In the medical field, where clinical decision support system have a significant impact on vital medical decisions, there is an urgent need for transparent and secure artificial intelligence solutions. This research offers a framework that combines requirement engineering with explainable artificial intelligence methods concepts to improve clinical decision support system security and transparency. The framework uses concern separation goal modeling (Knowledge Acquisition in automated specification), stakeholder analysis (Use Case Modeling), and concern separation (Aspect-Oriented Requirement Engineering) to ensure that system explanations are aligned with stakeholder needs while addressing privacy, compliance, and safety requirements. The proposed approach is evaluated using a real-world medical dataset demonstrating improvements in explanation consistency, requirement alignment, and robustness under security constraints. These results highlight the potential of integrating Requirements Engineering with XAI to support secure, interpretable, and accountable AI-driven clinical decision-making.
BackgroundCheckpoint inhibitor pneumonitis (CIP) is a rare but potentially fatal immune-related adverse event (irAE) that can interrupt immune checkpoint blockade in non-small cell lung cancer (NSCLC). With no validated pretreatment biomarkers and a diagnosis largely made by exclusion, upfront risk stratification is required. Recent advances in artificial intelligence (AI)-driven pathomics have made it feasible to infer tumor immune microenvironment (TIME)-relevant risk states in patients with NSCLC. Accordingly, we leveraged hematoxylin and eosin(H&E)-based digital pathomics combined with clinical variables to interrogate the TIME in patients who developed CIP and to enable pretreatment and early prediction of CIP.MethodsIn this retrospective study, 346 eligible patients from three hospitals were screened consecutively between January 2022 and January 2025. Patients were divided into CIP and non-CIP groups according to whether CIP occurred at a prespecified observation endpoint. We first developed a pathomics model that employed convolutional neural networks (CNNs) combined with multi-instance learning (MIL) to generate predictions at both the patch and whole slide image (WSI) levels on H&E-stained slides. Separately, we constructed a clinical model using logistic regression (LR) to process the structured clinical data accompanying each case. Subsequently, pathological and clinical information were integrated, where modeling was advanced from modality-specific feature learning to cross-modal representation learning, and final predictive modeling was completed. The predictive performance of different models was evaluated using the area under the Receiver Operating Characteristic (ROC) curve and benchmarked against unimodal models and standard ensemble methods.ResultsWhen the models were evaluated across both internal validation and external test datasets, the pathomics model demonstrated noticeably stronger performance than the clinical approach, achieving area under the curve (AUC) scores of 0.916, 0.875(test 1), and 0.843(test 2), respectively, while the clinical model posted more modest results of 0.880, 0.569(test 1), and 0.594(test 2). The most significant outcome, however, emerged from the multimodal fusion model, which produced the strongest results of all, with performance metrics of 0.930, 0.919(test 1), and 0.905(test 2) in the validation and test phases, respectively.ConclusionPretreatment H&E-derived pathomics, integrated with baseline clinical biomarkers, enable accurate prediction of CIP risk in locally advanced or metastatic NSCLC. This framework supports proactive surveillance and individualized immune checkpoint inhibitor (ICI) strategies and provides a scalable route to decode TIME-relevant states from routine pathology.
Abstract Recent breakthroughs in artificial intelligence have revolutionized the automation of frame-field-driven quad mesh generation, a critical surface representation paradigm in computer-aided engineering. However, existing neural frame-field generation methods, limited by the orthogonality of fields, struggle to preserve the geometric fidelity as well as quad quality around sharp features. To address these limitations, we propose NeuralPoly, an intelligent non-orthogonal frame-field generation method. We design a poly-vector encoding of the non-orthogonal field to leverage the representation power of neural network in capturing geometric features without manual tuning. Furthermore, we introduce a Hessian-based neural weighting scheme that autonomously resolves ambiguous alignments in flat and spherical regions. We then incorporates the poly-vector encoding and the proposed weighting scheme into the loss functions of a unified neural network architecture consists of a SIREN module for neural implicit representation and a ResUNet module for field prediction. Finally, we compare our method with state-of-the-art techniques in field-guided quad mesh generation. Quantitative and qualitative evaluations demonstrate that our approach achieves superior performance in both geometric fidelity and quad mesh quality.
Dear Authors and Readers,The closing issue of “Zeszyty Teoretyczne Rachunkowości” (ZTR, “The Theoretical Journal of Accounting”) for 2025, vol. 49, number 4, once again provides an engaging and multidimensional review of contemporary research trends in accounting. This Special Issue, titled Contemporary challenges, conditions and directions of development of accounting, gathers 13 studies that explore the ongoing transformation of the accounting discipline driven by technological advancements, sustainability demands, and evolving expectations from professionals and educators. The featured articles reflect a diverse range of approaches, from theoretical modelling and comparative analysis to bibliometric synthesis and empirical evaluation, offering a comprehensive perspective on the accounting field as it advances into a new digital and regulatory era.At the intersection of behavioural finance and accounting communication, Adeel Ali Qureshi and Mateusz Lemańczyk present a comprehensive literature review in their paper Attention metrics and stock market reactions to accounting events: A literature review. By combining bibliometric analysis with the TCCM frame- work, they investigate how investor attention, measured by media coverage, online search activity, and textual complexity, influences market reactions to accounting disclosures. Their findings highlight the increasing significance of behavioural insights and data analytics in understanding how financial information is perceived, processed, and priced.The paper by Mateja Brozović, Sanja Sever Mališ, and Dominik Piršić, titled Financial accounting analysis of leverage and profitability: Evidence from Croatian SMEs, expands the discussion to corporate financial performance. Using key financial ratios from small and medium-sized enterprises in Croatia, the authors analyse the relationship between leverage and profitability, providing empirical evidence that enhances understanding of the financial resilience and risk structures of SMEs, a vital yet often overlooked segment of the European economy.Renáta Hornická and Renáta Pakšiová examine the development of non-financial disclosure in their paper Scope of sustainability reporting in the largest companies in Slovakia in 2017 and 2022. By analysing textual data from the annual and sustainability reports of major Slovak firms, they document a noticeable growth in the scope and depth of ESG reporting following the introduction of the Non-Financial Reporting Directive. Their findings offer timely insight into how regulatory pressure drives increased corporate accountability and the institutionalisation of sustainability reporting in Central and Eastern Europe.A broader institutional and regulatory perspective on sustainability assurance is examined by Tanja Laković, Daniel Zdolšek, and Milica Vukčević in their paper Development of the regulatory framework for sustainability assurance: A comparative analysis of the transition from NFRD to CSRD in Slovenia and Montenegro. This comparative study highlights the challenges and opportunities of implementing the new EU Corporate Sustainability Reporting Directive in Montenegro, a non-EU member state. It highlights differences in readiness and institutional adaptation between EU member and candidate countries.From a theoretical perspective, Serhii Lehenchuk and Viktoriia Makarovych offer an innovative conceptual discussion in Theoretical foundations of accounting for intellectual investment property: Towards standard setting. Their paper develops a framework for recognising and measuring intellectual investment property, bridging gaps between traditional accounting and emerging forms of intangible capital. By proposing theoretical principles for potential standardisation, the study adds a significant perspective to debates on accounting for knowledge-based assets in the digital economy.The linguistic and communicative aspects of accountability are examined in Raili Lilo, Elina Paemurru, and Ülle Pärl’s paper, Accountability through linguistic features: A holistic theoretical framework for sustainability reports. Through a meta- -analysis of previous empirical studies, the authors incorporate insights from legitimacy, stakeholder, signalling, and institutional theories to illustrate how language can both promote and conceal accountability in sustainability reporting. Their comprehensive framework offers a valuable basis for analysing how textual choices such as tone, clarity, and structure can influence stakeholders’ perceptions of corporate responsibility and transparency.The public sector perspective is presented by Diana Papradanova and Ventsislav Vechev in their paper An evaluation of the accounting model for reporting public sector entities’ revenues in Bulgaria in the context of the International Public Sector Accounting Standards. The authors carry out a detailed comparative analysis of Bulgarian regulations and IPSAS provisions, highlighting conceptual differences and gaps that impede transparency and comparability. Their findings offer practical recommendations for aligning public-sector accounting practices with international standards and fiscal accountability principles.The human factor and digital transformation in accounting are central themes in Katarzyna Prędkiewicz and Krzysztof Biegun’s article, Factors that influence accountants’ acceptance of Artificial Intelligence: An extended Technology Acceptance Model, which incorporates technology anxiety and experience. The authors empirically expand the Technology Acceptance Model by including variables related to technological anxiety and professional experience, offering fresh insights into how accountants view, accept, and adopt AI tools in their work. Their findings emphasise both the opportunities and psychological barriers in the move towards automation and intelligent systems in accounting practice.The contribution by Ana Rep Romić, Marzena Remlein, and Sanja Sever Mališ, titled Information technology in accounting education: A bibliometric-systematic literature review (2006–2025), focuses on the intersection of pedagogy and digitalisation. Drawing on a bibliometric and systematic literature review spanning two decades of research, the authors map global trends in the integration of IT into accounting education. Their study identifies emerging competencies, evolving educational technologies, and the changing role of educators in developing digitally literate accounting professionals capable of responding to sustainability and AI-driven challenges.Kristina Rudžionienė, Aušrinė Tamulevičiūtė, and Aurelija Kustienė’s study, The relationship between CSR and earnings management in Lithuanian listed companies, explores how sustainability efforts relate to financial behaviour in a small, transitional economy. Contrary to prior expectations, their results indicate a positive link between corporate social responsibility and both accrual- and real-activity earnings management. This surprising outcome suggests that, in some cases, CSR initiatives might be strategically used to hide opportunistic actions. The study offers new empirical insights into ethical authenticity and transparency in financial reporting across Central and Eastern Europe.The intersection of family business and accounting research is explored in Amin Soheili’s paper Family business and accounting research: A structured literature review. Through a systematic review of seventy peer-reviewed papers published between 2000 and 2024, the author maps the theoretical and methodological development of accounting research within family business contexts. Using a SWOT framework, the study highlights the underrepresentation of socioemotional and qualitative dimensions. The review advocates a broader investigation into private and emerging-market family firms, emphasising the need for interdisciplinary approaches that account for the behavioural and relational dynamics of family-owned enterprises.Gintarė Špogienė, Daiva Tamulevičienė, and Kristina Rudžionienė analyse five leading Lithuanian retail chains in their paper Integrating corporate social responsibility into internal decision-making in leading retail chains in Lithuania: A responsibility accounting perspectiveThey highlight a gap between publicly disclosed CSR and the information that genuinely influences managerial decisions. To reduce “informational noise” and enhance accountability, they suggest adapting responsibility accounting and reporting (RAR) to incorporate stakeholder-impact assessment and to categorise decisions as financial, philanthropic, or socially responsible, aligning internal controls with public CSR commitments and fostering more transparent, ethics-based governance.Finally, considering preparedness for the EU’s sustainability regime, Aleksandra Sulik-Górecka, Marzena Strojek-Filus, and Daniel Iskra, in their article Assessment of Polish companies’ preparedness for ESG reporting in the context of its determinants as evaluated by report preparers, explore Polish companies’ readiness through a nationwide survey and non-parametric inference. Most respondents rated themselves as only moderately prepared, with preparedness significantly linked to firm size (but not industry), about 70% viewing ESG reporting as complex, and they highlight a need for investment in personnel and reporting technologies. The study places these findings in the context of the roll-out of CSRD/ESRS and presents them as a baseline for more in-depth quality analysis.Taken together, the articles in this Special Issue reflect the complexity of modern accounting as a discipline that is simultaneously technological, behavioural, regulatory, and ethical. The contributions show how accounting continues to broaden beyond its traditional financial scope, including data analytics, artificial intelligence, linguistic transparency, and sustainability assurance. Each paper not only advances academic discussion but also provides valuable insights for practitioners, educators, and policymakers, enhancing the quality, relevance, and integrity of accounting information.The Editorial Team extends its gratitude to all authors and reviewers for their valuable contributions and diligent work in preparing this issue. We also thank our readers for their continued interest and engagement with the journal. We hope that the studies presented here will inspire further discussion, research, and innovation in the ever-evolving field of accounting.Marzena Remlein* Ana Rep Romić**The Editorial Team of ZTR is pleased to announce that in ZTR’s 49th year of publication, its four quarterly issues contained 39 articles: 25 in English and 14 in Polish. Their authors come from eleven countries (Bulgaria, Estonia, Croatia, Montenegro, Lithuania, Poland, the Czech Republic, Slovakia, Slovenia, Sweden, and Ukraine). We thank all the authors for their cooperation with the Editorial Team and the reviewers of their articles. The manuscripts submitted to ZTR were reviewed in 2025 by 73 reviewers, including 52 from Poland and 21 from abroad. The Editorial Team would like to thank all specialists who provided anonymous reviews and insightful feedback. The list of Polish and foreign reviewers is included in this issue of ZTR and on our journal’s website at https://ztr.skwp.pl/ cms/reviewers. We encourage authors and readers to visit ZTR’s website at https://ztr.skwp.pl/, which contains extensive information about ZTR, including its presence in databases (including Scopus, Web of Science, BazEkon, EBSCO Business Source Ulti-mate, Erich Plus, CEEOL, Cejsh, CROSSREF, DOAJ, and ICI Journals Master List), as well as an invitation to a thematic issue of ZTR in 2026 titled Accounting’s Expanded Horizon: Redefining Internal Practices for Organizational Flourishing (for more, see Call for papers published in ZTR, Vol. 49, No. 2 and at https://ztr.skwp.pl/cms/CMS:647). On behalf of the entire ZTR Editorial Team, I wish all authors, reviewers, members of the Editorial Board, and readers of ZTR a lot of health, happi-ness, and peace, as well as numerous professional successes in 2026. Yours sincerely,Anna Szychta
Safaa M Alsanosi,1 Asayel Q Aldajani,2 Hasnaa A Gheliwi,2 Manar M Alotibi,2 Ghadi S Bokhari,2 Orjuwan A Almatrafi,2 Abdulelah K Alqawlaq,3 Jakleen Z Abujamai,3 Mohammed Shaikhomer,4 Yosra Z Alhindi,1 Asim M Alshanberi3 1Department of Pharmacology and Toxicology, Faculty of Medicine, Umm Al Qura University, Makkah, Saudi Arabia; 2Faculty of Medicine, Umm Al-Qura University, Makkah, Saudi Arabia; 3General Medicine Program, Batterjee Medical College, Jeddah, Saudi Arabia; 4Department of Internal Medicine, Faculty of Medicine, King Abdulaziz University, Jeddah, Saudi ArabiaCorrespondence: Safaa M Alsanosi, Department of Pharmacology and Toxicology, Faculty of Medicine, Umm Al Qura University, Makkah, Saudi Arabia, Email smsanosi@uqu.edu.saBackground: Artificial intelligence (AI) is advancing healthcare globally and in Saudi Arabia, enhancing patient care, diagnostics, and administrative efficiency, despite challenges such as data privacy and regulation. This study explores knowledge, attitudes, and perceptions (KAP) regarding AI in medication adherence among chronic patients in Makkah region, Saudi Arabia.Methods: A cross-sectional study was conducted among patients with chronic diseases in the Makkah region, Saudi Arabia, from 1 July to 31 December 2024. The study included adult patients with chronic diseases (≥ 18 years) receiving primary care in the Makkah region. KAP levels were analyzed using descriptive statistics and composite scores, with demographic associations evaluated through Pearson chi-square tests (p< 0.05).Results: A total of 385 participants were included in the study. Most participants were women (60%), and those belonging to the 50 years or older group comprised the highest percentage (51.2%). The most reported chronic conditions were diabetes (30.7%), hypertension (19.7%), and asthma (14%). Knowledge levels were at a good level among 72.7% of the study participants, and 45.5% expressed a positive attitude towards AI’s role. Perception was high among 50.9% of the respondents but low among 23.4%. Demographic factors, particularly age, significantly improved KAP (p-values of 0.048, 0.046, and 0.031, respectively). A positive attitude towards AI’s role in medication adherence was observed in 58.2% of the participants with good knowledge levels compared to only 11.4% of those with poor knowledge (p=0.001). Variations in perception levels regarding AI’s role in medication adherence were evident across demographics, with statistically significant associations found for age and overall knowledge level (p-values of 0.031 and 0.001, respectively).Conclusion: The results highlight AI’s potential to enhance medication adherence and healthcare efficiency while maintaining a human-centred approach. To ensure effective integration, it’s crucial to address concerns related to privacy, trust, and reduced human interaction. AI should be positioned as a supportive tool that complements—not replaces—human care, with transparent governance and targeted education playing key roles.Keywords: knowledge, attitude, perception, artificial intelligence, medication adherence, chronic patients
With the rapid development of artificial intelligence (AI) technologies, AI-generated content (AIGC) on social media platforms has significantly increased. This study collected text data related to AIGC from mainstream social media platforms and employed the Latent Dirichlet Allocation (LDA) topic model to uncover the thematic characteristics of AIGC. The analysis was further integrated with the Unified Theory of Acceptance and Use of Technology (UTAUT) and Social Cognitive Theory (SCT) to identify seven key conditional variables: the maturity of AIGC technology, users’ perception of the authenticity of AIGC, users’ perception of the usefulness of AIGC, users’ perception of the entertainment value of AIGC, the commercialization level of AIGC, the personalization level of AIGC recommendations on the platform, and the ecosystem management and interaction atmosphere of AIGC on the platform. Using fuzzy-set Qualitative Comparative Analysis (fsQCA), this study identified seven configurational paths that drive user engagement with AIGC on social media platforms, which were ultimately summarized into three core pathways: user perception—platform recommendation pathway, user perception—platform atmosphere pathway, and technology characteristics—user perception—platform recommendation—platform atmosphere pathway. The results indicate that users’ perceptions of the usefulness of AIGC are a key factor in driving user engagement with AIGC on social media platforms.
Aneeza Alam, Ahmad Sami Al-Shamayleh, Nisrean Thalji
et al.
Abstract A bone fracture is a medical condition characterized by a partial or complete break in the continuity of the bone. Fractures are primarily caused by injuries and accidents, affecting millions of people worldwide. The healing process for a fracture can take anywhere from one month to one year, leading to significant economic and psychological challenges for patients. The detection of bone fractures is crucial, and radiographic images are often relied on for accurate assessment. An efficient neural network method is essential for the early detection and timely treatment of fractures. In this study, we propose a novel transfer learning-based approach called MobLG-Net for feature engineering purposes. Initially, the spatial features are extracted from bone X-ray images using a transfer model, MobileNet, and then input into a tree-based light gradient boosting machine (LGBM) model for the generation of class probability features. Several machine learning (ML) techniques are applied to the subsets of newly generated transfer features to compare the results. K-nearest neighbor (KNN), LGBM, logistic regression (LR), and random forest (RF) are implemented using the novel features with optimized hyperparameters. The LGBM and LR models trained on proposed MobLG-Net (MobileNet-LGBM) based features outperformed others, achieving an accuracy of 99% in predicting bone fractures. A cross-validation mechanism is used to evaluate the performance of each model. The proposed study can improve the detection of bone fractures using X-ray images.
Since the launching of ChatGPT (generative) AI has been developed so much and fast that it has entered higher education (HE) and higher education institutions (HEIs). The article is meant to help HE(Is) how to deal with AI strategically and in leadership. It investigates which influences AI and the use of AI tools is having on HE(Is). Therefore 4 research questions are formulated: how does AI and AI tools influence HE(Is) in its mission, organization and context; should AI and its applications then be regarded as an strategic objective or only as a tool to realize the strategy; how is AI and the use of AI tools, as developed and described in an AI strategy, best managed to be adopted and integrated in an effective and responsible way, and finally which influence does AI and its tools have on the leadership and culture? In order to answer those questions, the article describes first our contemporary times, and the leadership needed, then delves into the history of the development of AI and its tools and investigates the current and future attitudes towards, degrees of implementation, and uses of AI and its tools among the internal and external stakeholders of HE(Is). The findings result from a global literature study of international surveys and 2 case studies. The selection is based on topical usefulness, international scope, (statistical) relevance and quality of research in general. In this way the article aims to help to develop an AI strategy and thus can be read as a policy paper underpinned by a meta-analysis. The main results are that, although the use of AI in HEIs is divided, the effective and responsible adoption and integration of AI is a new strategic objective in order to help to realize HE’s three-fold mission in a well-planned and managed way asking for a visionary leadership and a clear policy framework and guidelines, in which the words transparency, responsibility and critical thinking link AI use with an enhancement of unique human competences such as critical thinking.
Neural-network-based models have made considerable progress in many computer vision areas over recent years. However, many works have exposed their vulnerability to malicious input data manipulation—that is, to adversarial attacks. Although many recent works have thoroughly examined the adversarial robustness of classifiers, the robustness of Image Quality Assessment (IQA) methods remains understudied. This paper addresses this gap by proposing FM-GOAT (Frequency-Masked Gradient Orthogonalization Attack), a novel white box adversarial method tailored for no-reference IQA models. Using a novel gradient orthogonalization technique, FM-GOAT uniquely optimizes adversarial perturbations against multiple perceptual constraints to minimize visibility, moving beyond traditional <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>l</mi><mi>p</mi></msub></semantics></math></inline-formula>-norm bounds. We evaluate FM-GOAT on seven state-of-the-art NR-IQA models across three image and video datasets, revealing significant vulnerability to the proposed attack. Furthermore, we examine the applicability of adversarial purification methods to the IQA task, as well as their efficiency in mitigating white box adversarial attacks. By studying the activations from models’ intermediate layers, we explore their behavioral patterns in adversarial scenarios and discover valuable insights that may lead to better adversarial detection.
Since the beginning of the existence of the Internet, devices have been connected to it, and in the modern world everything has become more perfect and convenient. Also, the devices themselves have become more compact and accessible to ordinary people, so the Internet of Things or connected devices can be considered the driving force of Industry 4.0, which is also called digital, that is, artificial intelligence can perform certain operations without human assistance. New opportunities are opening up for companies and enterprises, they can significantly save their costs, both time and money, which will help to effectively perform work for the modern business sphere. An employee of the accounting service must be ready for new challenges, because the accounting documentation system is changing. Therefore, every modern accountant must be flexible to digital changes, acquire new knowledge, abilities and skills that are needed in the modern world. But not all new changes are immediately ideal, so there are a number of advantages and disadvantages of the Internet of Things in accounting. The advantage of the Internet of Things in accounting can be considered automation. Devices connected to the Internet collect the necessary data for the employee of the accounting service and, as a result, automatically enter financial data into the accounting system, this advantage allows for quick processing of information, and also reduces the risk of human errors in entering accounting data. But there are also disadvantages of using the Internet of Things in accounting. The main drawback is security and data protection. Ultimately, devices that are not sufficiently secured can be vulnerable to cyber attacks, which can lead to the leakage of financial and other important data stored by businesses. In order to succeed in the implementation of digital changes in accounting, it is necessary to analyze in detail the advantages and disadvantages of new technologies.
Anaylen Beatriz López Velasquez, Albino Goncalves de Sousa
The main objective of the study is to evaluate the impact of Artificial Intelligence (AI) Chatbots on the efficiency and quality of service management at Fundación Divino Niño. The study is qualitative and uses a field research design. It also focuses on describing the indicators of service management in this health institution, after having previously evaluated the AI Chatbots platforms and selected the one that best suits the needs of patient information requests considering the technical criteria. The results indicate that the implementation of AI Chatbots significantly improved efficiency in service management, response times were reduced by 40%, and user satisfaction increased by 25%. In addition, a decrease in operational costs was observed due to the automation of repetitive tasks by the people in charge of attending patients through WhatsApp and Social Media channels. User satisfaction increased notably and they provide a more satisfactory and personalized customer service experience, achieving a reduction in operating costs, freeing up resources that can be allocated to other critical areas of the Fundación Divino Niño. Finally, the impact has a significant impact by offering a sustainable and scalable solution for service management, which is especially beneficial for non-profit organizations.
Chiara Camponeschi, Benedetta Righino, Davide Pirolli
et al.
CD44 is a cell surface glycoprotein transmembrane receptor that is involved in cell–cell and cell–matrix interactions. It crucially associates with several molecules composing the extracellular matrix, the main one of which is hyaluronic acid. It is ubiquitously expressed in various types of cells and is involved in the regulation of important signaling pathways, thus playing a key role in several physiological and pathological processes. Structural information about CD44 is, therefore, fundamental for understanding the mechanism of action of this receptor and developing effective treatments against its aberrant expression and dysregulation frequently associated with pathological conditions. To date, only the structure of the hyaluronan-binding domain (HABD) of CD44 has been experimentally determined. To elucidate the nature of CD44s, the most frequently expressed isoform, we employed the recently developed deep-learning-based tools D-I-TASSER, AlphaFold2, and RoseTTAFold for an initial structural prediction of the full-length receptor, accompanied by molecular dynamics simulations on the most promising model. All three approaches correctly predicted the HABD, with AlphaFold2 outperforming D-I-TASSER and RoseTTAFold in the structural comparison with the crystallographic HABD structure and confidence in predicting the transmembrane helix. Low confidence regions were also predicted, which largely corresponded to the disordered regions of CD44s. These regions allow the receptor to perform its unconventional activity.
The rapid development of internet technology and artificial intelligence drives the demand for flexible sensors. Compared with vacuum technology and solution method, the fabrication of flexible sensors by pencil writing directly has advantages such as low cost, simple operation, low temperature, and no pollution. However, they are based on paper and rely on rigid fibers on the surface of it. The polymer is comfortable, portable and has excellent tensile properties, making it more suitable than paper as flexible substrates for wearable devices. In this paper, flexible pressure sensors and arrays are prepared by friction on polymers (Eco-flex、PDMS and bionic skin). The sensitivity of the prepared pressure sensor was 0.78 kPa−1 in the range of 20 kPa, and the response time was 400 ms, while the pressure detection ranged up to 160 kPa. Finally, it can be reused for 1000 cycles. As a wearable device, it can be applied to object grasping, muscle movement and respiratory monitoring. Furthermore, by combining the friction process with the transfer printing process, the stretchable flexible pressure sensor can be prepared on 3D cylindrical and curved hemispherical surfaces. Moreover, patterned sensors can also be prepared. It should be noted that the sensor can be cleaned after being discarded and has no pollution to the environment due to the mild type of materials, which is of certain significance for the development of flexible sensors towards green and low-cost development trends.
Materials of engineering and construction. Mechanics of materials
A physical model can be used to judge cementing quality to help drilling engineering. This article reports a physical model based on the XGboost algorithm to solve the cementing quality prediction problem of oil and gas wells. Through the physical model, the nonlinear, time-varying, and uncertain influencing factors, the high latitude of the data set, the lack of data, data imbalance and other characteristics are comprehensively analyzed. Finally, through numerical example verification, the physical model we reported can effectively predict the key factors affecting quality, improve process quality and reduce unit cost.
Marija Habijan, Irena Galić, Krešimir Romić
et al.
Accurate segmentation of cardiovascular structures plays an important role in many clinical applications. Recently, fully convolutional networks (FCNs), led by the UNet architecture, have significantly improved the accuracy and speed of semantic segmentation tasks, greatly improving medical segmentation and analysis tasks. The UNet architecture makes heavy use of contextual information. However, useful channel features are not fully exploited. In this work, we present an improved UNet architecture that exploits residual learning, squeeze and excitation operations, Atrous Spatial Pyramid Pooling (ASPP), and the attention mechanism for accurate and effective segmentation of complex cardiovascular structures and name it AB-ResUNet+. The channel attention block is inserted into the skip connection to optimize the coding ability of each layer. The ASPP block is located at the bottom of the network and acts as a bridge between the encoder and decoder. This increases the field of view of the filters and allows them to include a wider context. The proposed AB-ResUNet+ is evaluated on eleven datasets of different cardiovascular structures, including coronary sinus (CS), descending aorta (DA), inferior vena cava (IVC), left atrial appendage (LAA), left atrial wall (LAW), papillary muscle (PM), posterior mitral leaflet (PML), proximal ascending aorta (PAA), pulmonary aorta (PA), right ventricular wall (RVW), and superior vena cava (SVC). Our experimental evaluations show that the proposed AB-ResUNet+ significantly outperforms the UNet, ResUNet, and ResUNet++ architecture by achieving higher values in terms of Dice coefficient and mIoU.