Network Pipeline Research Articles

Neural networks pipeline for quality management in IVF laboratory

This study introduces a novel neural network-based pipeline for predicting clinical pregnancy rates in IVF treatments, integrating both clinical and laboratory data. We developed a metamodel combining deep neural networks and Kolmogorov-Arnold networks, leveraging their complementary strengths to enhance predictive accuracy and interpretability. The metamodel achieved robust performance metrics after training and fitting on 11500 clinical cases: accuracy = 0.72, AUC = 0.75, F1 score = 0.60, and Matthews Correlation Coefficient of 0.42. According to morpho-kinetical embryo evaluation, our model’s PRC of 0.66 significantly improves over existing time-lapse systems for pregnancy prediction, demonstrating better handling of imbalanced clinical data. The metamodel’s calibration metrics (Brier score = 0.20, expected calibration error = 0.06, maximum calibration error = 0.12, Hosmer-Lemeshow test p-value = 0.06) indicate robust reliability in predicting clinical pregnancy outcomes. We validated the model’s reproducibility using an independent dataset of 665 treatment cycles, showing close alignment between predicted and actual pregnancy rates (58.9% vs. 59.1%). With the Bayesian method, we proposed a robust framework for integrating historical data with real-time predictions from neural networks, enabling a transition from retrospective to prospective analysis. Our approach extends beyond conventional embryo selection, incorporating post-analytical phase evaluation in the IVF laboratory. This comprehensive framework enables detailed analysis across different patient subpopulations and time periods, facilitating the identification of systemic issues and IVF protocol optimization. The model’s ability to track pregnancy probabilities over time and staff members allows for both outcome prediction and retrospective and prospective assessment of IVF treatment efficacy, providing a data-driven strategy for continuous improvement in assisted reproductive technology.

Genetic networks suggest Asperger’s syndrome as a distinct subtype of autism spectrum disorders

BackgroundThe Diagnostic and Statistical Manual of Mental Disorders (DSM-V) issued new diagnostic criteria for autism spectrum disorders (ASD) which resulted in missing the diagnosis of some cases of Asperger’s syndrome (AS). This negatively affected the support received by those affected. In this study, we explored if AS could be biologically stratified from the broader spectrum through a gene co-expression network preservation analysis. MethodsWe analysed the GEO microarray data of 24 individuals with Asperger’s syndrome and 72 individuals with autism. Then, we used a weighted gene co-expression network (WGCNA) pipeline to construct gene co-expression networks. We explored whether these modules share the same co-expression patterns between autism and Asperger’s syndrome using network preservation analysis. ResultsOur results showed that all co-expression modules of autism are preserved into the Asperger’s syndrome. However, three modules of Asperger’s syndrome out of 30 modules were not preserved in autism.Gene enrichment analysis revealed that these modules were involved in chromatin remodelling, immune and neuroinflammatory response, synaptic and neuronal development. Brain enrichment analysis showed significant downregulation of neurodevelopment genes in different brain regions associated with impaired social recognition in Asperger’s syndrome. ConclusionsThe identified genetic and molecular profiles suggest that Asperger’s syndrome, despite sharing numerous similarities with autism, possesses a distinct genetic profile that makes it a distinct subtype of autism. This distinction could have significant implications for the management and treatment strategies tailored to individuals with Asperger’s syndrome.

Stage-Aware Interaction Network for Point Cloud Completion

Point cloud completion aims to restore full shapes of objects from partial scans, and a typical network pipeline is AutoEncoder, which has coarse-to-fine refinement modules. Although existing approaches using this kind of architecture achieve promising results, they usually neglect the usage of shallow geometry features in partial inputs and the fusion of multi-stage features in the upsampling process, which prevents network performances from further improving. Therefore, in this paper, we propose a new method with dense interactions between different encoding and decoding steps. First, we introduce the Decoupled Multi-head Transformer (DMT), which implements and integrates semantic prediction and resolution upsampling in a unified network module, which serves as a primary ingredient in our pipeline. Second, we propose an Encoding-aware Coarse Decoder (ECD) that compactly makes the top–down shape-decoding process interact with the bottom–up feature-encoding process to utilize both shallow and deep features of partial inputs for coarse point cloud generation. Third, we design a Stage-aware Refinement Group (SRG), which comprehensively understands local semantics from densely connected features across different decoding stages and gradually upsamples point clouds based on them. In general, the key contributions of our method are the DMT for joint semantic-resolution generation, the ECD for multi-scale feature fusion-based shape decoding, and the SRG for stage-aware shape refinement. Evaluations on two synthetic and three real-world datasets illustrate that our method achieves competitive performances compared with existing approaches.

A Hybrid Neuromorphic Object Tracking and Classification Framework for Real-Time Systems.

Deep learning inference that needs to largely take place on the "edge" is a highly computational and memory intensive workload, making it intractable for low-power, embedded platforms such as mobile nodes and remote security applications. To address this challenge, this article proposes a real-time, hybrid neuromorphic framework for object tracking and classification using event-based cameras that possess desirable properties such as low-power consumption (5-14 mW) and high dynamic range (120 dB). Nonetheless, unlike traditional approaches of using event-by-event processing, this work uses a mixed frame and event approach to get energy savings with high performance. Using a frame-based region proposal method based on the density of foreground events, a hardware-friendly object tracking scheme is implemented using the apparent object velocity while tackling occlusion scenarios. The frame-based object track input is converted back to spikes for TrueNorth (TN) classification via the energy-efficient deep network (EEDN) pipeline. Using originally collected datasets, we train the TN model on the hardware track outputs, instead of using ground truth object locations as commonly done, and demonstrate the ability of our system to handle practical surveillance scenarios. As an alternative tracker paradigm, we also propose a continuous-time tracker with C++ implementation where each event is processed individually, which better exploits the low latency and asynchronous nature of neuromorphic vision sensors. Subsequently, we extensively compare the proposed methodologies to state-of-the-art event-based and frame-based methods for object tracking and classification, and demonstrate the use case of our neuromorphic approach for real-time and embedded applications without sacrificing performance. Finally, we also showcase the efficacy of the proposed neuromorphic system to a standard RGB camera setup when simultaneously evaluated over several hours of traffic recordings.

DeepTree: Modeling Trees With Situated Latents.

In this article, we propose DeepTree, a novel method for modeling trees based on learning developmental rules for branching structures instead of manually defining them. We call our deep neural model "situated latent" because its behavior is determined by the intrinsic state -encoded as a latent space of a deep neural model- and by the extrinsic (environmental) data that is "situated" as the location in the 3D space and on the tree structure. We use a neural network pipeline to train a situated latent space that allows us to locally predict branch growth only based on a single node in the branch graph of a tree model. We use this representation to progressively develop new branch nodes, thereby mimicking the growth process of trees. Starting from a root node, a tree is generated by iteratively querying the neural network on the newly added nodes resulting in the branching structure of the whole tree. Our method enables generating a wide variety of tree shapes without the need to define intricate parameters that control their growth and behavior. Furthermore, we show that the situated latents can also be used to encode the environmental response of tree models, e.g., when trees grow next to obstacles. We validate the effectiveness of our method by measuring the similarity of our tree models and by procedurally generated ones based on a number of established metrics for tree form.

Prediction of Pipe Failure Rate in Heating Networks Using Machine Learning Methods

The correct prediction of heating network pipeline failure rates can increase the reliability of the heat supply to consumers in the cold season. However, due to the large number of factors affecting the corrosion of underground steel pipelines, it is difficult to achieve high prediction accuracy. The purpose of this study is to identify connections between the failure rate of heating network pipelines and factors not taken into account in traditional methods, such as residual pipeline wall thickness, soil corrosion activity, previous incidents on the pipeline section, flooding (traces of flooding) of the channel, and intersections with communications. To achieve this goal, the following machine learning algorithms were used: random forest, gradient boosting, support vector machines, and artificial neural networks (multilayer perceptron). The data were collected on incidents related to the breakdown of heating network pipelines in the cities of Kazan and Ulyanovsk. Based on these data, four intelligent models have been developed. The accuracy of the models was compared. The best result was obtained for the gradient boosting regression tree, as follows: MSE = 0.00719, MAE = 0.0682, and MAPE = 0.06069. The feature «Previous incidents on the pipeline section» was excluded from the training set as the least significant.

Classification of substances by health hazard using deep neural networks and molecular electron densities

In this paper we present a method that allows leveraging 3D electron density information to train a deep neural network pipeline to segment regions of high, medium and low electronegativity and classify substances as health hazardous or non-hazardous. We show that this can be used for use-cases such as cosmetics and food products. For this purpose, we first generate 3D electron density cubes using semiempirical molecular calculations for a custom European Chemicals Agency (ECHA) subset consisting of substances labelled as hazardous and non-hazardous for cosmetic usage. Together with their 3-class electronegativity maps we train a modified 3D-UNet with electron density cubes to segment reactive sites in molecules and classify substances with an accuracy of 78.1%. We perform the same process on a custom food dataset (CompFood) consisting of hazardous and non-hazardous substances compiled from European Food Safety Authority (EFSA) OpenFoodTox, Food and Drug Administration (FDA) Generally Recognized as Safe (GRAS) and FooDB datasets to achieve a classification accuracy of 64.1%. Our results show that 3D electron densities and particularly masked electron densities, calculated by taking a product of original electron densities and regions of high and low electronegativity can be used to classify molecules for different use-cases and thus serve not only to guide safe-by-design product development but also aid in regulatory decisions.Scientific contributionWe aim to contribute to the diverse 3D molecular representations used for training machine learning algorithms by showing that a deep learning network can be trained on 3D electron density representation of molecules. This approach has previously not been used to train machine learning models and it allows utilization of the true spatial domain of the molecule for prediction of properties such as their suitability for usage in cosmetics and food products and in future, to other molecular properties. The data and code used for training is accessible at https://github.com/s-singh-ivv/eDen-Substances.

448 Deformable Medial Modeling to Generate Novel Shape Features of the Placenta Using Automated versus Manual Segmentations

OBJECTIVES/GOALS: In this study, we implemented deformable medial modeling as a morphometric approach in first trimester placentas to characterize morphometric differences between fully automated and manual segmentations. METHODS/STUDY POPULATION: Twenty placentas from singleton pregnancies between 11-14 weeks’ gestation were manually and automatically segmented from 3D ultrasound volumes. Automated segmentations were produced by a trained convolutional neural network pipeline. Dice overlap scores and volumes were computed between manual and automated segmentations. Deformable medial modeling was applied to both manual and automated segmentations to produce the following metrics: maternal and chorionic surface areas (SA), thickness, circumference, and diameter along the generated medial surface. Placental non-planarity was also determined as the greatest medial surface height difference. A paired t-test and simple linear regression was performed between manual and automated segmentations for each shape metric. RESULTS/ANTICIPATED RESULTS: Mean placental volume measurements between manual and automated segmentations were similar, with a percent difference of 3.28% and a mean Dice overlap score of 0.85 ± 0.07. There were strong, statistically significant (p <0.01) linear correlations with chorionic and maternal SA, SA difference, thickness, circumference, medial surface diameter, and medial surface height difference. No significant differences were noted with chorionic SA, thickness, circumference, maximum medial surface diameter, or medial surface height difference. However, statistically significant differences (p <0.01-0.03) were noted in maternal SA, SA difference, and mean medial surface diameter. Despite these differences, mean percent difference for all morphometric parameters was less than or equal to 10%. DISCUSSION/SIGNIFICANCE: A deformable medial model evaluate unique global and regional shape placental features with highly correlated values between manual versus automated placental segmentations. However, clinical studies are needed to determine if minor differences would impact the clinical utility of these features as potential indicators of placental function.

Effect of corrosion rate on pipe material's chemical composition in water distribution network

This study investigated the effect of corrosion on the galvanized iron due to exposure to supply water in the water distribution network. A galvanized iron sample was installed in the distribution network pipeline for 315 days. Water samples were tested for their physicochemical characteristics. The corrosion rate of galvanized iron was also determined. XRD analysis of the by-products was carried out after exposure. The EDX analysis of the galvanized iron sample was carried out without and after exposure to supply water. As a result of the experiment, the sample material's composition changed from C 11.0 % to 7.7 %, O 14.0 % to 25.3 %, Fe 10.8 % to 65.8 %, and Zn 63.2 % to 0.6 % by weight. It was illustrated by the XRD graph that when Zn reacted with supply water, by-products of Zinc Sulphate Hydroxide, Zinc Oxide, Zinc Sulphate, or Zinc Chloride were produced. It was also noticed that Zn was reduced by 62.60 %. SEM analysis indicates that Zinc coating has disappeared from the top layer. So, it can be concluded that after exposure to supply water in the distribution network, Zn reduced almost to zero. It indicates due to corrosion Zn coating was disappears from the pipe.

Automatic ARDS surveillance with chest X-ray recognition using convolutional neural networks

ObjectiveThis study aims to design, validate and assess the accuracy a deep learning model capable of differentiation Chest X-Rays between pneumonia, acute respiratory distress syndrome (ARDS) and normal lungs. Materials and methodsA diagnostic performance study was conducted using Chest X-Ray images from adult patients admitted to a medical intensive care unit between January 2003 and November 2014. X-ray images from 15,899 patients were assigned one of three prespecified categories: “ARDS”, “Pneumonia”, or “Normal”. ResultsA two-step convolutional neural network (CNN) pipeline was developed and tested to distinguish between the three patterns with sensitivity ranging from 91.8% to 97.8% and specificity ranging from 96.6% to 98.8%. The CNN model was validated with a sensitivity of 96.3% and specificity of 96.6% using a previous dataset of patients with Acute Lung Injury (ALI)/ARDS. DiscussionThe results suggest that a deep learning model based on chest x-ray pattern recognition can be a useful tool in distinguishing patients with ARDS from patients with normal lungs, providing faster results than digital surveillance tools based on text reports. ConclusionA CNN-based deep learning model showed clinically significant performance, providing potential for faster ARDS identification. Future research should prospectively evaluate these tools in a clinical setting.

Super resolution-based methodology for self-supervised segmentation of microscopy images.

Data-driven Artificial Intelligence (AI)/Machine learning (ML) image analysis approaches have gained a lot of momentum in analyzing microscopy images in bioengineering, biotechnology, and medicine. The success of these approaches crucially relies on the availability of high-quality microscopy images, which is often a challenge due to the diverse experimental conditions and modes under which these images are obtained. In this study, we propose the use of recent ML-based image super-resolution (SR) techniques for improving the image quality of microscopy images, incorporating them into multiple ML-based image analysis tasks, and describing a comprehensive study, investigating the impact of SR techniques on the segmentation of microscopy images. The impacts of four Generative Adversarial Network (GAN)- and transformer-based SR techniques on microscopy image quality are measured using three well-established quality metrics. These SR techniques are incorporated into multiple deep network pipelines using supervised, contrastive, and non-contrastive self-supervised methods to semantically segment microscopy images from multiple datasets. Our results show that the image quality of microscopy images has a direct influence on the ML model performance and that both supervised and self-supervised network pipelines using SR images perform better by 2%-6% in comparison to baselines, not using SR. Based on our experiments, we also establish that the image quality improvement threshold range [20-64] for the complemented Perception-based Image Quality Evaluator(PIQE) metric can be used as a pre-condition by domain experts to incorporate SR techniques to significantly improve segmentation performance. A plug-and-play software platform developed to integrate SR techniques with various deep networks using supervised and self-supervised learning methods is also presented.

Synthetic PET from CT improves diagnosis and prognosis for lung cancer: Proof of concept

[18F]Fluorodeoxyglucose positron emission tomography (FDG-PET) and computed tomography (CT) are indispensable components in modern medicine. Although PET can provide additional diagnostic value, it is costly and not universally accessible, particularly in low-income countries. To bridge this gap, we have developed a conditional generative adversarial network pipeline that can produce FDG-PET from diagnostic CT scans based on multi-center multi-modal lung cancer datasets (n= 1,478). Synthetic PET images are validated across imaging, biological, and clinical aspects. Radiologists confirm comparable imaging quality and tumor contrast between synthetic and actual PET scans. Radiogenomics analysis further proves that the dysregulated cancer hallmark pathways of synthetic PET are consistent with actual PET. We also demonstrate the clinical values of synthetic PET in improving lung cancer diagnosis, staging, risk prediction, and prognosis. Taken together, this proof-of-concept study testifies to the feasibility of applying deep learning to obtain high-fidelity PET translated from CT.

A Deep Learning Framework for Diagnosis and Survival Prognosis of Central Nervous System Tumors

Diagnosis and prognosis of central nervous system (CNS) tumors rely on manual and algorithmic approaches that are subject to variations, inefficiencies and bias. With 5-year survival rates as low as 6%, timely tumor assessment is crucial to ensure patient health. Machine learning and deep learning techniques have been used over the past decade to reduce the assessment burden and augment physician diagnoses. Traditional machine learning models require manual feature extraction and engineering, introducing further variability of the results. On the other hand, automated deep learning models require complicated engineering and workflow implementations that need to be managed and updated for new datasets. This study endeavors to define a generalizable, two-phase neural network pipeline and user interface that can be applied to clinical assessment of any tumor with 3D MRI scans, allowing for improved personalization of patient diagnosis and treatment options. In the first diagnostic phase, radiomic features are used to predict tumor severity and extent. The outputs of this phase are then passed to a survival prognosis model along with the patients’ clinical data to predict the 5-year overall survival rate. The diagnostic model achieved a F1-Score, a measure of classification accuracy, of 94% and the prognostic model achieved a risk-adjusted, time-dependent Harrell’s Concordance Index score of 0.92, indicating the framework’s generalizability to new datasets. The mobile app developed as part of this study offer ease of access to physicians and radiologists in reviewing predicted results and subsequent patient interactions.

A New Brain Network Construction Paradigm for Brain Disorder Via Diffusion-Based Graph Contrastive Learning.

Brain network analysis plays an increasingly important role in studying brain function and the exploring of disease mechanisms. However, existing brain network construction tools have some limitations, including dependency on empirical users, weak consistency in repeated experiments and time-consuming processes. In this work, a diffusion-based brain network pipeline, DGCL is designed for end-to-end construction of brain networks. Initially, the brain region-aware module (BRAM) precisely determines the spatial locations of brain regions by the diffusion process, avoiding subjective parameter selection. Subsequently, DGCL employs graph contrastive learning to optimize brain connections by eliminating individual differences in redundant connections unrelated to diseases, thereby enhancing the consistency of brain networks within the same group. Finally, the node-graph contrastive loss and classification loss jointly constrain the learning process of the model to obtain the reconstructed brain network, which is then used to analyze important brain connections. Validation on two datasets, ADNI and ABIDE, demonstrates that DGCL surpasses traditional methods and other deep learning models in predicting disease development stages. Significantly, the proposed model improves the efficiency and generalization of brain network construction. In summary, the proposed DGCL can be served as a universal brain network construction scheme, which can effectively identify important brain connections through generative paradigms and has the potential to provide disease interpretability support for neuroscience research.

Stripe Sensitive Convolution for Omnidirectional Image Dehazing.

The haze in a scenario may affect the 360 photo/video quality and the immersive 360 ° virtual reality (VR) experience. The recent single image dehazing methods, to date, have been only focused on plane images. In this work, we propose a novel neural network pipeline for single omnidirectional image dehazing. To create the pipeline, we build the first hazy omnidirectional image dataset, which contains both synthetic and real-world samples. Then, we propose a new stripe sensitive convolution (SSConv) to handle the distortion problems due to the equirectangular projections. The SSConv calibrates distortion in two steps: 1) extracting features using different rectangular filters and, 2) learning to select the optimal features by a weighting of the feature stripes (a series of rows in the feature maps). Subsequently, using SSConv, we design an end-to-end network that jointly learns haze removal and depth estimation from a single omnidirectional image. The estimated depth map is leveraged as the intermediate representation and provides global context and geometric information to the dehazing module. Extensive experiments on challenging synthetic and real-world omnidirectional image datasets demonstrate the effectiveness of SSConv, and our network attains superior dehazing performance. The experiments on practical applications also demonstrate that our method can significantly improve the 3-D object detection and 3-D layout performances for hazy omnidirectional images.

Method for Assessing Damage to Gas Distribution Network Pipelines based on Nonlinear Guided Wave

Method for Assessing Damage to Gas Distribution Network Pipelines based on Nonlinear Guided Wave

Analyzing the impact of blade geometrical parameters on energy recovery and efficiency of centrifugal pump as turbine installed in the pressure-reducing station

Improving renewable energy efficiency has received great attention globally by increasing demand for energy supplies and air pollution issues. The Soft Pressure Regulation System (SPRS) as a potential technology provides the possibility of regulating pressure and energy recovery in pressure-reducing stations. Pump as Turbine (PAT) is a good choice for installation in Water Distribution Network (WDN) pipelines due to its availability and economic benefits, and it can play a key role in decreasing startup time and return on investment. In this paper, the performance of centrifugal PAT under WDN conditions has been numerically and experimentally evaluated, and simulation validation was performed, too. Analyzing the work conditions of the pressure reduction station shows that the hourly flow rate trend changes considerably based on end-user demand. A major portion of the SPRS's operation time occurs in off-design conditions of the PAT, and efficiency is drastically decreased considering the lack of flow control tools. The effect of blade inlet angle, blade warp angle, blade curvature, and change of these parameters simultaneously was investigated to increase the high-efficiency working range. Losses resulting from flow separation and formation of vortices are significantly decreased by simultaneous change of blade geometrical parameters in the range of design point and higher flow rates (Q ≥ 1.0QBEP). Since the highest frequency of flow rate based on statistical analysis occurs in this range, the total power generation of SPRS improves significantly. The PAT efficiency with the improved impeller at part-load condition (0.8QBEP), design condition (QBEP), and over-load condition (1.2QBEP) is improved by 0.83 %, 2.69 %, and 6.71 % in comparison with the original impeller, respectively.

Robust Person Identification and Following in a Mobile Robot Based on Deep Learning and Optical Tracking

There is an exciting synergy between deep learning and robotics, combining the perception skills a deep learning system can achieve with the wide variety of physical responses a robot can perform. This article describes an embedded system integrated into a mobile robot capable of identifying and following a specific person reliably based on a convolutional neural network pipeline. In addition, the design incorporates an optical tracking system for supporting the inferences of the neural networks, allowing the determination of the position of a person using an RGB depth camera. The system runs on an NVIDIA Jetson TX2 board, an embedded System-on-Module capable of performing computationally demanding tasks onboard and handling the complexity needed to run a solid tracking and following algorithm. A robotic mobile base with the Jetson TX2 board attached receives velocity orders to move the system toward the target. The proposed approach has been validated on a mobile robotic platform that successfully follows a determined person, relying on the robustness of the combination of deep learning with optical tracking for working efficiently in a real environment.

Automated prediction of acute promyelocytic leukemia from flow cytometry data using a graph neural network pipeline.

Our study aimed to develop a machine learning (ML) model to accurately classify acute promyelocytic leukemia (APL) from other types of acute myeloid leukemia (other AML) using multicolor flow cytometry (MFC) data. Multicolor flow cytometry is used to determine immunophenotypes that serve as disease signatures for diagnosis. We used a data set of MFC files from 27 patients with APL and 41 patients with other AML, including those with uncommon immunophenotypes. Our ML pipeline involved training a graph neural network (GNN) to output graph-level labels and identifying the most crucial MFC parameters and cells for predictions using an input perturbation method. The top-performing GNN achieved 100% accuracy on the training/validation and test sets on classifying APL from other AML and used MFC parameters similarly to expert pathologists. Pipeline performance is amenable to use in a clinical decision support system, and our deep learning architecture readily enables prediction explanations. Our ML pipeline shows robust performance on predicting APL and could be used to screen for APL using MFC data. It also allowed for intuitive interrogation of the model's predictions by clinicians.

Accurate detection of weld seams for laser welding in real‐world manufacturing

AbstractWelding is a fabrication process used to join or fuse two mechanical parts. Modern welding machines have automated lasers that follow a predefined weld seam path between the two parts to create a bond. Previous efforts have used simple computer vision edge detectors to automatically detect the weld seam on an image at the junction of two metals to be welded. However, these systems lack reliability and accuracy resulting in manual human verification of the detected edges. This paper presents a neural network architecture that automatically detects the weld seam edge between two metals with high accuracy. We augment this system with a preclassifier that filters out anomalous workpieces (e.g., incorrect placement). Finally, we justify our design choices by evaluating against several existing deep network pipelines as well as proof through real‐world use. We also describe in detail the process of deploying the system in a real‐world shop floor including evaluation and monitoring. We make public a large well‐labeled laser seam dataset to perform deep learning‐based edge detection in industrial settings.