Architecture Model Research Articles

Advancing Kidney Transplantation: A Machine Learning Approach to Enhance Donor-Recipient Matching.

(1) Background: Globally, the kidney donor shortage has made the allocation process critical for patients awaiting a kidney transplant. Adopting Machine Learning (ML) models for donor-recipient matching can potentially improve kidney allocation processes when compared with traditional points-based systems. (2) Methods: This study developed an ML-based approach for donor-recipient matching. A comprehensive evaluation was conducted using ten widely used classifiers (logistic regression, decision tree, random forest, support vector machine, gradient boosting, boost, CatBoost, LightGBM, naive Bayes, and neural networks) across three experimental scenarios to ensure a robust approach. The first scenario used the original dataset, the second used a merged version of the dataset, and the last scenario used a hierarchical architecture model. Additionally, a custom ranking algorithm was designed to identify the most suitable recipients. Finally, the ML-based donor-recipient matching model was integrated into a web-based platform called Nephron. (3) Results: The gradient boost model was the top performer, achieving a remarkable and consistent accuracy rate of 98% across the three experimental scenarios. Furthermore, the custom ranking algorithm outperformed the conventional cosine and Jaccard similarity methods in identifying the most suitable recipients. Importantly, the platform not only facilitated efficient patient selection and prioritisation for kidney allocation but can be flexibly adapted for other solid organ allocation systems built on similar criteria. (4) Conclusions: This study proposes an ML-based approach to optimize donor-recipient matching within the kidney allocation process. Successful implementation of this methodology demonstrates significant potential to enhance both efficiency and fairness in kidney transplantation.

Personalized food consumption detection with deep learning and Inertial Measurement Unit sensor

For individuals diagnosed with diabetes mellitus, it is crucial to keep a record of the carbohydrates consumed during meals, as this should be done at least three times daily, amounting to an average of six meals. Unfortunately, many individuals tend to overlook this essential task. For those who use an artificial pancreas, carbohydrate intake proves to be a critical factor, as it can activate the insulin pump in the artificial pancreas to deliver insulin to the body. To address this need, we have developed personalized deep learning model that can accurately detect carbohydrate intake with a high degree of accuracy. Our study employed a publicly available dataset gathered by an Inertial Measurement Unit (IMU), which included accelerometer and gyroscope data. The data was sampled at a rate of 15 Hz, necessitating preprocessing. For our tailored to the patient model, we utilized a recurrent network comprising Long short-term memory (LSTM) layers. Our findings revealed a median F1 score of 0.99, indicating a high level of accuracy. Additionally, the confusion matrix displayed a difference of only 6 s, further validating the model’s accuracy. Therefore, we can confidently assert that our model architecture exhibits a high degree of accuracy. Our model performed well above 90% on the dataset, with most results between 98%–99%. The recurrent networks improved the problem-solving capabilities significantly, though some outliers remained. The model’s average prediction latency was 5.5 s, suggesting that later meal predictions result in extended meal progress predictions. The dataset’s limitation of mostly single-day data points raises questions about multi-day performance, which could be explored by collecting multi-day data, including night periods. Future enhancements might involve transformer networks and shorter time windows to improve model responsiveness and accuracy. Therefore, we can confidently assert that our model exhibits a high degree of accuracy.

Fractal Self-Similarity in Semantic Convergence: Gradient of Embedding Similarity across Transformer Layers

This paper presents a mathematical analysis of semantic convergence in transformer-based language models, drawing inspiration from the concept of fractal self-similarity. We introduce and prove a novel theorem characterizing the gradient of embedding similarity across layers. Specifically, we establish that there exists a monotonically increasing function that provides a lower bound on the rate at which the average cosine similarity between token embeddings at consecutive layers and the final layer increases. This establishes a fundamental property: semantic alignment of token representations consistently increases through the network, exhibiting a pattern of progressive refinement, analogous to fractal self-similarity. The key challenge addressed is the quantification and generalization of semantic convergence across diverse model architectures and input contexts. To validate our findings, we conduct experiments on BERT and DistilBERT models, analyzing embedding similarities for diverse input types. While our experiments are limited to these models, we empirically demonstrate consistent semantic convergence within these architectures. Quantitatively, we find that the average rates of semantic convergence are approximately 0.0826 for BERT and 0.1855 for DistilBERT. We observe that the rate of convergence varies based on token frequency and model depth, with rare words showing slightly higher similarities (differences of approximately 0.0167 for BERT and 0.0120 for DistilBERT). This work advances our understanding of transformer models’ internal mechanisms and provides a mathematical framework for comparing and optimizing model architectures.

Advances in Protein-Ligand Binding Affinity Prediction via Deep Learning: A Comprehensive Study of Datasets, Data Preprocessing Techniques, and Model Architectures.

Drug discovery is a complex and expensive procedure involving several timely and costly phases through which new potential pharmaceutical compounds must pass to get approved. One of these critical steps is the identification and optimization of lead compounds, which has been made more accessible by the introduction of computational methods, including deep learning (DL) techniques. Diverse DL model architectures have been put forward to learn the vast landscape of interaction between proteins and ligands and predict their affinity, helping in the identification of lead compounds. This survey fills a gap in previous research by comprehensively analyzing the most commonly used datasets and discussing their quality and limitations. It also offers a comprehensive classification of the most recent DL methods in the context of protein-ligand binding affinity prediction, providing a fresh perspective on this evolving field. We thoroughly examine commonly used datasets for BAP and their inherent characteristics. Our exploration extends to various preprocessing steps and DL techniques, including graph neural networks, convolutional neural networks, and transformers, which are found in the literature. We conducted extensive literature research to ensure that the most recent deep learning approaches for BAP were included by the time of writing this manuscript. The systematic approach used for the present study highlighted inherent challenges to BAP via DL, such as data quality, model interpretability, and explainability, and proposed considerations for future research directions. We present valuable insights to accelerate the development of more effective and reliable DL models for BAP within the research community. The present study can considerably enhance future research on predicting affinity between protein and ligand molecules, hence further improving the overall drug development process.

Higuchi fractal dimension and deep learning on near-infrared spectroscopy for determination of free fatty acid (FFA) content in oil palm fruit

Free fatty acid (FFA) content is an essential parameter with a significant influence on the quality of oil palm, with lower levels showing higher quality. The measurement of this parameter is often carried out with conventional methods in chemical laboratories using solvents and reagents, but these techniques have several limitations. Therefore, this study aimed to develop a rapid and non-destructive technique for determining FFA content based on near-infrared (NIR) spectroscopy. This technique comprised the use of Higuchi fractal dimension (HFD) and deep learning to process NIR spectra as a new chemometric analysis. The sample population consisted of 350 oil palm fruits at various maturity ages collected from the Cikabayan Oil Palm Plantation. Each sample had its NIR spectrum acquired with a wavelength of 1000–1500 nm. Good performance was shown by a low root mean squared error (RMSE) value and a high coefficient of determination (R2). Based on numerical analysis, this study proposed HFD with a maximum discrete time interval (kmax) value of 10 and deep learning with long short-term memory (LSTM) as a robust architectural model. The results showed that the proposed technique could predict oil palm fruit FFA levels with RMSE and R2 values of 0.167 and 0.959, respectively. Furthermore, it could identify FFA content at each process step, before and after harvesting. Based on these results, its implementation in oil palm control management was expected to enhance effectiveness and efficiency, leading to the production of high-quality products.

Deep Learning Methods for Chest X-Ray Imaging-Based COVID-19 Pneumonia Detection

The COVID-19 pandemic currently underway has highlighted a need for fast and accurate screening tools to help identify the disease, particularly in resource-poor settings. In this paper, a deep convolutional neural network (CNN) model is proposed for the automatic detection of COVID-19 pneumonia using chest X-ray images that helps in reducing the cost and processing time compared to traditional testing procedures. It uses several deep-structured medical image data types and improvement of the deep learning model architecture to achieve good diagnostic accuracy. Our hybrid CNN architecture integrates VGG-16 for feature extraction and ResNet-50 for pattern complexity assessment in detecting fine changes characteristic of COVID-19 pneumonia. Our dataset for this study is a collection of few thousands labelled chest X-ray images categorized in three classes, which are COVID-19, viral pneumonia and healthy cases. More advanced data preprocessing methods such as normalized, augmentation, and filtered noise has also been carried out which enhances in the model performance. We have high accuracy, recall, and good F1-score in the experimental results proving model robustness are applicable in real-world clinical scenarios. These research results speak to the importance of AI for improving diagnosis workflows, especially in fast-to-deploy large scale scenarios needed during a pandemic to cope with healthcare burdens.

Automated cardiac coverage assessment in cardiovascular magnetic resonance imaging using an explainable recurrent 3D dual-domain convolutional network.

Cardiovascular magnetic resonance (CMR) imaging has become a modality with superior power for the diagnosis and prognosis of cardiovascular diseases. One of the essential quality controls of CMR images is to investigate the complete cardiac coverage, which is necessary for the volumetric and functional assessment. This study examines the full cardiac coverage using a 3D dual-domain convolutional model and then improves this model using an innovative explainable salient region detection model and a recurrent architecture. Salient regions are extracted from the short-axis cine CMR stacks using a three-step proposed algorithm. Changing the architecture of the 3D dual-domain convolutional model to a recurrent one and taking advantage of the salient region detection model creates a kind of attention mechanism that leads to improved results. The results obtained from the images of over 6200 participants of the UK Biobank population cohort study show the superiority of the proposed model over the previous studies. The dataset is the largest regarding the number of participants to control the cardiac coverage. The accuracies of the proposed model in identifying the presence/absence of basal/apical slices are 96.22% and 95.42%, respectively. The proposed recurrent architecture of the 3D dual-domain convolutional model can force the model to focus on the most informative areas of the images using the extracted salient regions, which can help the model improve accuracy. The performance of the proposed fully automated model indicates that it can be used for image quality control in population cohort datasets and real-time post-imaging quality assessments. Codes are available at https://github.com/mohammadhashemii/CMR_Cardiac_Coverage_Control.

Study path analyses for quality assurance and support of study planning

AbstractUtilizing student lifecycle data provided by campus management systems yields the opportunity to conduct study path analyses. Methods of artificial intelligence (AI) and data science can be used to analyze study paths, identify indicators for success, and gain insights into problems and issues of student cohorts following different study paths. Meanwhile, AI can also be used to support students through informed study planning. This article presents the project AIStudyBuddy with its focus on utilizing rule-based AI and process mining to support study planning and cohort monitoring. The concept of a reference architecture and data model for study path analytics as well as details on the development of the two user applications, StudyBuddy for students and BuddyAnalytics for study program designers, are presented. By exploring how AI and process mining can be applied in the scope of the two applications, the article addresses the question of how AI can be used for quality assurance in study planning and student cohort monitoring.

E-WebGuard: Enhanced neural architectures for precision web attack detection

Web applications have become a favored tool for organizations to disseminate vast amounts of information to the public. With the increasing adoption and inherent openness of these applications, there is an observed surge in web-based attacks exploited by adversaries. However, most of the web attack detection works are based on public datasets that are outdated or do not cover a sufficient quantity of web application attacks. Furthermore, most of them are binary detection (i.e., normal or attack) and there is little work on multi-class web attack detection. This highlights the crucial need for automated web attack detection models to bolster web security. In this study, a suite of integrated machine learning and deep learning models is designed to detect web attacks. Specifically, this study employs the Character-level Support Vector Machine (Char-SVM), Character-level Long Short-Term Memory (Char-LSTM), Convolutional Neural Network - SVM (CNN-SVM), and CNN-Bi-LSTM models to differentiate between standard HTTP requests and HTTP-based attacks in both the CSIC 2010 and SR-BH 2020 datasets. Note that the CSIC 2010 dataset involves binary classification, while the SR-BH 2020 dataset involves multi-class classification, specifically with 13 classes. Notably, the input data is first converted to the character level before being fed into any of the proposed model architectures. In the binary classification task, the Char-SVM model with a linear kernel outperforms other models, achieving an accuracy rate of 99.60%. The CNN-Bi-LSTM model closely follows with a 99.41% accuracy, surpassing the performance of the CNN-LSTM model presented in previous research. In the context of multi-class classification, the CNN-Bi-LSTM model demonstrates outstanding performance with a 99.63% accuracy rate. Furthermore, the multi-class classification models, namely Char-LSTM and CNN-Bi-LSTM, achieve validation accuracies above 98%, outperforming the two machine learning-based methods mentioned in the original research.

BANDRP: a bilinear attention network for anti-cancer drug response prediction based on fingerprint and multi-omics.

Predicting anti-cancer drug response can help with personalized cancer treatment and is an important topic in modern oncology research. Although some methods have been used for anti-cancer drug response prediction, how to effectively integrate various features related to cancer cell lines, drugs, and their known responses is still affected by the redundant information of input features and the complex interactions between features. In this study, we propose a bilinear attention model, named BANDRP, based on multiple omics data of cancer cell lines and multiple molecular fingerprints of drugs to predict potential anti-cancer drug responses. Compared with existing models, BANDRP uses gene expression data to calculate pathway enrichment scores to enrich the features of cancer cell lines and can automatically learn the interactive information of cancer cell lines and drugs through bilinear attention networks. Benchmarking and independent tests demonstrate that BANDRP surpasses baseline models and exhibits robust generalization performance. Ablation experiments affirm the optimality of the current model architecture and feature selection scheme for our prediction task. Furthermore, analytical experiments and case studies on unknown anti-cancer drug response predictions underscore BANDRP's potential as a potent and reliable framework for predicting anti-cancer drug response.

An initial game-theoretic assessment of enhanced tissue preparation and imaging protocols for improved deep learning inference of spatial transcriptomics from tissue morphology.

The application of deep learning to spatial transcriptomics (ST) can reveal relationships between gene expression and tissue architecture. Prior work has demonstrated that inferring gene expression from tissue histomorphology can discern these spatial molecular markers to enable population scale studies, reducing the fiscal barriers associated with large-scale spatial profiling. However, while most improvements in algorithmic performance have focused on improving model architectures, little is known about how the quality of tissue preparation and imaging can affect deep learning model training for spatial inference from morphology and its potential for widespread clinical adoption. Prior studies for ST inference from histology typically utilize manually stained frozen sections with imaging on non-clinical grade scanners. Training such models on ST cohorts is also costly. We hypothesize that adopting tissue processing and imaging practices that mirror standards for clinical implementation (permanent sections, automated tissue staining, and clinical grade scanning) can significantly improve model performance. An enhanced specimen processing and imaging protocol was developed for deep learning-based ST inference from morphology. This protocol featured the Visium CytAssist assay to permit automated hematoxylin and eosin staining (e.g. Leica Bond), 40×-resolution imaging, and joining of multiple patients' tissue sections per capture area prior to ST profiling. Using a cohort of 13 pathologic T Stage-III stage colorectal cancer patients, we compared the performance of models trained on slide prepared using enhanced versus traditional (i.e. manual staining and low-resolution imaging) protocols. Leveraging Inceptionv3 neural networks, we predicted gene expression across serial, histologically-matched tissue sections using whole slide images (WSI) from both protocols. The data Shapley was used to quantify and compare marginal performance gains on a patient-by-patient basis attributed to using the enhanced protocol versus the actual costs of spatial profiling. Findings indicate that training and validating on WSI acquired through the enhanced protocol as opposed to the traditional method resulted in improved performance at lower fiscal cost. In the realm of ST, the enhancement of deep learning architectures frequently captures the spotlight; however, the significance of specimen processing and imaging is often understated. This research, informed through a game-theoretic lens, underscores the substantial impact that specimen preparation/imaging can have on spatial transcriptomic inference from morphology. It is essential to integrate such optimized processing protocols to facilitate the identification of prognostic markers at a larger scale.

M6ATM: a deep learning framework for demystifying the m6A epitranscriptome with Nanopore long-read RNA-seq data.

N6-methyladenosine (m6A) is one of the most abundant and well-known modifications in messenger RNAs since its discovery in the 1970s. Recent studies have demonstrated that m6A is involved in various biological processes, such as alternative splicing and RNA degradation, playing an important role in a variety of diseases. To better understand the role of m6A, transcriptome-wide m6A profiling data are indispensable. In recent years, the Oxford Nanopore Technology Direct RNA Sequencing (DRS) platform has shown promise for RNA modification detection based on current disruptions measured in transcripts. However, decoding current intensity data into modification profiles remains a challenging task. Here, we introduce the m6A Transcriptome-wide Mapper (m6ATM), a novel Python-based computational pipeline that applies deep neural networks to predict m6A sites at a single-base resolution using DRS data. The m6ATM model architecture incorporates a WaveNet encoder and a dual-stream multiple-instance learning model to extract features from specific target sites and characterize the m6A epitranscriptome. For validation, m6ATM achieved an accuracy of 80% to 98% across in vitro transcription datasets containing varying m6A modification ratios and outperformed other tools in benchmarking with human cell line data. Moreover, we demonstrated the versatility of m6ATM in providing reliable stoichiometric information and used it to pinpoint PEG10 as a potential m6A target transcript in liver cancer cells. In conclusion, m6ATM is a high-performance m6A detection tool, and our results pave the way for future advancements in epitranscriptomic research.

Optimizing IOL Calculators with Deep Learning Prediction of Total Corneal Astigmatism.

Background/Objectives: This study aims to identify the most accurate regression model for predicting total corneal astigmatism (TCA) from anterior corneal astigmatism (ACA) and to fine-tune the best model's architecture to further optimize predictive accuracy. Methods: A retrospective review of 19,468 eyes screened for refractive surgery was conducted using electronic medical records. Corneal topography data were acquired using the Pentacam HR. Various types (7) and subtypes (21) of regression learners were tested, with a deep neural network (DNN) emerging as the most suitable. The DNN was further refined, experimenting with 23 different architectures. Model performance was evaluated using root mean squared error (RMSE), R2, average residual error, and circular error. The final model only used age, ACA magnitude, and ACA axis to predict TCA magnitude and axis. Results were compared to predictions from one of the leading TCA prediction formulas. Results: Our model achieved higher performance for TCA magnitude prediction (R2 = 0.9740, RMSE = 0.0963 D, and average residual error = 0.0733 D) compared to the leading formula (R2 = 0.8590, RMSE = 0.2257 D, and average residual error = 0.1928 D). Axis prediction error also improved by an average of 8.1° (average axis prediction error = 4.74° versus 12.8°). The deep learning approach consistently demonstrated smaller errors and tighter clustering around actual values compared to the traditional formula. Conclusion: Deep learning techniques significantly outperformed traditional methods for TCA prediction accuracy using the Pentacam HR. This approach may lead to more precise TCA calculations and better IOL selection, potentially enhancing surgical outcomes.

Energy evaluation for a dynamic and reconfigurable TWDM‐PON on a novel network traffic adaptable pattern

SummaryEven with existing methodologies, dynamic resource allocation, service inconsistency, operational costs, and excess energy/power consumption in optical access networks remain the key challenges for the next‐generation passive optical network (NG‐PON) design and development. This paper introduces a novel method for analyzing energy/power consumption in time and wavelength division multiplexed (TWDM) passive optical networks (PONs) with a ring‐based topology. We propose subscriber‐based channel allocation models (CAMs) for on‐demand services, allocate bandwidth dynamically, and evaluate their downstream power consumption. To determine the necessary optical devices for the various CAMs, we investigate subscriber take‐up rates by optimizing the utilization of optical components based on active subscribers. Implementing a novel Poisson distribution‐based network traffic model addresses energy‐related concerns and explores optimization techniques to minimize power and improve energy efficiency. With a take‐up rate of 64 active subscribers, simulation results show significant power consumption reductions ranging from 62.2 to 67.7%, achieved by the proposed subscriber‐based architecture models with varying subscriber counts, when compared with the conventional model. The model with the most subscribers (2048 subscribers) saves the most energy by reducing power consumption to 292 W from the conventional model's (2048 subscribers) 440 W. Furthermore, the model with the most subscribers saves 33.4% of energy, outperforming other subscriber‐based architecture models without affecting QoS even during peak hours of the day. These numerical results show that TWDM‐PONs with the most subscribers exhibit improved energy efficiency than the conventional and other proposed subscriber‐based architecture models.

Analysis of synergistic influence of multi-scale design parameters on nearly-zero energy office blocks performance based on architectural morphological classification and parametric modeling

Analysis of synergistic influence of multi-scale design parameters on nearly-zero energy office blocks performance based on architectural morphological classification and parametric modeling

Turbocharging protein binding site prediction with geometric attention, inter-resolution transfer learning, and homology-based augmentation

BackgroundLocating small molecule binding sites in target proteins, in the resolution of either pocket or residue, is critical in many drug-discovery scenarios. Since it is not always easy to find such binding sites using conventional methods, different deep learning methods to predict binding sites out of protein structures have been developed in recent years. The existing deep learning based methods have several limitations, including (1) the inefficiency of the CNN-only architecture, (2) loss of information due to excessive post-processing, and (3) the under-utilization of available data sources.MethodsWe present a new model architecture and training method that resolves the aforementioned problems. First, by layering geometric self-attention units on top of residue-level 3D CNN outputs, our model overcomes the problems of CNN-only architectures. Second, by configuring the fundamental units of computation as residues and pockets instead of voxels, our method reduced the information loss from post-processing. Lastly, by employing inter-resolution transfer learning and homology-based augmentation, our method maximizes the utilization of available data sources to a significant extent.ResultsThe proposed method significantly outperformed all state-of-the-art baselines regarding both resolutions—pocket and residue. An ablation study demonstrated the indispensability of our proposed architecture, as well as transfer learning and homology-based augmentation, for achieving optimal performance. We further scrutinized our model’s performance through a case study involving human serum albumin, which demonstrated our model’s superior capability in identifying multiple binding sites of the protein, outperforming the existing methods.ConclusionsWe believe that our contribution to the literature is twofold. Firstly, we introduce a novel computational method for binding site prediction with practical applications, substantiated by its strong performance across diverse benchmarks and case studies. Secondly, the innovative aspects in our method— specifically, the design of the model architecture, inter-resolution transfer learning, and homology-based augmentation—would serve as useful components for future work.

Analysis of Brain Age Gap across Subject Cohorts and Prediction Model Architectures.

Background: Brain age prediction from brain MRI scans and the resulting brain age gap (BAG)-the difference between predicted brain age and chronological age-is a general biomarker for a variety of neurological, psychiatric, and other diseases or disorders. Methods: This study examined the differences in BAG values derived from T1-weighted scans using five state-of-the-art deep learning model architectures previously used in the brain age literature: 2D/3D VGG, RelationNet, ResNet, and SFCN. The models were evaluated on healthy controls and cohorts with sleep apnea, diabetes, multiple sclerosis, Parkinson's disease, mild cognitive impairment, and Alzheimer's disease, employing rigorous statistical analysis, including repeated model training and linear mixed-effects models. Results: All five models consistently identified a statistically significant positive BAG for diabetes (ranging from 0.79 years with RelationNet to 2.13 years with SFCN), multiple sclerosis (2.67 years with 3D VGG to 4.24 years with 2D VGG), mild cognitive impairment (2.13 years with 2D VGG to 2.59 years with 3D VGG), and Alzheimer's dementia (5.54 years with ResNet to 6.48 years with SFCN). For Parkinson's disease, a statistically significant BAG increase was observed in all models except ResNet (1.30 years with 2D VGG to 2.59 years with 3D VGG). For sleep apnea, a statistically significant BAG increase was only detected with the SFCN model (1.59 years). Additionally, we observed a trend of decreasing BAG with increasing chronological age, which was more pronounced in diseased cohorts, particularly those with the largest BAG, such as multiple sclerosis (-0.34 to -0.2), mild cognitive impairment (-0.37 to -0.26), and Alzheimer's dementia (-0.66 to -0.47), compared to healthy controls (-0.18 to -0.1). Conclusions: Consistent with previous research, Alzheimer's dementia and multiple sclerosis exhibited the largest BAG across all models, with SFCN predicting the highest BAG overall. The negative BAG trend suggests a complex interplay of survival bias, disease progression, adaptation, and therapy that influences brain age prediction across the age spectrum.

Recent advances in the integration of protein mechanics and machine learning

Mechanics underlies protein properties and behavior. From a theoretical standpoint, it is possible to derive these based on physical rules. This is appealing because they provide insights into physiology and disease, as well as aid in protein engineering; however, the convoluted nature of the biological system and current computational speeds limit its feasibility. Machine learning (ML) architectures are known for their ability to make inferences on complex data, such as the relationship between protein mechanics, properties, and behavior. Substantial efforts have been made to learn such correlations in tasks such as the prediction of structure, stability, natural frequency, mechanical strength, folding rate, solubility, and function. Each of these properties is interconnected through protein mechanics, and it is not surprising that the methods used in these tasks overlap highly in model input and architecture. In this review, we evaluate ML methods for the seven aforementioned prediction tasks to identify current trends in ML research in the field of protein sciences, focusing on the input and model architecture of each method. A short overview of de novo protein design is also provided. Finally, we highlight trends in the application of ML methods in the field of protein science, as well as directions for future improvements.

The Strategic Adoption of 5G Technology From a Business Model Perspective

Even if 5G is considered an enabling technology, its transformative potential and widespread adoption remain limited. This paper explores how Mobile Network Operators (MNOs) innovate their business models by leveraging 5G. Through an exploratory multiple-case study of Italian MNOs, key determinants concerning innovation in value proposition, creation, and delivery are obtained inductively from interviews and grouped into different 5G business model archetypes. These latter are "Advanced connectivity provider", "Vertical Service Provider", and "Orchestrator". The first focuses on offering 5G as an enabling technology on which end users can build their own applications. The second selects target industries and allows access to a specific service. The third doesn’t own specific knowledge but it acts as the intermediary between a customer and a capable partner’s network, ensuring that the project is carried out and targeting multiple verticals. The research provides both theoretical and practical contributions: from a theoretical standpoint, it contributes to the business model innovation literature by offering empirical work that investigates the strategic adoption of a new technological paradigm focusing on the architecture of the business model rather than a measurement of performance afterwards. Moreover, it contributes to the 5G literature with an empirical study that focuses on the strategic implications of the phenomenon rather than just the technical aspects, and to the enabling technology literature by offering a study that examines an enabling technology from a firm-specific perspective. From the standpoint of practitioners, it can offer managerial guidance on developing suitable 5G business models to promote technology adoption and maintain competitiveness by harnessing the potential applications envisioned with 5G.

Machine learning-based approach for assessing the seismic vulnerability of reinforced concrete frame buildings

Recent artificial intelligence (AI) advancements have significantly impacted the field of structural engineering by offering new solutions to enhance safety, efficiency, and cost-effectiveness in the design, construction, and maintenance of infrastructure and urban areas. This study has investigated a machine learning-based approach to estimate the seismic vulnerability of reinforced concrete (RC) building frames. The study utilized a probabilistic technique to assess the simulation of a 4-story RC building frame in the face of epistemic uncertainty. Data was collected using the Monte-Carlo approach, which created a machine learning (ML) model for predicting structural damage. The problem was formulated as both a regression-based and a classification-based model. An artificial neural network (ANN) feed-forward network architecture model was used to solve the classification problems. The optimal number of neurons in the hidden layer was determined to produce the best estimation model. Three different models were combined for the regression model, including LASSO regression, random forest, and gradient boosting, by implementing the stacking generalization. The investigation results indicate that the stacked predicted model exhibits less variance than other ML models. The classification and regression-based algorithms can forecast damage states with a high degree of accuracy, ranging from 87 to 95 percent. In conclusion, the study has demonstrated the effectiveness of machine learning techniques for predicting the seismic vulnerability of RC building frames. With the continued advancement of AI and big data, these methods will likely play a crucial role in the structural engineering field.