or their distributions, to each individual voxel. GX Simulator can apply parametric heating models involving average properties of the magnetic field lines crossing a given voxel, as well as compute and investigate the spatial and spectral properties of radio, (sub)millimeter, EUV, and X-ray emissions calculated from the model, and quantitatively compare them with observations. The package includes a fully automatic model production pipeline that, based on minimal users input, downloads the required SDO/HMI vector magnetic field data, performs potential or nonlinear force-free field extrapolations, populates the magnetic field skeleton with parameterized heated plasma coronal models that assume either steady-state or impulsive plasma heating, and generates non-LTE density and temperature distribution models of the chromosphere that are constrained by photospheric measurements. The standardized models produced by this pipeline may be further customized through specialized IDL scripts, or a set of interactive tools provided by the graphical user interface. Here, we describe the GX Simulator framework and its applications.