Conventional Model Research Articles

3D modeling of normal skin and cutaneous squamous cell carcinoma. A comparative study in 2D cultures, spheroids, and 3D bioprinted systems.

The current cancer research and drug testing are primarily based on 2D cell cultures and animal models. However, these methods have limitations and yield distinct drug response patterns. This study addressed this gap by developing an innovativein vitrohuman three-dimensional (3D) normal skin model and a multicellular model of human cutaneous squamous cell carcinoma (cSCC) using 3D bioprinting technology. Comparative analyzes were performed between bioprinted 3D-cSCC model, consisting of HaCaT keratinocytes, primary normal human dermal fibroblasts and A431 cancer cells (tricellular), bioprinted 3D-A431 model composed of A431 cancer cells only (monocellular), A431 cancer cell spheroids, and conventional 2D models. The models were structurally characterized by light microscopy, immunofluorescence (LIVE/DEAD assay, confocal microscopy) and immunohistochemistry (hematoxylin/eosin, p63, vimentin, Ki67, epidermal growth factor receptor stainings). The spatial arrangement of the 3D models was analyzed using the ARIVIS scientific image analysis platform. All models were also functionally assessed by cetuximab (CTX) response testing with the MTS assay. 3D-cSCC models were maintained for up to 16 weeks. Morphological and histological examinations confirmed the presence of skin-like layers in the bioprinted 3D models of normal skin, and the intricate and diverse features of the bioprinted skin cancer model, replicating the criticalin vivocharacteristics. In both mono- and tricellular bioprinted tumor constructs, there was a gradual formation and continuous growth of spheroid-like clusters of cancer cells, significantly influencing the morphology of the models. Cancer cells in the 3D bioprinted constructs showed reduced sensitivity to CTX compared to spheroids and 2D cultures. This study underscores the potential of 3D multicellular models in elucidating drug responses and gaining a better understanding the intricate interplay of cellular components within the tumor microenvironment. Developing the multicellular 3D tumor model paves the way for new research critical to advancing fundamental cancer research and future clinical applications, particularly drug response testing.

A proposed reconstruction method of a 3D animation scene based on a fuzzy long and short-term memory algorithm.

With the development of computer technology leading to a broad range of virtual technology implementations, the construction of virtual tasks has become highly demanded and has increased rapidly, especially in animation scenes. Constructing three-dimensional (3D) animation characters utilizing properties of actual characters could provide users with immersive experiences. However, a 3D face reconstruction (3DFR) utilizing a single image is a very demanding operation in computer graphics and vision. In addition, limited 3D face data sets reduce the performance improvement of the proposed approaches, causing a lack of robustness. When datasets are large, face recognition, transformation, and animation implementations are relatively practical. However, some reconstruction methods only consider the one-to-one processes without considering the correlations or differences in the input images, resulting in models lacking information related to face identity or being overly sensitive to face pose. A face model composed of a convolutional neural network (CNN) regresses 3D deformable model coefficients for 3DFR and alignment tasks. The manuscript proposes a reconstruction method for 3D animation scenes employing fuzzy LSMT-CNN (FLSMT-CNN). Multiple collected images are employed to reconstruct 3D animation characters. First, the serialized images are processed by the proposed method to extract the features of face parameters and then improve the conventional deformable face modeling (3DFDM). Afterward, the 3DFDM is utilized to reconstruct animation characters, and finally, high-precision reconstructions of 3D faces are achieved. The FLSMT-CNN has enhanced both the precision and strength of the reconstructed 3D animation characters, which provides more opportunities to be applied to other animation scenes.

Detection and classification of progressive supranuclear palsy from MRI images using deep learning

A timely and reliable computer aided MR image-based evaluations is required for the detection and classification of Progressive supranuclear palsy (PSP). It is a neurodegenerative ailment that is clinically very difficult to identify due to a high degree of overlap in characterized symptoms. Unlike other ailments, the primary constraint regarding PSP is the limited research work in this area. The main aim of our study is to establish a generalized model by comparing traditional custom CNN and transfer learning models such as DenseNet121, ResNet50, InceptionV3, VGG16, EfficientNetB0, Xception, MobileNet and InceptionResNetV2 which are fine tuned for the detection and classification of PSP with higher accuracy rates. Existing research has primarily focused on reducing the time complexity of neural networks and has only had success with low-level features. Furthermore, obtaining a significant volume of distributed labelled data is difficult. In our research, 125 T1 protocol based high resolution MRI images of 65 PSP and 60 normal control patients were considered. The image dataset is pre-processed, normalized and augmented before deploying them to the respective networks. The results propose that ‘InceptionResNetV2’ model can be considered as a generalised model for the detection and classification of PSP. The network offered a classification accuracy of 95%, Precision of 100%, Sensitivity of 92.8%, F1 score of 96.2% and a Specificity of 100% which is significantly higher compared to other models considered in the study and also the existing conventional ML models, thereby providing a prerequisite for significant diagnostic implementation.

A novel machine learning prediction model for metastasis in breast cancer.

Breast cancer (BC) metastasis is the common cause of high mortality. Conventional prognostic criteria cannot accurately predict the BC metastasis risk. The machine learning technologies can overcome the disadvantage of conventional models. We developed a model to predict BC metastasis using the random survival forest (RSF) method. Based on demographic data and routine clinical data, we used RSF-recursive feature elimination to identify the predictive variables and developed a model to predict metastasis using RSF method. The area under the receiver operating characteristic curve (AUROC) and Kaplan-Meier survival (KM) analyses were plotted to validate the predictive effect when C-index was plotted to assess the discrimination and Brier scores was plotted to assess the calibration of the predictive model. We developed a metastasis prediction model comprising three variables (pathological stage, aspartate aminotransferase, and neutrophil count) selected by RSF-recursive feature elimination. The model was reliable and stable when assessed by the AUROC (0.932 in training set and 0.905 in validation set) and KM survival analyses (p < .0001). The C-indexes (0.959) and Brier score (0.097) also validated the good predictive ability of this model. This model relies on routine data and examination indicators in real-time clinical practice and exhibits an accurate prediction performance without increasing the cost for patients. Using this model, clinicians can facilitate risk communication and provide precise and efficient individualized therapy to patients with breast cancer.

Performance reporting design in artificial intelligence studies using image-based TNM staging and prognostic parameters in rectal cancer: a systematic review.

The integration of artificial intelligence (AI) and magnetic resonance imaging in rectal cancer has the potential to enhance diagnostic accuracy by identifying subtle patterns and aiding tumor delineation and lymph node assessment. According to our systematic review focusing on convolutional neural networks, AI-driven tumor staging and the prediction of treatment response facilitate tailored treat-ment strategies for patients with rectal cancer. This paper summarizes the current landscape of AI in the imaging field of rectal cancer, emphasizing the performance reporting design based on the quality of the dataset, model performance, and external validation. AI-driven tumor segmentation has demonstrated promising results using various convolutional neural network models. AI-based predictions of staging and treatment response have exhibited potential as auxiliary tools for personalized treatment strategies. Some studies have indicated superior performance than conventional models in predicting microsatellite instability and KRAS status, offer-ing noninvasive and cost-effective alternatives for identifying genetic mutations. Image-based AI studies for rectal can-cer have shown acceptable diagnostic performance but face several challenges, including limited dataset sizes with standardized data, the need for multicenter studies, and the absence of oncologic relevance and external validation for clinical implantation. Overcoming these pitfalls and hurdles is essential for the feasible integration of AI models in clinical settings for rectal cancer, warranting further research.

Machine learning-based prediction of lithium-ion battery life cycle for capacity degradation modelling

With a particular focus on capacity degradation modeling, this paper offers a ground-breaking examination of the use of machine learning approaches for the precise prediction of lithium-ion battery life cycles. Because lithium-ion batteries are essential to many technological applications, it's critical to comprehend and anticipate their life cycles in order to maximize performance and guarantee sustainable energy solutions. The study starts with an extensive examination of the literature, assessing current approaches critically and setting the stage for the introduction of models based on machine learning. The process entails the methodical collecting of data across a range of operational settings, environmental variables, and charging-discharging cycles. Thorough preprocessing guarantees the dataset's consistency and quality for further machine learning model training. Predictive models are created using a variety of machine learning algorithms, including regression models, support vector machines, and deep neural networks. In order to improve prediction accuracy, the paper focuses on the reasoning behind model selection, parameter tuning, and the incorporation of ensemble approaches. In order to uncover important elements influencing the life cycles of lithium-ion batteries and provide important insights into degradation mechanisms, feature selection approaches are used. Using cross-validation techniques and real-world lithium-ion battery datasets, the built machine learning models go through rigorous evaluation and validation processes to determine their robustness, capacity for generalization, and performance metrics. Comparing machine learning-based predictions with conventional models, the results are presented and discussed, offering insights into the interpretability of the models and the identification of important affecting elements. In order to promote proactive maintenance and optimize battery usage, predictive models are integrated into real-time monitoring systems. The consequences for battery management systems are examined. The paper continues by discussing the challenges that come with using machine learning to estimate the life cycle of batteries and outlining possible directions for further research and development, such as scalability, interpretability, and the incorporation of emerging technologies. This research contributes to the ongoing efforts to increase the reliability and sustainability of lithium-ion battery technologies by highlighting the potential impact of machine learning on energy storage system optimization.

Adopting the metaverse in the luxury hotel business: a cost–benefit perspective

Purpose This study aims to examine the perceived values of the metaverse when adopting it in the luxury hospitality business. Based on the cost–benefit perspective, this research provides solid theoretical contributions and actionable managerial recommendations. Design/methodology/approach An exploratory sequential mixed-method design was used. For the qualitative phase, 21 hotel managers and 24 hotel guests (who often stay in four-star and five-star hotels and resorts) were interviewed after showing them a series of videos about using the metaverse in the hotel business. Based on the results of the qualitative phase, the analytic hierarchy process method was used, and 476 valid questionnaires were analyzed. Findings The results highlight the perceived benefits (personalized services, immersive experience and positive brand image) and costs (lack of human touch, time and effort and security and privacy) of metaverse adoption for hotel managers and their guests. In addition, the study determines the weight of each value attribute of metaverse adoption for each travel stage (pre-travel, during travel and post-travel). Practical implications Regarding metaverse adoption, the research offers practical suggestions for luxury hotels. For instance, the cost of equipment and the time and effort required are perceived costs of metaverse adoption. To address these challenges, hotels may offer free equipment (e.g. VR headsets) and training to their guests to stimulate the use of the metaverse. Originality/value This study addresses a gap in the literature by presenting a conceptual framework for examining metaverse adoption in the luxury hotel scenario. Unlike using conventional models like the technology acceptance model or the unified theory of acceptance and use of technology to investigate a technology’s adoption, this study stands out by unraveling the topic through the lens of value proposition. The latter often comes from an efficient value co-creation process, which is indeed shaped by an adequate appreciation of the congruence of perceived values (i.e. perceived benefits and costs) of metaverse from hotel manager and guest perspectives.

A deep learning approach for predicting the architecture of 3D textile fabrics

In this paper, a deep learning approach to 3D textile geometry simulations is presented. Two different network architectures with convolutional and recurrent properties are explored. The deep neural networks were trained to generate a fully compacted 3D textile unit cell based on the weave initial architecture. The AI training was conducted on a set of precomputed weaving case studies generated by digital element based weaving simulation software. The proposed strategy demonstrated effectiveness in estimation of 3D textile architectures. The designed system was able to operate within 10% error for stiffness properties prediction. The main benefit of the proposed approach over conventional modelling is its computational efficiency. Rapid weaving simulations provide an opportunity to explore the effects of different yarn architectures, matrix materials, and manufacturing techniques on the mechanical properties of woven composites, leading to a better understanding of their behaviour and their potential for use in new applications.

Research on the Application of the New Blended Teaching Model Based on Virtual Simulation Technology in Foreign Language Teaching

The reform of “Internet+” in higher education has been driven by the rapid development of information technology in recent years. While the blended teaching model has brought about greater openness in educational concepts and more humane teaching methods, its effectiveness in addressing students’ weak self-directed learning, lack of initiative, and insufficient practical abilities remains limited. In light of these challenges, this study thoroughly investigates the current status and primary problems of foreign language professional courses. The objective is to integrate virtual simulation technology with the traditional blended teaching model, creating a new blended teaching model for implementation in teaching practice. Two parallel classes from the 2020 English major at the researcher’s institution were selected as research subjects, with an experimental group and a control group established. The experimental group utilized the new blended teaching model based on virtual simulation technology, while the control group followed the conventional blended teaching model. After implementing the teaching methods, the effects were evaluated through two aspects: the scores of the final theoretical exams and the results of a survey. The experimental results reveal that the experimental group achieved higher scores in the theoretical exams compared to the control group, with a statistically significant difference (P&lt;0.05). Additionally, the students in the experimental group demonstrated notable improvements in their self-directed learning abilities, learning attitude, and practical application skills after being exposed to the new blended teaching model. These findings indicate that the online-offline blended teaching model, which incorporates virtual simulation experiments, effectively combines the advantages of blended teaching and virtual simulation technology. Furthermore, it successfully enhances students’ academic performance. Thus, it is evident that this new teaching model caters to the individual learning needs of students and exemplifies the integration of theory and practice.

Interaction between Pipe Rotation and Cuttings Transport in Extended-Reach Drilling: Mechanism, Model, and Applications

Summary High drag and torque on drillstring and difficulty in hole cleaning are two major challenges in extended-reach drilling, and these two challenges are usually coupled with each other. However, previous studies considered the drillstring mechanics and hole cleaning as two relatively independent issues and studied them separately, which cannot fully reveal the complex mechanisms of pipe sticking and obtain proper drilling parameters to ensure hole cleaning and reduce drag and torque. To solve this problem, in this paper we present the concept of pipe-cuttings interaction and elaborate on the two interaction mechanisms, which are the effect of pipe rotation on cuttings transport and the effect of cuttings distribution on pipe load. Second, we introduce the stirring diffusion factor and pipe-cuttings contact stress to quantify the pipe-cuttings interaction, and we obtain their mathematical expressions through nonlinear regression of numerical simulation results. Third, we establish the mechanical pipe-cuttings interaction model by modifying and combining the tubular mechanical model and the cuttings transport model. Finally, we validate the interaction model and apply it to a case study of extended-reach drilling in the South China Sea. The results show that the interaction model is of high prediction accuracy and is superior to the conventional tubular mechanical and cuttings transport models because the interaction between pipe rotation and cuttings transport is sufficiently considered. The problem of poor hole cleaning for a 12¼-in. section is serious, and pipe sticking occurs frequently for the extended-reach well in the case study because the stationary bed height is more than 10% in actual drilling. It is usually difficult to ensure hole cleaning of a 12¼-in. section by only optimizing a single parameter of rate of penetration (ROP), flow rate, or rotational speed, but optimizing two of the parameters at the same time can achieve better results. Lastly, we establish a recommended chart of drilling parameters, which can aid engineers in making comprehensive recommendations on drilling parameters.

Model Tests on Jacked Pile Penetration Characteristics Considering a Static Press-in Piling Machine

This study incorporates a static press-in piling machine into the conventional laboratory model tests for jacked piles. By conducting a comparative analysis between two tests, one involving the static press-in piling machine and the other focusing solely on pile jacking, this study aims to unveil the variations in penetration characteristics with pile sinking depth during the process of pile jacking under the constraint imposed by the static press-in piling machine. When considering the impact of the piling machine, the pile pressing force, pile sinking resistance, pile axial force, and unit side friction resistance of the pile body are higher compared to test results that only focus on pile jacking. There is an acceleration in the total side friction resistance within the depth range of 20 to 30 cm. Additionally, the reduction rate of axial force during the entire pile jacking process is 2% higher, with a general reduction in the “side resistance degradation” phenomenon. The soil pressure around the pile exhibits an initial increase followed by a decrease. The authors believe that the model box test of the jacked pile, considering the pile machine, would be more aligned with engineering practice.

LOCAL SELF-GOVERNMENT INSTITUTIONS AND CLIMATE CHANGE: THE CASE OF MEENANGADI PANCHAYAT IN KERALA

India faces significant vulnerability to the effects of climate change, attributed to its varied geography, substantial population, and reliance on agriculture. The nation encounters numerous climate-related challenges, including more frequent and severe extreme weather events like heatwaves, droughts, floods, and cyclones. Dealing with this intricate matter necessitates a comprehensive and cooperative strategy that extends beyond conventional top-down governance models. Local governments have a vital role in formulating and executing climate change adaptation strategies. They assess local vulnerabilities, identify priority areas, and formulate plans to build resilience and reduce risks. This can involve measures such as infrastructure improvements, land-use planning, water management, and public health initiatives tailored to the specific needs of the community. Local governments often collaborate with stakeholders, including businesses and residents, to achieve greenhouse gas reduction targets. India has pledged to become carbon neutral by the year 2070, and achieving carbon neutrality is a complex and multi-faceted endeavour that requires coordinated efforts across sectors and levels of government. The carbon-neutral Meenangadi in the Wayanad district of Kerala is a perfect example of successful local self-government taking measures to achieve carbon neutrality.

Detection of COVID-19, pneumonia, and tuberculosis from radiographs using AI-driven knowledge distillation

Chest radiography is an essential diagnostic tool for respiratory diseases such as COVID-19, pneumonia, and tuberculosis because it accurately depicts the structures of the chest. However, accurate detection of these diseases from radiographs is a complex task that requires the availability of medical imaging equipment and trained personnel. Conventional deep learning models offer a viable automated solution for this task. However, the high complexity of these models often poses a significant obstacle to their practical deployment within automated medical applications, including mobile apps, web apps, and cloud-based platforms. This study addresses and resolves this dilemma by reducing the complexity of neural networks using knowledge distillation techniques (KDT). The proposed technique trains a neural network on an extensive collection of chest X-ray images and propagates the knowledge to a smaller network capable of real-time detection. To create a comprehensive dataset, we have integrated three popular chest radiograph datasets with chest radiographs for COVID-19, pneumonia, and tuberculosis. Our experiments show that this knowledge distillation approach outperforms conventional deep learning methods in terms of computational complexity and performance for real-time respiratory disease detection. Specifically, our system achieves an impressive average accuracy of 0.97, precision of 0.94, and recall of 0.97.

Neural Network-Based Investigation of Periodic Noise Reduction Methods for High-Resolution Infrared Line Scanning Images

In the realm of neural network-based noise reduction, conventional models predominantly address Gaussian and blur artifacts across entire images. However, they encounter notable challenges when directly applied to periodic noise characteristics of high-resolution infrared sequential imagery. The high resolution also complicates the construction of suitable datasets. Our study introduces an innovative strategy that transforms two-dimensional images into one-dimensional signals, eliminating the need for processing the full image. We have developed a simulated dataset that closely mirrors natural infrared line scanning images derived from the FLIR dataset. To address low-frequency periodic noise, we propose two neural-network-based denoising approaches. The first employs a neural network to deduce noise from the one-dimensional signal, while the second utilizes discrete Fourier transforms for noise prediction within the frequency domain. Our experimental results highlight the Restormer model’s exemplary performance in direct noise prediction, where denoised images attain a PSNR of around 41 and an SSIM close to 0.9 on simulated data, leaving minimal residual noise in the actual denoised images. Further, we investigate the influence of Fourier coefficients, as predicted by neural networks, on the denoising process in the second approach. Employing 12 frequency coefficients, the Restormer and NAFNet models both reach a PSNR near 34 and an SSIM of approximately 0.842.

Advanced analytics for predicting traffic collision severity assessment

Accurate prediction of accident risks plays a crucial role in proactively implementing safety measures and allocating resources effectively. This paper introduces an innovative approach aimed at improving accident risk prediction by harnessing unique data sources and extracting insights from diverse yet sparse datasets. Traditional models often face limitations due to a lack of diversity and scope in the available data, which hinders their predictive capabilities. In response to this challenge, our study integrates a broad spectrum of heterogeneous data, encompassing traffic flow, weather conditions, road infrastructure details, and historical accident records. To overcome the difficulties associated with sparse data, we employ advanced data science techniques such as feature engineering, imputation, and machine learning. The paper introduces a novel dataset that amalgamates various data types, establishing a robust foundation for our predictive model. Through meticulous analysis, we derive valuable insights from these diverse sources, significantly enhancing our ability to assess accident risks. The proposed approach offers numerous advantages, including the capacity to predict accidents in areas that were previously underrepresented and under varying conditions. We rigorously evaluate the model's performance through extensive experimentation and validate its accuracy using real-world accident data. Our results indicate substantial improvements in prediction accuracy compared to conventional models. This research contributes significantly to the field of accident risk prediction by highlighting the potential benefits of integrating heterogeneous sparse data and leveraging advanced data science techniques. The study underscores the importance of tapping into novel data sources and extracting concealed patterns and insights to promote safety and optimize resource allocation in accident-prone regions, fostering more secure environments.

Using Convolutional Neural Networks for Blocking Prediction in Elastic Optical Networks

This paper presents a study on connection-blocking prediction in Elastic Optical Networks (EONs) using Convolutional Neural Networks (CNNs). In EONs, connections are established and torn down dynamically to fulfill the instantaneous requirements of the users. The dynamic allocation of the connections may cause spectrum fragmentation and lead to network performance degradation as connection blocking increases. Predicting potential blocking situations can be helpful during EON operations. For example, this prediction could be used in real networks to trigger proper spectrum defragmentation mechanisms at suitable moments, thereby enhancing network performance. Extensive simulations over the well-known NSFNET (National Science Foundation Network) backbone network topology were run by generating realistic traffic patterns. The obtained results are later used to train the developed machine learning models, which allow the prediction of connection-blocking events. Resource use was continuously monitored and recorded during the process. Two different Convolutional Neural Network models, a 1D CNN (One-Dimensional Convolutional Neural Network) and 2D CNN (Two-Dimensional Convolutional Neural Network), are proposed as the predicting methods, and their behavior is compared to other conventional models based on an SVM (Support Vector Machine) and KNN (K Nearest Neighbors). The results obtained show that the proposed 2D CNN predicts blocking with the best accuracy (92.17%), followed by the SVM, the proposed 1D CNN, and KNN. Results suggest that 2D CNN can be helpful in blocking prediction and might contribute to increasing the efficiency of future EON networks.

Deep learning-based radiomics model from pretreatment ADC to predict biochemical recurrence in advanced prostate cancer.

To develop deep-learning radiomics model for predicting biochemical recurrence (BCR) of advanced prostate cancer (PCa) based on pretreatment apparent diffusion coefficient (ADC) maps. Data were collected retrospectively from 131 patients diagnosed with advanced PCa, randomly divided into training (n = 93) and test (n = 38) datasets. Pre-treatment ADC images were segmented using a pre-trained artificial intelligence (AI) model to identify suspicious PCa areas. Three models were constructed, including a clinical model, a conventional radiomics model and a deep-radiomics model. The receiver operating characteristic (ROC), precision-recall (PR) curve and decision curve analysis (DCA) were used to assess predictive performance in test dataset. The net reclassification index (NRI) and integrated discrimination improvement (IDI) were employed to compare the performance enhancement of the deep-radiomics model in relation to the other two models. The deep-radiomics model exhibited a significantly higher area under the curve (AUC) of ROC than the other two (P = 0.033, 0.026), as well as PR curve (AUC difference 0.420, 0.432). The DCA curve demonstrated superior performance for the deep-radiomics model across all risk thresholds than the other two. Taking the clinical model as reference, the NRI and IDI was 0.508 and 0.679 for the deep-radiomics model with significant difference. Compared with the conventional radiomics model, the NRI and IDI was 0.149 and 0.164 for the deep-radiomics model without significant difference. The deep-radiomics model exhibits promising potential in predicting BCR in advanced PCa, compared to both the clinical model and the conventional radiomics model.

An ECG-based artificial intelligence model for assessment of sudden cardiac death risk.

Conventional ECG-based algorithms could contribute to sudden cardiac death (SCD) risk stratification but demonstrate moderate predictive capabilities. Deep learning (DL) models use the entire digital signal and could potentially improve predictive power. We aimed to train and validate a 12 lead ECG-based DL algorithm for SCD risk assessment. Out-of-hospital SCD cases were prospectively ascertained in the Portland, Oregon, metro area. A total of 1,827 pre- cardiac arrest 12 lead ECGs from 1,796 SCD cases were retrospectively collected and analyzed to develop an ECG-based DL model. External validation was performed in 714 ECGs from 714 SCD cases from Ventura County, CA. Two separate control group samples were obtained from 1342 ECGs taken from 1325 individuals of which at least 50% had established coronary artery disease. The DL model was compared with a previously validated conventional 6 variable ECG risk model. The DL model achieves an AUROC of 0.889 (95% CI 0.861-0.917) for the detection of SCD cases vs. controls in the internal held-out test dataset, and is successfully validated in external SCD cases with an AUROC of 0.820 (0.794-0.847). The DL model performs significantly better than the conventional ECG model that achieves an AUROC of 0.712 (0.668-0.756) in the internal and 0.743 (0.711-0.775) in the external cohort. An ECG-based DL model distinguishes SCD cases from controls with improved accuracy and performs better than a conventional ECG risk model. Further detailed investigation is warranted to evaluate how the DL model could contribute to improved SCD risk stratification.

Rock-mass heterogeneous rheological properties caused the formation of deep tension fractures

Tension failure is a unique phenomenon in solid Earth that occurs on scales ranging from large plate rift valleys to small laboratory rocks. On a slope scale, deep unloading tension fractures are distinct from conventional unloading fractures and are a unique geological phenomenon in valleys with high in-situ stress. To accurately reproduce the development and evolution of deep unloading tension fractures and to support major excavation projects, a series of works including geological investigation, laboratory tests, intrinsic model establishment, and numerical simulation were carried out in this study. The unloading rheological tests, accounting for time-dependent effects, uncover the heterogeneity in the rheological attributes of rock-mass strength parameters during valley downcutting. This heterogeneity manifests as a transition in rock-mass strength parameters that cohesion weakening-friction strengthening (CWFS). Drawing on the results of laboratory tests, a novel viscoelastic-plastic model, termed the WSR model, was proposed. This model takes into account both CWFS and rheological considerations. It has been applied to simulate deep unloading tension fractures in the Jinping I hydropower station (JP-I) and compared with conventional models. The results show that WSR model accurately reproduced the development and evolution of deep unloading tension fractures and the heterogeneity of rock-mass deformation during age evolution leads to the formation of deep unloading tension fractures. In this study, the rock-mass heterogeneous rheological properties were summarized as the heterogeneity of the rock-mass strength parameters during age deterioration and the heterogeneity of rock-mass deformation during age evolution; the WSR model was proposed to characterize the heterogeneous rheological property of rock-mass strength parameters and to reproduce the development and evolution of deep unloading tension fractures in the JP-I. This novel contribution to deep unloading tension fractures emphasizes that the rock-mass heterogeneous rheological properties lead to the formation of deep unloading tension fractures.

A PINN-based modelling approach for hydromechanical behaviour of unsaturated expansive soils

Hydromechanical behaviour of unsaturated expansive soils is complex, and current constitutive models failed to accurately reproduce it. Different from conventional modelling, this study proposes a novel physics-informed neural networks (PINN)-based model utilising long short-term memory as the baseline algorithm and incorporating a physical constraint (water retention) to modify the loss function. Firstly, a series of laboratory tests on Zaoyang expansive clay, including wetting and drying cycle tests and triaxial tests, was compiled into a dataset and subsequently fed into the PINN-based model. Subsequently, a specific equation representing the soil water retention curve (SWRC) for expansive clay was derived by accounting for the influence of the void ratio, which was integrated into the PINN-based model as a physical law. The ultimate predictions from the PINN-based model are compared with experimental data of unsaturated expansive clay with excellent agreement. This study demonstrates the capability of the proposed PINN in modelling the hydromechanical response of unsaturated soils and provides an innovative approach to establish constitutive models in the unsaturated soil mechanics field.