Search

The electronic properties and optical response of ice and water are intricately shaped by their molecular structure, including the quantum mechanical nature of the hydrogen atoms. Despite numerous previous studies, a comprehensive understanding of the nuclear quantum effects (NQEs) on the electronic structure of water and ice at finite temperatures remains elusive. Here, we utilize molecular simulations that harness efficient machine-learning potentials and many-body perturbation theory to assess how NQEs impact the electronic bands of water and hexagonal ice. By comparing path-integral and classical simulations, we find that NQEs lead to a larger renormalization of the fundamental gap of ice, compared to that of water, ultimately yielding similar bandgaps in the two systems, consistent with experimental estimates. Our calculations suggest that the increased quantum mechanical delocalization of protons in ice, relative to water, is a key factor leading to the enhancement of NQEs on the electronic structure of ice.

Geosynthetic-reinforced and pile-supported (GRPS) embankments are becoming morepopular as a solution for addressing soil structuralinstability. The interaction between the geosyntheticpile-subsoil-embankment elements is crucial to theload transfer mechanism and performance of GRPSembankments. Several analytical models for GRPSembankment design have been proposed, but theirperformance and applicability still require further validation. This research presents a three-dimensional numerical investigation of the load transfer mechanism of GRPS embankments using the fnite diference approach, considering the combined interactionbetween the soil embankment, geosynthetics, pile,and subsoil. The importance of these crucial aspects in the GRPS embankment design technique is highlighted, as well as their infuence and sensitivity.The following elements, in descending order, infuence the load and settlement efcacies of the GRPSembankments: soft soil stifness, embankment height,geosynthetic stifness, and embankment soil density,according to this research. Furthermore, the use ofgeosynthetics reduces diferential settlements andmitigates soil yielding above the pile heads. Thenumerical fndings are then compared to four wellknown design standards, with the subsurface stifness, geosynthetic stifness, embankment height, andfll soil density all being varied simultaneously tomeasure their performance. The fndings of the comparison revealed that these techniques difer greatly intheir ability to forecast load efcacy and diferentialsettlement. Depending on the geometric propertiesof the embankment and material properties, all of theselected design methods produce over-predictions orunder-predictions., QC 20240829

This work presents a simplified method for the nonlinear analysis of the load–displacement response of piles in multi-layered soils. As a starting step, a new interface model based on the disturbed state concept (DSC) is put forth to simulate the interface shear stress-displacement relationship by considering the nonlinear hardening–softening behaviour. In the new model, input parameters can be conveniently calibrated using conventional interface shear tests or on-site tests. The good agreement between predictions and experimental data from interface direct shear tests validated the performance of the proposed DSC model. The DSC model performed better in terms of predictions when compared to the hyperbolic one. Next, the soil-structure interface model and bearing capacity theory are coupled to provide a theoretical framework for the analysis of pile load-transfer in saturated and unsaturated multi-layered soils, where the DSC model is employed to represent base resistance as well as skin friction. This work also discusses the profile of steady-state in-situ matric suction, soil–water characteristic curve, and pore-water pressure of unsaturated soils. The proposed method has the advantage of being used in practice as it is simple to obtain input parameters from laboratory tests, as well as Standard Penetration or Cone Penetration Tests. The proposed framework is finally applied to the analysis of five well-documented case studies. The proposed approach and the static load test results from the field measurements are found to be in satisfactory agreement, indicating that the proposed method performs well. The proposed method is suggested to be utilised for preliminary analysis, planning a suitable programme of loading tests, as well as optimizing the pile design by back analysis of the load test results., QC 20240617

High-speed passenger vessels, integrated river and sea vessels, container vessels, oil tankers, and other underwater vehicles operating in maritime traffic are among the types of vessels that must be equipped with AIS and VHF. The safety of navigation is one of the major problems in the maritime sector, particularly in Vietnam. Furthermore, marine traffic in the seaport zone is a common and difficult issue to manage in areas with a high volume of vessel traffic, mostly in places where the infrastructure supporting navigation is inadequately developed to meet the rapidly growing demands of the contemporary world. Therefore, it is necessary to create an integrated maritime management system to improve the efficiency of data exploitation and support maritime safety. To address this challenge, this study suggests a Maritime Traffic State Prediction (MTSP) model to predict traffic conditions in the channels where real-time data collection is insufficient in some specific locations. We recommend a deep learning method using Long Short-Term Memory (LSTM) networks to predict the safe path of the vessel in case of missing data segments. The findings have shown that the proposed approach encourages the mining of historical vessel data for maritime traffic, is ready to be applied, and can easily be implemented in a computer program or a web-based app.

In this work, we have proposed an approach for improving the GCN for predicting ratings in social networks. Our model is expanded from the standard model with several layers of transformer architecture. The main focus of the paper is on the encoder architecture for node embedding in the network. Using the embedding layer from the graph-based convolution layer, the attention mechanism could rearrange the feature space to get a more efficient embedding for the downstream task. The experiments showed that our proposed architecture achieves better performance than GCN on the traditional link prediction task.

Human resources in the context of the 4.0 industrial revolution play an important role, especially for youth-oriented organizations such as the Ho Chi Minh Communist Youth Union. The article presents and focuses on innovative content in terms of methods and content for the activities of the Ho Chi Minh Communist Youth Union in the context of digital transformation and the 4.0 industrial revolution following the requirements and tasks of improving the quality of young human resources.  Solutions to innovate the content and operating methods of the Ho Chi Minh Communist Youth Union focus on solutions to innovate the content and operating methods of the Union based on the capacity of young human resources. Keywords:  Young Human Resources, Digital Transformation, Innovation Methods, Operation Content And Operation Method Of Ho Chi Minh Communist Youth Union, Industrial Revolution 4.0.

BackgroundCancer biomarkers play a pivotal role in the diagnosis, prognosis, and treatment response prediction of the disease. In this study, we analyzed the expression levels of RhoB and DNp73 proteins in rectal cancer, as captured in immunohistochemical images, to predict the 5-year survival time of two patient groups: one with preoperative radiotherapy and one without.MethodsThe utilization of deep convolutional neural networks in medical research, particularly in clinical cancer studies, has been gaining substantial attention. This success primarily stems from their ability to extract intricate image features that prove invaluable in machine learning. Another innovative method for extracting features at multiple levels is the wavelet-scattering network. Our study combines the strengths of these two convolution-based approaches to robustly extract image features related to protein expression.ResultsThe efficacy of our approach was evaluated across various tissue types, including tumor, biopsy, metastasis, and adjacent normal tissue. Statistical assessments demonstrated exceptional performance across a range of metrics, including prediction accuracy, classification accuracy, precision, and the area under the receiver operating characteristic curve.ConclusionThese results underscore the potential of dual convolutional learning to assist clinical researchers in the timely validation and discovery of cancer biomarkers.

BackgroundCancer biomarkers play a pivotal role in the diagnosis, prognosis, and treatment response prediction of the disease. In this study, we analyzed the expression levels of RhoB and DNp73 proteins in rectal cancer, as captured in immunohistochemical images, to predict the 5-year survival time of two patient groups: one with preoperative radiotherapy and one without.MethodsThe utilization of deep convolutional neural networks in medical research, particularly in clinical cancer studies, has been gaining substantial attention. This success primarily stems from their ability to extract intricate image features that prove invaluable in machine learning. Another innovative method for extracting features at multiple levels is the wavelet-scattering network. Our study combines the strengths of these two convolution-based approaches to robustly extract image features related to protein expression.ResultsThe efficacy of our approach was evaluated across various tissue types, including tumor, biopsy, metastasis, and adjacent normal tissue. Statistical assessments demonstrated exceptional performance across a range of metrics, including prediction accuracy, classification accuracy, precision, and the area under the receiver operating characteristic curve.ConclusionThese results underscore the potential of dual convolutional learning to assist clinical researchers in the timely validation and discovery of cancer biomarkers.

BackgroundCancer biomarkers play a pivotal role in the diagnosis, prognosis, and treatment response prediction of the disease. In this study, we analyzed the expression levels of RhoB and DNp73 proteins in rectal cancer, as captured in immunohistochemical images, to predict the 5-year survival time of two patient groups: one with preoperative radiotherapy and one without.MethodsThe utilization of deep convolutional neural networks in medical research, particularly in clinical cancer studies, has been gaining substantial attention. This success primarily stems from their ability to extract intricate image features that prove invaluable in machine learning. Another innovative method for extracting features at multiple levels is the wavelet-scattering network. Our study combines the strengths of these two convolution-based approaches to robustly extract image features related to protein expression.ResultsThe efficacy of our approach was evaluated across various tissue types, including tumor, biopsy, metastasis, and adjacent normal tissue. Statistical assessments demonstrated exceptional performance across a range of metrics, including prediction accuracy, classification accuracy, precision, and the area under the receiver operating characteristic curve.ConclusionThese results underscore the potential of dual convolutional learning to assist clinical researchers in the timely validation and discovery of cancer biomarkers.

Geosynthetic-reinforced and pile-supported (GRPS) embankments are becoming more popular as a solution for addressing soil structural instability. The interaction between the geosynthetic-pile-subsoil-embankment elements is crucial to the load transfer mechanism and performance of GRPS embankments. Several analytical models for GRPS embankment design have been proposed, but their performance and applicability still require further validation. This research presents a three-dimensional numerical investigation of the load transfer mechanism of GRPS embankments using the finite difference approach, considering the combined interaction between the soil embankment, geosynthetics, pile, and subsoil. The importance of these crucial aspects in the GRPS embankment design technique is highlighted, as well as their influence and sensitivity. The following elements, in descending order, influence the load and settlement efficacies of the GRPS embankments: soft soil stiffness, embankment height, geosynthetic stiffness, and embankment soil density, according to this research. Furthermore, the use of geosynthetics reduces differential settlements and mitigates soil yielding above the pile heads. The numerical findings are then compared to four well-known design standards, with the subsurface stiffness, geosynthetic stiffness, embankment height, and fill soil density all being varied simultaneously to measure their performance. The findings of the comparison revealed that these techniques differ greatly in their ability to forecast load efficacy and differential settlement. Depending on the geometric properties of the embankment and material properties, all of the selected design methods produce over-predictions or under-predictions., QC 20240701

The effect of soil density on the soil–water characteristic curve (SWCC) is becoming a topic of universal interest due to the heterogeneity of soils and environmental variables. In this study, a simple and effective model based on the idea of translating the particle-size distribution curve to the SWCC is proposed for predicting SWCC change with initial density. There is only one new parameter introduced, and it is easily calibrated using two SWCCs obtained from test data. The SWCCs for the same soil at different initial void ratios can be estimated using the developed model. Several existing models are also thoroughly examined, with an emphasis on the advantages and disadvantages of each model. The validity of the proposed model was then verified by comparing it to three other models and experimental data for eight different types of soils. The proposed model also outperforms other existing models in this extensive study, providing good and consistent prediction performance across various soils. The proposed model is then applied to different engineering challenges involving the estimation of bearing resistance of unsaturated soils. Three typical examples chosen for illustration in this paper are the effect of soil density variation on unsaturated shear strength, bearing capacity of a shallow foundation, and ultimate bearing resistance of an energy pile. The findings of the investigation reveal that the proposed model can be utilized to solve a variety of problems involving soil–structure interaction in unsaturated soils., QC 20240701

The prediction of shear strength of unsaturated soils remains a significant challenge due to their complex multi-phase nature. In this paper, a review of prior experimental studies is first presented in order to outline important pieces of evidence, limitations and some design considerations. Then, an overview of existing shear strength equations is summarised, with a brief discussion. A micromechanical model with stress equilibrium conditions and multi-phase interaction considerations is presented to provide a new equation for predicting the shear strength of unsaturated soils. The validity of the proposed model is examined using published shear strength data for different soil types. The shear strength predicted by the analytical model was found to be in good agreement with the experimental data and to provide high performance in comparison with existing models. Evaluation of the results using two criteria – the average relative error and the normalised sum of squared error – proved the effectiveness and validity of the proposed equation. Using the proposed model, a non-linear relationship between shear strength, saturation degree, volumetric water content and matric suction was observed., QC 20240701

The design of geosynthetic-reinforced and pile-supported (GRPS) embankments is traditionally optimized by searching for the most cost-effective solution among several workable candidates. The candidates are usually based on experiences of engineers, and the real optimal design could be therefore missed. This paper intends to address the above-mentioned issue by systematically optimizing the design of GRPS embankments considering simultaneously the cost and the safety in the entire design space. It is thus a multi-objective optimization (MOO) problem that differs from the studies only focusing on minimizing the construction cost. A practical MOO procedure is proposed in this paper, and it is applied to an illustrative GRPS embankment case. A set of nondominated optimal designs (Pareto front) are obtained at first, allowing an informed design decision. Then, four candidates located on the Pareto front are highlighted. Each of them represents an attractive design: the safest, the least-cost, the best trade-off (knee point) considering the two objectives, and the cheapest one for a target safety requirement. Finally, the optimal design can be selected from these four candidates depending on specific project purposes. For the case study, the knee point design leads to improvements in both the two defined objectives (i.e., decreased cost and increased safety) compared to the initial design, showing great benefits of performing a MOO analysis. By using the procedure, the optimal designs are also efficiently determined for the cases of different embankment heights., QC 20240701

Confined fluids and electrolyte solutions in nanopores exhibit rich and surprising physics and chemistry that impact the mass transport and energy efficiency in many important natural systems and industrial applications. Existing theories often fail to predict the exotic effects observed in the narrowest of such pores, called single-digit nanopores (SDNs), which have diameters or conduit widths of less than 10 nm, and have only recently become accessible for experimental measurements. What SDNs reveal has been surprising, including a rapidly increasing number of examples such as extraordinarily fast water transport, distorted fluid-phase boundaries, strong ion-correlation and quantum effects, and dielectric anomalies that are not observed in larger pores. Exploiting these effects presents myriad opportunities in both basic and applied research that stand to impact a host of new technologies at the water-energy nexus, from new membranes for precise separations and water purification to new gas permeable materials for water electrolyzers and energy-storage devices. SDNs also present unique opportunities to achieve ultrasensitive and selective chemical sensing at the single-ion and single-molecule limit. In this review article, we summarize the progress on nanofluidics of SDNs, with a focus on the confinement effects that arise in these extremely narrow nanopores. The recent development of precision model systems, transformative experimental tools, and multiscale theories that have played enabling roles in advancing this frontier are reviewed. We also identify new knowledge gaps in our understanding of nanofluidic transport and provide an outlook for the future challenges and opportunities at this rapidly advancing frontier.

Background: Over a decade, tissues dissected adjacent to primary tumors have been considered "normal " or healthy samples (NATs). However, NATs have recently been discovered to be distinct from both tumorous and normal tissues. The ability to predict the survival rate of cancer patients using NATs can open a new door to selecting optimal treatments for cancer and discovering biomarkers. Methods: This paper introduces an artificial intelligence (AI) approach that uses NATs for predicting the 5-year survival of pre-operative radiotherapy patients with rectal cancer. The new approach combines pre-trained deep learning, nonlinear dynamics, and long short-term memory to classify immunohistochemical images of RhoB protein expression on NATs. Results: Ten-fold cross-validation results show 88% accuracy of prediction obtained from the new approach, which is also higher than those provided from baseline methods. Conclusion: Preliminary results not only add objective evidence to recent findings of NATs molecular characteristics using state-of-the-art AI methods, but also contribute to the discovery of RhoB expression on NATs in rectal-cancer patients. Clinical impact: The ability to predict the survival rate of cancer patients is extremely important for clinical decision-making. The proposed AI tool is promising for assisting oncologists in their treatments of rectal cancer patients.

Background: Over a decade, tissues dissected adjacent to primary tumors have been considered "normal " or healthy samples (NATs). However, NATs have recently been discovered to be distinct from both tumorous and normal tissues. The ability to predict the survival rate of cancer patients using NATs can open a new door to selecting optimal treatments for cancer and discovering biomarkers. Methods: This paper introduces an artificial intelligence (AI) approach that uses NATs for predicting the 5-year survival of pre-operative radiotherapy patients with rectal cancer. The new approach combines pre-trained deep learning, nonlinear dynamics, and long short-term memory to classify immunohistochemical images of RhoB protein expression on NATs. Results: Ten-fold cross-validation results show 88% accuracy of prediction obtained from the new approach, which is also higher than those provided from baseline methods. Conclusion: Preliminary results not only add objective evidence to recent findings of NATs molecular characteristics using state-of-the-art AI methods, but also contribute to the discovery of RhoB expression on NATs in rectal-cancer patients. Clinical impact: The ability to predict the survival rate of cancer patients is extremely important for clinical decision-making. The proposed AI tool is promising for assisting oncologists in their treatments of rectal cancer patients.

Background: Over a decade, tissues dissected adjacent to primary tumors have been considered "normal " or healthy samples (NATs). However, NATs have recently been discovered to be distinct from both tumorous and normal tissues. The ability to predict the survival rate of cancer patients using NATs can open a new door to selecting optimal treatments for cancer and discovering biomarkers. Methods: This paper introduces an artificial intelligence (AI) approach that uses NATs for predicting the 5-year survival of pre-operative radiotherapy patients with rectal cancer. The new approach combines pre-trained deep learning, nonlinear dynamics, and long short-term memory to classify immunohistochemical images of RhoB protein expression on NATs. Results: Ten-fold cross-validation results show 88% accuracy of prediction obtained from the new approach, which is also higher than those provided from baseline methods. Conclusion: Preliminary results not only add objective evidence to recent findings of NATs molecular characteristics using state-of-the-art AI methods, but also contribute to the discovery of RhoB expression on NATs in rectal-cancer patients. Clinical impact: The ability to predict the survival rate of cancer patients is extremely important for clinical decision-making. The proposed AI tool is promising for assisting oncologists in their treatments of rectal cancer patients.

Purpose: The Newly-Escaped-Poverty-List householders (NEPLH) in Vietnam served as the survey subjects for this study, which sought to determine the impact of "Personal Financial Literacy" on "The Probability of Using Lending Services of Vietnam Bank for Social Policies (P_VBSP)" and "The Probability of Using Informal Private Lending Services (P_ILC)". Design/methodology/approach: Survey data were collected from 186 respondents, by structured online questionnaire with convenient non-probability sampling technique. Aspects of personal financial literacy (FLi) such as personal financial knowledge (FL1), personal financial skills (FL2), personal financial attitude (FL3), and personal financial behavior (FL4) are measured. When respondents were divided into groups based on subgroup criteria (CVj), such as gender, age, marital status, role in the family, career, education level, religion, and number of years since being removed from the list of poor households, ANOVA techniques were used to examine differences in various aspects of personal financial literacy. The impact of FLi and subgroup variables (CVj) on P_VBSP and P_ILC was assessed using the binary logistic regression analysis technique. Findings: Research results show that there is a statistical difference in the average score of FL1, FL2, FL3 and FL4. The research findings also show that FL2, FL4, and "Career Status – CAR" have a positive impact on P_VBSP, while "Aged group - AGE", "Education level - EDU", and "Number of years since being removed from the list of poor households - YEAR" have a negative impact on P_VBSP. Meanwhile, the P_ILC is positive influenced by the AGE factor and moves in the opposite direction of FL2, FL4, EDU, and YEAR. Research, Practical & Social implications: The research findings provide the basis for governance implications as well as policy recommendations to promote financial literacy among the NEPLH in Vietnam, encouraging them access policy lending services from the Vietna

Accurate simulation of detector responses to hadrons is paramount for all physics programs at the Large Hadron Collider (LHC). Central to this simulation is the modeling of hadronic interactions. Unfortunately, the absence of first-principle theoretical guidance has made this a formidable challenge. The state-of-the-art simulation tool, \textsc{Geant4}, currently relies on phenomenology-inspired parametric models. Each model is designed to simulate hadronic interactions within specific energy ranges and for particular types of hadrons. Despite dedicated tuning efforts, these models sometimes fail to describe the data in certain physics processes accurately. Furthermore, fine-tuning these models with new measurements is laborious. Our research endeavors to leverage generative models to simulate hadronic interactions. While our ultimate goal is to train a generative model using experimental data, we have taken a crucial step by training conditional normalizing flow models with \textsc{Geant4} simulation data. Our work marks a significant stride toward developing a fully differentiable and data-driven model for hadronic interactions in High Energy and Nuclear Physics., Comment: 7 pages, 5 figures, Proceeding of 26th International Conference On Computing High Energy & Nuclear Physics (CHEP 2023)

BackgroundCancer biomarkers play a pivotal role in the diagnosis, prognosis, and treatment response prediction of the disease. In this study, we analyzed the expression levels of RhoB and DNp73 proteins in rectal cancer, as captured in immunohistochemical images, to predict the 5-year survival time of two patient groups: one with preoperative radiotherapy and one without.MethodsThe utilization of deep convolutional neural networks in medical research, particularly in clinical cancer studies, has been gaining substantial attention. This success primarily stems from their ability to extract intricate image features that prove invaluable in machine learning. Another innovative method for extracting features at multiple levels is the wavelet-scattering network. Our study combines the strengths of these two convolution-based approaches to robustly extract image features related to protein expression.ResultsThe efficacy of our approach was evaluated across various tissue types, including tumor, biopsy, metastasis, and adjacent normal tissue. Statistical assessments demonstrated exceptional performance across a range of metrics, including prediction accuracy, classification accuracy, precision, and the area under the receiver operating characteristic curve.ConclusionThese results underscore the potential of dual convolutional learning to assist clinical researchers in the timely validation and discovery of cancer biomarkers.

Association between RhoB protein expression and rectal cancer in radiotherapy (RT) resistance has been hypothesized. However, there is no strong clinical evidence to confirm the prognostic power of the protein in this disease. Here, we combine advanced artificial intelligence and signal processing methods to examine RhoB expression captured by immunohistochemical imaging of tumor tissue in a cohort of rectal cancer patients with preoperative RT. Prediction results obtained from the proposed approach with 10-fold cross-validation accuracy rates between 85% and 94% not only discover the potential role of RhoB for rectal cancer prognosis, but also significantly outperform individual pretrained deep-learning models with accuracy between 58% and 67%.

RhoB protein belongs to the Rho GTPase family, which plays an important role in governing cell signaling and tissue morphology. Its expression is known to have implications in pathologic processes of diseases. In particular, the role of RhoB in rectal cancer is not well understood. Investigation in the regulation and communication of this protein, detected by immunohistochemical staining on the mi-croscope, can help gain insightful information leading to optimal disease treatment options. Herein, deep learning-based image analysis and the decomposition of multiway arrays were used to study the predictive factor of RhoB in two cohorts of patients with rectal cancer having survival rates of <5 and >5 years. The results show distinctions between the tensor decomposition factors of the two cohorts. (Am J Pathol 2023, 193: 579-590; https://doi.org/10.1016/j.ajpath.2023.01.007)

Background: Over a decade, tissues dissected adjacent to primary tumors have been considered "normal " or healthy samples (NATs). However, NATs have recently been discovered to be distinct from both tumorous and normal tissues. The ability to predict the survival rate of cancer patients using NATs can open a new door to selecting optimal treatments for cancer and discovering biomarkers. Methods: This paper introduces an artificial intelligence (AI) approach that uses NATs for predicting the 5-year survival of pre-operative radiotherapy patients with rectal cancer. The new approach combines pre-trained deep learning, nonlinear dynamics, and long short-term memory to classify immunohistochemical images of RhoB protein expression on NATs. Results: Ten-fold cross-validation results show 88% accuracy of prediction obtained from the new approach, which is also higher than those provided from baseline methods. Conclusion: Preliminary results not only add objective evidence to recent findings of NATs molecular characteristics using state-of-the-art AI methods, but also contribute to the discovery of RhoB expression on NATs in rectal-cancer patients. Clinical impact: The ability to predict the survival rate of cancer patients is extremely important for clinical decision-making. The proposed AI tool is promising for assisting oncologists in their treatments of rectal cancer patients.

This study describes an approach for augmenting urban residential preference and hedonic house price models by incorporating Status-Quality Trade Off theory (SQTO). SQTO seeks explain the dynamic of urban structure using a multipolar, in which the location and strength of poles is driven by notions of residential status and dwelling quality. This paper presents in outline an approach for identifying status poles and for quantifying their effect on land and residential property prices. The results show how the incorporation of SQTO results in an enhanced understanding of variations in land / property process with increased spatial nuance. A number of future research areas are identified related to the status pole weights and the development of status pole index.

We introduce a novel variant of GNN for particle tracking called Hierarchical Graph Neural Network (HGNN). The architecture creates a set of higher-level representations which correspond to tracks and assigns spacepoints to these tracks, allowing disconnected spacepoints to be assigned to the same track, as well as multiple tracks to share the same spacepoint. We propose a novel learnable pooling algorithm called GMPool to generate these higher-level representations called "super-nodes", as well as a new loss function designed for tracking problems and HGNN specifically. On a standard tracking problem, we show that, compared with previous ML-based tracking algorithms, the HGNN has better tracking efficiency performance, better robustness against inefficient input graphs, and better convergence compared with traditional GNNs., Comment: 7 pages, 5 figures, submitted to the 21st International Workshop on Advanced Computing and Analysis Techniques in Physics Research

We introduce a novel variant of GNN for particle tracking called Hierarchical Graph Neural Network (HGNN). The architecture creates a set of higher-level representations which correspond to tracks and assigns spacepoints to these tracks, allowing disconnected spacepoints to be assigned to the same track, as well as multiple tracks to share the same spacepoint. We propose a novel learnable pooling algorithm called GMPool to generate these higher-level representations called "super-nodes", as well as a new loss function designed for tracking problems and HGNN specifically. On a standard tracking problem, we show that, compared with previous ML-based tracking algorithms, the HGNN has better tracking efficiency performance, better robustness against inefficient input graphs, and better convergence compared with traditional GNNs., Comment: 7 pages, 5 figures, submitted to the 21st International Workshop on Advanced Computing and Analysis Techniques in Physics Research

Homophily, the tendency of humans to attract each other when sharing similar features, traits, or opinions, has been identified as one of the main driving forces behind the formation of structured societies. Here we ask to what extent homophily can explain the formation of social groups, particularly their size distribution. We propose a spin-glass-inspired framework of self-assembly, where opinions are represented as multidimensional spins that dynamically self-assemble into groups; individuals within a group tend to share similar opinions (intragroup homophily), and opinions between individuals belonging to different groups tend to be different (intergroup heterophily). We compute the associated nontrivial phase diagram by solving a self-consistency equation for "magnetization" (combined average opinion). Below a critical temperature, there exist two stable phases: one ordered with nonzero magnetization and large clusters, the other disordered with zero magnetization and no clusters. The system exhibits a first-order transition to the disordered phase. We analytically derive the group-size distribution that successfully matches empirical group-size distributions from online communities.

The relationship between genotype and phenotype plays a crucial role in determining the function and robustness of biological systems. Here the evolution progresses through the change in genotype, whereas the selection is based on the phenotype, and the genotype-phenotype relation also evolves. The theory for such phenotypic evolution remains poorly developed, in contrast to evolution under the fitness landscape determined by genotypes. Here we provide a statistical-physics formulation of this problem by introducing replicas for genotype and phenotype. We apply it to an evolution model in which phenotypes are given by spin configurations; genotypes are an interaction matrix for spins to give the Hamiltonian, and the fitness depends only on the configuration of a subset of spins called the target. We describe the interplay between the genetic variations and phenotypic variances by noise in this model by our approach that extends the replica theory for spin glasses to include a spin replica for phenotypes and a coupling replica for genotypes. Within this framework we obtain a phase diagram of the evolved phenotypes against the noise and selection pressure, where each phase is distinguished by the fitness and overlaps for genotypes and phenotypes. Among the phases, a robust fitted phase, relevant to biological evolution, is achieved under the intermediate level of noise (temperature), where robustnesses to noise and to genetic mutation are correlated, as a result of replica symmetry. We also find a tradeoff between maintaining a high fitness level of phenotype and acquiring a robust pattern of genes as well as the dependence of this tradeoff on the ratio between the size of the functional (target) part to that of the remaining nonfunctional (non-target) one. The selection pressure needed to achieve high fitness increases with the fraction of target spins.

Background: Over a decade, tissues dissected adjacent to primary tumors have been considered "normal " or healthy samples (NATs). However, NATs have recently been discovered to be distinct from both tumorous and normal tissues. The ability to predict the survival rate of cancer patients using NATs can open a new door to selecting optimal treatments for cancer and discovering biomarkers. Methods: This paper introduces an artificial intelligence (AI) approach that uses NATs for predicting the 5-year survival of pre-operative radiotherapy patients with rectal cancer. The new approach combines pre-trained deep learning, nonlinear dynamics, and long short-term memory to classify immunohistochemical images of RhoB protein expression on NATs. Results: Ten-fold cross-validation results show 88% accuracy of prediction obtained from the new approach, which is also higher than those provided from baseline methods. Conclusion: Preliminary results not only add objective evidence to recent findings of NATs molecular characteristics using state-of-the-art AI methods, but also contribute to the discovery of RhoB expression on NATs in rectal-cancer patients. Clinical impact: The ability to predict the survival rate of cancer patients is extremely important for clinical decision-making. The proposed AI tool is promising for assisting oncologists in their treatments of rectal cancer patients.

Quick, early, and precise detection is important for diagnosis to control the spread of COVID-19 infection. Artificial Intelligence (AI) technology could certainly be used as a modulating tool to ease the detection, and help with the preventive steps further. Convolutional neural networks (CNNs) have achieved state-of-the-art performance in many visual recognition tasks. Nevertheless, most of these state-of-the-art networks highly rely on the availability of a high amount of labeled data, being an essential step in supervised machine learning tasks. Conventionally, this manual, mundane, and time-consuming process of annotating images is done by humans. Learning to localize or detect COVID-19 infection masks in our specific case study typically requires the collection of CT scan data that has been labeled with bounding boxes or similar annotations, which generally is limited. A technique that could perform such learning with much less annotations, and transfer the learned proposals that are algorithm-driven to generate more synthetic annotated samples would be helpful & quite valuable. We present such a technique inspired by weakly trained mask region based convolutional neural networks (R-CNN) architecture for localization, in which the number of images with their pixel-level masks can be a small proportion of the total dataset, and then further improvise CNNs by inversely generating dense annotations on-the-go using an algorithmic-based computational approach. We focus on alleviating the bottleneck associated with deep learning models needing annotated data for training in an intuitive reverse engineering fashion through this work. Our proposed solution can certainly provide the prospect of automated labeling on-the-fly, thereby reducing much of the manual work. As a result, one can quickly train a precise COVID-19 infection detector with the leverage of autonomous frame-by-frame machine generated annotations. The model achieved mean precision accuracy (%) of 0

BackgroundCancer biomarkers play a pivotal role in the diagnosis, prognosis, and treatment response prediction of the disease. In this study, we analyzed the expression levels of RhoB and DNp73 proteins in rectal cancer, as captured in immunohistochemical images, to predict the 5-year survival time of two patient groups: one with preoperative radiotherapy and one without.MethodsThe utilization of deep convolutional neural networks in medical research, particularly in clinical cancer studies, has been gaining substantial attention. This success primarily stems from their ability to extract intricate image features that prove invaluable in machine learning. Another innovative method for extracting features at multiple levels is the wavelet-scattering network. Our study combines the strengths of these two convolution-based approaches to robustly extract image features related to protein expression.ResultsThe efficacy of our approach was evaluated across various tissue types, including tumor, biopsy, metastasis, and adjacent normal tissue. Statistical assessments demonstrated exceptional performance across a range of metrics, including prediction accuracy, classification accuracy, precision, and the area under the receiver operating characteristic curve.ConclusionThese results underscore the potential of dual convolutional learning to assist clinical researchers in the timely validation and discovery of cancer biomarkers.

Aim: The process of biomarker discovery is being accelerated with the application of artificial intelligence (AI), including machine learning. Biomarkers of diseases are useful because they are indicators of pathogenesis or measures of responses to therapeutic treatments, and therefore, play a key role in new drug development. Proteins are among the candidates for biomarkers of rectal cancer, which need to be explored using state-of-the-art AI to be utilized for prediction, prognosis, and therapeutic treatment. This paper aims to investigate the predictive power of Ras homolog family member B (RhoB) protein in rectal cancer. Methods: This study introduces the integration of pretrained convolutional neural networks and support vector machines (SVMs) for classifying biopsy samples of immunohistochemical expression of protein RhoB in rectal-cancer patients to validate its biologic measure in biopsy. Features of the immunohistochemical expression images were extracted by the pretrained networks and used for binary classification by the SVMs into two groups of less and more than 5-year survival rates. Results: The fusion of neural search architecture network (NASNet)-Large for deep-layer feature extraction and classifier using SVMs provided the best average classification performance with a total accuracy = 85%, prediction of survival rate of more than 5 years = 90%, and prediction of survival rate of less than 5 years = 75%. Conclusions: The finding obtained from the use of AI reported in this study suggest that RhoB expression on rectal-cancer biopsy can be potentially used as a biomarker for predicting survival outcomes in rectal-cancer patients, which can be informative for clinical decision making if the patient would be recommended for preoperative therapy.

Homophily, the tendency of humans to attract each other when sharing similar features, traits, or opinions, has been identified as one of the main driving forces behind the formation of structured societies. Here we ask to what extent homophily can explain the formation of social groups, particularly their size distribution. We propose a spin-glass-inspired framework of self-assembly, where opinions are represented as multidimensional spins that dynamically self-assemble into groups; individuals within a group tend to share similar opinions (intragroup homophily), and opinions between individuals belonging to different groups tend to be different (intergroup heterophily). We compute the associated nontrivial phase diagram by solving a self-consistency equation for "magnetization" (combined average opinion). Below a critical temperature, there exist two stable phases: one ordered with nonzero magnetization and large clusters, the other disordered with zero magnetization and no clusters. The system exhibits a first-order transition to the disordered phase. We analytically derive the group-size distribution that successfully matches empirical group-size distributions from online communities.

The ability to rapidly develop materials with desired properties has a transformative impact on a broad range of emerging technologies. In this work, we introduce a new framework based on the diffusion model, a recent generative machine learning method to predict 3D structures of disordered materials from a target property. For demonstration, we apply the model to identify the atomic structures of amorphous carbons ($a$-C) as a representative material system from the target X-ray absorption near edge structure (XANES) spectra--a common experimental technique to probe atomic structures of materials. We show that conditional generation guided by XANES spectra reproduces key features of the target structures. Furthermore, we show that our model can steer the generative process to tailor atomic arrangements for a specific XANES spectrum. Finally, our generative model exhibits a remarkable scale-agnostic property, thereby enabling generation of realistic, large-scale structures through learning from a small-scale dataset (i.e., with small unit cells). Our work represents a significant stride in bridging the gap between materials characterization and atomic structure determination; in addition, it can be leveraged for materials discovery in exploring various material properties as targeted.

Zeolites are inorganic materials known for their diversity of applications, synthesis conditions, and resulting polymorphs. Although their synthesis is controlled both by inorganic and organic synthesis conditions, computational studies of zeolite synthesis have focused mostly on organic template design. In this work, we use a strong distance metric between crystal structures and machine learning (ML) to create inorganic synthesis maps in zeolites. Starting with 253 known zeolites, we show how the continuous distances between frameworks reproduce inorganic synthesis conditions from the literature without using labels such as building units. An unsupervised learning analysis shows that neighboring zeolites according to our metric often share similar inorganic synthesis conditions, even in template-based routes. In combination with ML classifiers, we find synthesis-structure relationships for 14 common inorganic conditions in zeolites, namely Al, B, Be, Ca, Co, F, Ga, Ge, K, Mg, Na, P, Si, and Zn. By explaining the model predictions, we demonstrate how (dis)similarities towards known structures can be used as features for the synthesis space. Finally, we show how these methods can be used to predict inorganic synthesis conditions for unrealized frameworks in hypothetical databases and interpret the outcomes by extracting local structural patterns from zeolites. In combination with template design, this work can accelerate the exploration of the space of synthesis conditions for zeolites.

The use of social robots as instruments for social mediation has been gaining traction in the field of Human-Robot Interaction (HRI). So far, the design of such robots and their behaviors is often driven by technological platforms and experimental setups in controlled laboratory environments. To address complex social relationships in the real world, it is crucial to consider the actual needs and consequences of the situations found therein. This includes understanding when a mediator is necessary, what specific role such a robot could play, and how it moderates human social dynamics. In this paper, we discuss six relevant roles for robotic mediators that we identified by investigating a collection of videos showing realistic group situations. We further discuss mediation behaviors and target measures to evaluate the success of such interventions. We hope that our findings can inspire future research on robot-assisted social mediation by highlighting a wider set of mediation applications than those found in prior studies. Specifically, we aim to inform the categorization and selection of interaction scenarios that reflect real situations, where a mediation robot can have a positive and meaningful impact on group dynamics.

Stream-based active learning (AL) strategies minimize the labeling effort by querying labels that improve the classifier’s performance the most. So far, these strategies neglect the fact that an oracle or expert requires time to provide a queried label. We show that existing AL methods deteriorate or even fail under the influence of such verification latency. The problem with these methods is that they estimate a label’s utility on the currently available labeled data. However, when this label would arrive, some of the current data may have gotten outdated and new labels have arrived. In this article, we propose to simulate the available data at the time when the label would arrive. Therefore, our method Forgetting and Simulating (FS) forgets outdated information and simulates the delayed labels to get more realistic utility estimates. We assume to know the label’s arrival date a priori and the classifier’s training data to be bounded by a sliding window. Our extensive experiments show that FS improves stream-based AL strategies in settings with both, constant and variable verification latency.

Unsaturated soil shear strength is a crucial and useful parameter for predicting geostructure stability, soil erosion, seasonal variation, and land management. Unsaturated shear strength measurement, however, is frequently costly, complex, and time-consuming. The main objective of this paper is to present a new generalized equation for the shear strength estimation of unsaturated soils. The proposed equation is derived from a micromechanical equilibrium condition considering the interaction of air, water, and solid phases. The particle contact area effect is taken into account in the proposed model, even though it is thought to be negligible in existing equations. In comparison with existing equations, the proposed one has the benefit of being able to capture the nonlinear influence of saturation degree and matric suction on unsaturated shear strength. The proposed equation is compared with several other existing equations as well as experimental data for four different soil types to verify its validity. The findings indicate that the proposed equation has a potential application in estimating unsaturated soil shear strength and that it outperforms existing equations. The stress state also has a substantial impact on the shear strength properties of unsaturated soils, which was extended to include in the proposed equation. The results reveal that the proposed equation is capable of accurately predicting the variation of the soil–water characteristic curve and unsaturated soil shear strength as a function of the stress state., QC 20240701

Shear strength of unsaturated soils is an important engineering property that is required for addressing geotechnical problems, the prediction of which remains to be a challenging task for design engineers due to the complex interaction problem. This study presents a new shear strength equation based on the micromechanical model and the disturbed state concept for unsaturated soils. The original point of this study is considering the solid contact area ratio which was neglected in most of the existing equations. Using the proposed model, the non-linear relationship between the matric suction, saturation degree, and the shear strength of unsaturated soils are described. Validation of the shear strength model was verified against the experimental data and several current models on six different types of soils. The results indicate that the proposed model has a good performance in predicting the shear strength of unsaturated soils, and generally is better than other existing models. In response to varying climatic conditions, the analytical model was then extended to consider the effect of temperature on the shear strength of unsaturated soils. The comparison between predicted and measured results was carried out on compacted silt for three different temperatures. The results show that the proposed model is capable of accurately predicting changes in unsaturated shear strength as a function of temperature., QC 20240701

Geosynthetic-reinforced and pile-supported (GRPS) systems have already proven their good performance in supporting embankments constructed over soft soil. The load transfer mechanism in GRPS embankments depends on the complex interaction between the soil in place, the structural elements and the embankment's soil type (cohesive or cohesionless). However, the cohesion influence of the embankment soil has not been well investigated as it is often not considered in the design of such systems. The main aim of this study is to present an analytical model for GRPS embankments that combines several phenomena such as the concentric arches model in cohesive fill soils, the hyperbolic model for the isochrone geogrid curve, and subsoil's consolidation. Three-dimensional numerical analyses are also conducted to evaluate the embankment soil cohesion influence on the soil arching. Both the numerical and analytical results agree that the cohesive embankment fills strengthen the soil arching effect and increase the efficacy if compared with cohesionless embankment fills. A comparison of the analytical model with measured data and other design methods for full-scale field tests proved the proposed model efficiency. The proposed analytical model therefore can be applicable for GRPS embankments with cohesive and non-cohesive fill soils., QC 20240701

This paper presents a probabilistic analysis of a geosynthetics-reinforced and pile-supported (GRPS) embankment by using reliability, sensitivity and uncertainty propagation techniques. Compared to the traditional deterministic evaluation methods, the analyses performed in this work allow designers to rationally consider the uncertainties of material properties and quantify their influences on the embankment performance. Additionally, complementary information/results could be obtained which enables an informed design decision for engineers. This feature is further enhanced in the paper by proposing a new analysis procedure in which four carefully selected probabilistic techniques are combined with an efficient deterministic model. The contribution of the proposed procedure lies in providing a variety of valuable results (e.g., failure probability and sensitivity index) related to the GRPS embankment performance but with a limited computational time. An application of the procedure is presented in the paper. The obtained results revealed that the studied GRPS embankments have relatively low failure probabilities considering a usual traffic loading but could be risky under extreme loading conditions. Based on the sensitivity analysis, all the random parameters are ranked according to their quantified importance. Some discussions are also provided, trying to link the procedure with practical designs and to explain the uncertainty modeling with more details., QC 20240701

In geotechnical projects involving soft soils, cement-mixed soils are an excellent way to improve the mechanical behaviour and engineering performance of soils. The water-to-cement ratio, cement content, and porosity are the most critical parameters impacting the strength of cement-mixed soils, but their effects are less well understood. The proportional influence of cement amount, porosity, and moisture content on cement-mixed soil strength is investigated in this research. The results showed that as the water-to-cement ratio increases, the influence of cementation and porosity on unconfined compressive strength decreases. The blended volume ratio is proposed as a new and adequate index for evaluating cement-mixed soil strength based on the experimental results. The blended volume ratio is the product of the void/cement volume ratio and the water/cement volume ratio. While the current index has not fully addressed the important effects of the water-to-cement ratio and porosity, the blended volume ratio allows describing the concurrent influences of porosity, cement amount, and water amount on the strength development. The experimental results showed that the cement-mixed soil strength is a function of the blended volume ratio. An empirical equation based on the blended volume ratio was then proposed for predicting the strength of cement-mixed soils. The validation of the proposed equation is verified by strength analysis for several different test data sets. A good agreement between measured and predicted results indicated that the proposed model is applicable to predict well the strength of cement-mixed soils., QC 20240701

Full Text, Thesis (PhD Doctorate), Doctor of Philosophy (PhD), School of Environment and Sc, Science, Environment, Engineering and Technology, Wide bandgap semiconductors are currently considered as the building blocks which revolutionized many areas of nanoscience and nanotechnologies. Among them, silicon carbide has been quickly emerging as one of the most interesting materials due to its remarkable physical, chemical, mechanical, electrical properties along with its biocompatibility and its feasibility for a wafer-scale production using standard MEMS processes. Inspired by these advantages, considerable number of studies have been devoted to the development of various SiC-based electronic devices over the past years. Despite these great achievements, however, it is realized that existing sensing devices usually rely on SiC in the forms of bulk or thin film, which still exhibit relatively poor sensitivity and unstable operation due to the limitation of their surface-to-volume ratios. Another key bottleneck of these existing sensing devices lies in the impossibility to integrate SiC onto soft and stretchable substrates due to its rigid and brittle nature, which significantly hamper the broader applications of SiC in the fast-emerging field of bioelectronics. Therefore, the research activities in the framework of my Ph.D. candidature aimed to establish reliable and sustainable protocols to overcome these two critical challenges. The focuses of my research work are placed on the fabrication of SiC nanowires on SiO2 insulators and stretchable SiC microstructures integrated onto polymeric substrates with a high level of structural perfection. Towards these goals, I have successfully fabricated Si NWs on insulator by using a combination of electron beam lithography and thermal wet oxidation techniques. The temperature sensing devices based on SiC NWs exhibit a high TCR compared to their bulk counterparts. On the other hand, I successfully established an effective protocol to transfer rigid SiC microstructures onto polymeric substrates and subsequently, enhanced their stretchability by taking advantages of stru

Rectal or colorectal cancer is one of the leading causes of cancer-related death. With the advancement in surgical techniques, the survival rate has been improved. Predicting the survival rate is an important factor for enabling optimal treatments to prolong rectal-cancer patients lives. Methods of artificial intelligence and machine learning have been applied for assisting physicians in cancer research. In this study, we investigated the use of pretrained convolutional neural networks and support vector machines for predicting the survival rate of a cohort of rectal-cancer patients using metastatic immunohistochemistry samples staining for protein RhoB. The combination of convolutional neural networks and support vector machines achieved better classification results than using individual pretrained deep networks in most cases, and where manual pathological analysis is encountered with great difficulty. In particular, the combination of ResNet-101 and SVM produced an average accuracy of 86% for non-radiotherapy, and Inception-v3 and SVM resulted in an average accuracy of 85% for radiotherapy.

Wide bandgap semiconductors are currently considered as the building blocks which revolutionized many areas of nanoscience and nanotechnologies. Among them, silicon carbide has been quickly emerging as one of the most interesting materials due to its remarkable physical, chemical, mechanical, electrical properties along with its biocompatibility and its feasibility for a wafer-scale production using standard MEMS processes. Inspired by these advantages, considerable number of studies have been devoted to the development of various SiC-based electronic devices over the past years. Despite these great achievements, however, it is realized that existing sensing devices usually rely on SiC in the forms of bulk or thin film, which still exhibit relatively poor sensitivity and unstable operation due to the limitation of their surface-to-volume ratios. Another key bottleneck of these existing sensing devices lies in the impossibility to integrate SiC onto soft and stretchable substrates due to its rigid and brittle nature, which significantly hamper the broader applications of SiC in the fast-emerging field of bioelectronics. Therefore, the research activities in the framework of my Ph.D. candidature aimed to establish reliable and sustainable protocols to overcome these two critical challenges. The focuses of my research work are placed on the fabrication of SiC nanowires on SiO2 insulators and stretchable SiC microstructures integrated onto polymeric substrates with a high level of structural perfection. Towards these goals, I have successfully fabricated Si NWs on insulator by using a combination of electron beam lithography and thermal wet oxidation techniques. The temperature sensing devices based on SiC NWs exhibit a high TCR compared to their bulk counterparts. On the other hand, I successfully established an effective protocol to transfer rigid SiC microstructures onto polymeric substrates and subsequently, enhanced their stretchability by taking advantages of stru, Thesis (PhD Doctorate), Doctor of Philosophy (PhD), School of Environment and Sc, Science, Environment, Engineering and Technology, Full Text

The complexity in the styles of 1200 Byzantine icons painted between 13th and 16th from Greece, Russia and Romania was investigated through the Kolmogorov algorithmic information theory. The aim was to identify specific quantitative patterns which define the key characteristics of the three different painting schools. Our novel approach using the artificial surface images generated with Inverse FFT and the Midpoint Displacement (MD) algorithms, was validated by comparison of results with eight fractal and non-fractal indices. From the analyzes performed, normalized Kolmogorov compression complexity (KC) proved to be the best solution because it had the best complexity pattern differentiations, is not sensitive to the image size and the least affected by noise. We conclude that normalized KC methodology does offer capability to differentiate the icons within a School and amongst the three Schools.

Geosynthetic-reinforced and pile-supported (GRPS) systems provide an economic and effective solution for embankments. The load transfer mechanisms are tridimensional ones and depend on the interaction between linked elements, such as piles, soil, and geosynthetics. This paper presents an extensive parametric study using three-dimensional numerical calculations for geosynthetic-reinforced and pile-supported embankments. The numerical analysis is conducted for both cohesive and non-cohesive embankment soils to emphasize the fill soil cohesion effect on the load and settlement efficacy of GRPS embankments. The influence of the embankment height, soft ground elastic modulus, improvement area ratio, geosynthetic tensile stiffness and fill soil properties are also investigated on the arching efficacy, GR membrane efficacy, differential settlement, geosynthetic tension, and settlement reduction performance. The numerical results indicated that the GRPS system shows a good performance for reducing the embankment settlements. The ratio of the embankment height to the pile spacing, subsoil stiffness, and fill soil properties are the most important design parameters to be considered in a GRPS design. The results also suggested that the fill soil cohesion strengthens the soil arching effect, and increases the loading efficacy. However, the soil arching mobilization is not necessarily at the peak state but could be reached at the critical state. Finally, the geosynthetic strains are not uniform along the geosynthetic, and the maximum geosynthetic strain occurs at the pile edge. The geosynthetic deformed shape is a curve that is closer to a circular shape than a parabolic one., QC 20240701

Geosynthetic-reinforced column-supported system is an economic and effective solution to support embankments constructed on soft soils. In this solution, both end-bearing and floating columns are commonly used in practice. For deep soil foundation depths, floating columns are more economical than end-bearing piles. The design of a floating column foundation involves complex soil–structure interactions and there are still no clear uniform guidelines available for the design of embankments supported by floating columns. The main focus of this paper is to present a design method for the geosynthetic-reinforced floating column-supported (GRFCs) embankments. The main features of the proposed method are combining the bearing capacity theory for the floating columns, the arching theory for fill soils, the tensioned membrane theory for the geosynthetic, and considering interaction models between geosynthetic, soil, and piles. Using the proposed method, the influences of the pile geometry, soft clay, geosynthetic, and embankment fills properties were investigated. It was observed that the geosynthetic membrane inclusion enhances the load transfer mechanism and reduces significantly the differential settlements of floating pile-supported embankments. The floating columns with a higher ultimate bearing capacity cause more soil arching. In general, the soil shear strength properties and column geometry (length, diameter, column spacing) have a strong influence on the GRFCs embankment behaviour. Finally, the proposed method is compared with the BS 8006–1 and EBGEO design standards considering several experimental and numerical models to investigate its validity. The results showed that the proposed method is able of very good prediction performance and allows conducting the design optimization of GRFCs embankment., QC 20240701