Artificial Intelligence Review最新文献

{"title":"AI-enabled clinically aligned fewer-lead ECG approaches for detecting cardiovascular diseases","authors":"Tasriva Sikandar,&nbsp;Md. Tariq Hasan,&nbsp;AKM Azad,&nbsp;Asaduzzaman Khan,&nbsp;Mohammad Ali Moni","doi":"10.1007/s10462-026-11577-8","DOIUrl":"10.1007/s10462-026-11577-8","url":null,"abstract":"<div><p>The 12-lead electrocardiogram (ECG) is the clinical gold standard for cardiovascular disease (CVD) detection, though its use in self-care and resource-limited settings is constrained by complex configurations and reliance on specialised diagnostic infrastructure. Commercially available portable/wearable artificial intelligence-enabled ECGs (AI-ECGs) using fewer leads offer greater accessibility yet face clinical integration challenges due to misalignment with gold standard lead configurations, and restricted detection capabilities and clinical interpretability. Consequently, recent research emphasises development of clinically aligned fewer-lead AI-ECG approaches, where the leads are aligned with the 12-lead ECG and used individually or in combinations. However, these approaches remain largely pre-commercial, and comprehensive evaluation is needed to determine their potential and limitations for clinical and self-care acceptability. This review critically assesses 316 studies in terms of detected CVD types, selected leads and rationale for lead selection, and diagnostic performance of AI methods. Key findings reveal limitations in detectable CVD types, diagnostic consistency, incorporation of cardiac electrophysiology and lead interdependency insights, along with the limited adoption of advanced multi-task learning, adaptive methods and explainable AI models. This review also outlines future directions to develop compact and clinically aligned fewer-lead AI-ECGs, bridging the gap between innovation and practical application in clinical and self-care settings.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":18.8,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11577-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148381043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"L.L. Thurstone, The law of effect and the dynamics of deep learning “a deep history of deep learning”","authors":"Stephen Josè Hanson,&nbsp;Catherine Hanson","doi":"10.1007/s10462-026-11560-3","DOIUrl":"10.1007/s10462-026-11560-3","url":null,"abstract":"<div><p>Deep learning (DL), a variant of the neural network algorithms originally proposed in the early twentieth century, has resulted in a renaissance of artificial intelligence. Despite the growing dominance of DL networks, little is understood about the learning mechanism that makes these networks so effective across such a wide range of applications. Drawing on a century of psychological learning theory (e.g., LL Thurstone), an account is offered of the learning mechanism that may enable DL networks to perform so successfully across so many different tasks. Specifically, evidence is provided that learning in DL networks is fit best by a hyperbolic function. This function is independent of hyper/meta parameters–not a scaling function but a learning curve to a specific equilibrium, a function that also entails an autocatalytic mechanism through which complex structures, abstracted from sensory features, can in principle create and support cognitive function in both biological and artificial systems. Keywords: Learning theory, Deep Learning, hyperbolic, autocatalytic, AI, Thurstone </p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":18.8,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11560-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148465005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rebellious AI can be controlled by submissive AI: an overview of background, identifying major problems, and taxonomizing possible solutions","authors":"Saeed Banaeian Far,&nbsp;Mohammad Reza Chalak Qazani,&nbsp;Azadeh Imani Rad","doi":"10.1007/s10462-026-11567-w","DOIUrl":"10.1007/s10462-026-11567-w","url":null,"abstract":"<div><p>Artificial Intelligence (AI) has become an integral component of modern life, significantly enhancing the quality of life and efficiency of processes through AI-powered voice assistants, recommender systems in smart devices, internet-based search engines, and Generative AI (GenAI) chatbots. However, many technologists and futurists caution that advancements in Strong AI (Superintelligence) and GenAI could lead to the emergence of Rebellious AI (RAI), posing unprecedented risks to society. Addressing this critical concern, the present study offers novel solutions to increase awareness and mitigate these risks through a three-pronged approach. <i>First</i>, the study outlines the evolutionary stages of AI: Weak, General, Strong, and Generative, providing a comprehensive review of recent academic contributions and state-of-the-art applications for each phase. This study clarifies the progression of AI technologies and their respective societal implications. <i>Second</i>, the concept of \"Submissive AI\" (SAI) is revisited and expanded, presenting refined definitions and novel strategies for employing this form of AI to manage and control the potential threats posed by RAI. This introduces original methods for harnessing SAI as a counterbalance to the rise of autonomous and potentially hazardous AI systems. <i>Third</i>, the study explores four fundamental challenges associated with RAI, such as (<i>i</i>) autonomy and control over human life, (<i>ii</i>) unintended economic manipulation, (<i>iii</i>) unchecked authority over research and innovation, and (<i>iv</i>) the difficulties in regulating such systems. To address these challenges, the study categorizes proposed solutions into three key strategies: regulatory frameworks that should be established by industrial consortia and governments, decentralization of AI control through blockchain-based technologies, and the strong, correct implementation of SAI mechanisms relying on kill switches. Ultimately, a brief discussion about the ambiguous future of AI is presented, such as AI as an Everything (AIaaX) and the role of AI in accelerating human civilization transition. In synthesizing these contributions, this research offers a significant advancement in the discourse on AI safety, proposing innovative strategies for addressing the complex and growing risks of RAI.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11567-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148247733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Linking AI agent architectures to trust antecedents in human-autonomous teams: a scoping review","authors":"Andrew Zamecnik,&nbsp;Wicaksono Febriantoro,&nbsp;Mutlu Cukurova,&nbsp;Srećko Joksimović,&nbsp;Georg Grossmann,&nbsp;George Siemens","doi":"10.1007/s10462-026-11572-z","DOIUrl":"10.1007/s10462-026-11572-z","url":null,"abstract":"<div><p>The rapid adoption of AI agents is transforming how human teams collaborate in problem-solving across academia, industry, and government. Although AI agents offer promising enhancements to team performance and productivity, trust fundamentally shapes human interactions with technology, influencing the effectiveness and success of human-autonomous teaming (HAT). While there is a wealth of literature on trust between socio-technical relationships, significant gaps persist. In particular, how trust in HAT varies across different agent architectures. This gap limits the ability to identify patterns in existing research, assess the consistency of trust conceptualisations, and understand how methodological choices relate to different agent designs. To address these gaps, this paper systematically maps the current landscape, identifying key bibliometric trends, prevalent agent architectures, and the alignment of trust antecedents with those architectures in recent HAT studies. Our scoping review highlights a lack of clear and systematic links between agent capabilities and the trust factors examined in HAT studies. We emphasise the need for research designs that explore diverse agent roles and team configurations while testing how specific architectures foster particular trust antecedents to strengthen both the validity and applicability of HAT findings.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":18.8,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11572-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148466544","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic radiology report generation: a systematic review of emerging AI architectures and multimodal technologies","authors":"Moustapha Boubacar Hamma,&nbsp;Zakariae Alami Merrouni,&nbsp;Bouchra Frikh,&nbsp;Brahim Ouhbi","doi":"10.1007/s10462-026-11556-z","DOIUrl":"10.1007/s10462-026-11556-z","url":null,"abstract":"<div><p>Generating radiology reports from medical images is a labour-intensive task, exacerbated by the global shortage of radiologists and the complexity of image interpretation. Automatic radiology report generation (ARRG) has emerged to address this challenge, with AI-driven methods showing promise in accelerating diagnosis and improving workflow efficiency. While previous reviews primarily focus on conventional AI and deep learning models and often adopt non-systematic methodologies, this work provides a systematic PRISMA-based review of 51 studies published between 2015 and 2025, offering a rigorous synthesis of model architectures, learning paradigms, data strategies, and evaluation metrics. We categorize approaches from classical CNN-RNN frameworks to emerging paradigms including knowledge graphs, large language models (LLMs), retrieval-augmented generation (RAG), AI agents, and multimodal vision-language models. Our analysis highlights key challenges in clinical translation, including model explainability, factual accuracy, generalization across modalities, and deployment feasibility. We also critically evaluate benchmark datasets and performance metrics, identifying inconsistencies that hinder fair comparison and clinical adoption. Finally, we propose actionable research directions to advance clinically reliable, explainable, and generalizable ARRG systems. By linking technical developments to real-world clinical requirements, this review serves as a forward-looking reference for researchers and practitioners aiming to bridge the gap from AI prototypes to trusted clinical tools.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11556-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148247707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An artificial intelligence-driven fuzzy decision-making framework integrating natural language processing and rule-based logic for preventing agile fatigue","authors":"Serkan Eti,&nbsp;Onur Kardeş,&nbsp;Serhat Yüksel,&nbsp;Hasan Dinçer,&nbsp;Öznur Gülen Ertosun,&nbsp;Özge Doğuç","doi":"10.1007/s10462-026-11562-1","DOIUrl":"10.1007/s10462-026-11562-1","url":null,"abstract":"<div>\u0000 \u0000 <p>Although agile management practices offer advantages such as flexibility and speed, their excessive or improper implementation can lead to agile fatigue, negatively affecting organizational efficiency and employee performance. While the literature extensively examines the antecedents of agile fatigue, studies that systematically prioritize prevention strategies across different organizational contexts remain limited. Therefore, this study aims to identify strategic priorities for preventing agile fatigue and to develop an AI-based fuzzy multi-criteria decision-making framework to support objective and data-driven decision-making from a cross-sectoral perspective rather than focusing on a single industry. The proposed model employs AI-supported natural language processing (NLP) and rule-based decision logic to automatically extract key concepts from literature summaries and generate a structured criteria set. Expert opinions obtained from a limited but heterogeneous group of agile-experienced professionals are weighted based on demographic characteristics using the z-score based normalized ideal distance method (z-NIDM), while criteria weights are calculated via the CRITIC method and strategic alternatives are ranked using the ARLON method. The framework integrates Sierpinski Triangle, Pythagorean, and Fermatean fuzzy sets to model uncertainty in a robust and comparative manner. The findings indicate that business process flexibility (0.178) and cultural adaptation and communication (0.169) are the most critical criteria, while culture and communication strategies (0.194) and business process and structural strategies (0.168) emerge as the most effective interventions. From a practical perspective, these results provide analytically grounded and transferable guidance for managers across different industries by prioritizing structural and cultural actions over isolated technological or training initiatives, offering a strategic roadmap for sustainable agile transformation and the prevention of agile fatigue without claiming universal generalizability.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 7","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11562-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148237104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial intelligence for the diagnosis and detection of smoking and alcohol-related cancers: a systematic review","authors":"Nandini Modi,&nbsp;Sonam Gandotra,&nbsp;Yogesh Kumar,&nbsp;Noopur Tyagi","doi":"10.1007/s10462-026-11552-3","DOIUrl":"10.1007/s10462-026-11552-3","url":null,"abstract":"<div>\u0000 \u0000 <p>Cancer is one of the major global health challenge and tobacco use and alcohol consumption significantly contribute as risk factors, predominantly for stomach, liver, lungs, esophagus and oral cancers. The interplay between lifestyle factors and genetic traits further complicates disease progression, underscoring the need for early detection and precise risk assessment. The presented systematic review explores how artificial intelligence (AI) and deep learning (DL) techniques such as advanced convolutional neural networks (CNNs) and transfer learning are used to improve the diagnosis of cancers linked to smoking and alcohol consumption. It aims to provide a detailed review of existing studies on the detection of these cancers and associated symptoms, highlighting the role of AI-driven methods with high accuracy. The article discusses advancements that focus on Explainable AI (XAI) to enhance transparency, multi-modal data integration for comprehensive diagnostics and strategies to improve model generalizability across diverse populations. The main emphasis of the review article is placed on AI methods used for identifying symptoms and pathophysiological patterns associated with smoking and alcohol induced cancers, as well as techniques for detecting these substances in patient records and imaging data.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 7","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11552-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148237151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From benchmarks to deployment: a comprehensive review of agentic AI evaluation","authors":"Tanzila Kehkashan,&nbsp;Muhammad Abdullah,&nbsp;Ahmad Sami Al-Shamayleh,&nbsp;Nikola Ivković,&nbsp;Nor Azman Ismail,&nbsp;Sharifah Sakinah Syed Ahmad,&nbsp;Abdul Rehman,&nbsp;Adnan Akhunzada","doi":"10.1007/s10462-026-11571-0","DOIUrl":"10.1007/s10462-026-11571-0","url":null,"abstract":"<div>\u0000 \u0000 <p>This review systematically examines evaluation methodologies for agentic AI systems, agentic AI systems capable of multi-step planning, tool usage, and environmental interaction across diverse domains. Current evaluation practices exhibit a critical disconnect between benchmark performance and deployment viability, where agents achieving high scores on standardized benchmarks frequently fail in real world applications due to fundamental inadequacies in assessment methodologies that prioritize task completion over deployment critical dimensions such as cost efficiency, safety compliance, maintainability, and workflow integration. We critically analyze 15 major agent benchmarks including AgentBench, WebArena, SWE-bench, PaperBench, MLGym, BrowserGym, HumanEval, MBPP, GAIA, ToolBench, Terminal-Bench, Mind2Web, ALFWorld, BabyAI, and HotPotQA, examining their methodologies, metrics, datasets, and application domains with software development serving as our primary case study. Our analysis reveals that evaluation methodology not model capability constitutes the primary bottleneck limiting reliable agent deployment. We demonstrate how test passing metrics systematically ignore code quality, security vulnerabilities, and integration complexity while binary success metrics obscure planning coherence, resource efficiency, and safety violations (0/15 benchmarks integrate security or safety into scoring. This review provides a cross-domain taxonomy exposing evaluation inadequacies, trajectory level evaluation frameworks addressing cost reproducibility validity trade-offs, systematic identification of metric insufficiencies including absence of safety-aware and cost aware scoring, and synthesis of emerging evaluation paradigms with adoption barriers. Quantitatively, 0/15 benchmarks integrate safety or security into scoring, 0/15 include cost-efficiency metrics in their primary evaluation protocol, and 13/15 rely exclusively on binary success measures confirming that evaluation methodology, not model capability, is the primary bottleneck to reliable deployment. Progress toward trustworthy agentic AI fundamentally depends on evolving evaluation infrastructure beyond binary metrics toward comprehensive, multidimensional assessment frameworks that capture these deployment-essential dimensions.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11571-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148247708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interpretable Automated Feature Engineering: A Comprehensive Review with a Focus on Dynamic and Stationary Environments","authors":"Dilshan Sonnadara,&nbsp;Thanh Tung Khuat,&nbsp;Katarzyna Musial-Gabrys,&nbsp;Bogdan Gabrys","doi":"10.1007/s10462-026-11573-y","DOIUrl":"10.1007/s10462-026-11573-y","url":null,"abstract":"<div>\u0000 \u0000 <p>This paper presents a comprehensive review of Automated Feature Engineering (AFE) methods, with a focus on their application in static data environments and explores recent advancements in dynamic Feature Engineering (FE) for evolving data contexts. AFE is essential in modern Machine Learning (ML) pipelines, as it addresses the drawbacks of traditional FE, which often relies heavily on domain expertise and is labor intensive. By automating this process, AFE has the potential to enhance the efficiency, performance, and interpretability of ML workflows. The paper analyzes the methods, strengths, and limitations of different AFE methods. The review also identifies critical research gaps and proposes directions for future advancements. This work serves as a foundational resource for researchers aiming to optimize ML performance through novel AFE strategies.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 7","pages":""},"PeriodicalIF":13.9,"publicationDate":"2026-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11573-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148237252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Explainable artificial intelligence techniques for interpretation of food models: a review","authors":"Leonardo Arrighi,&nbsp;Ingrid Alves de Moraes,&nbsp;Marco Zullich,&nbsp;Michele Simonato,&nbsp;Douglas Fernandes Barbin,&nbsp;Sylvio Barbon Junior","doi":"10.1007/s10462-026-11553-2","DOIUrl":"10.1007/s10462-026-11553-2","url":null,"abstract":"<div>\u0000 \u0000 <p>Artificial Intelligence (AI) has become essential for analyzing complex data and solving highly-challenging tasks. It is being applied across numerous disciplines beyond computer science, including Food Engineering, where there is a growing demand for accurate and reliable predictions to meet stringent food quality standards. However, this requires increasingly complex AI models, raising concerns. In response, eXplainable AI (XAI) has emerged to provide insights into AI decision-making, aiding model interpretation by developers and users. Nevertheless, XAI remains underutilized in Food Engineering, limiting model reliability. For instance, in food quality control, AI models using spectral imaging can detect contaminants or assess freshness levels, but their opaque decision-making process hinders adoption. XAI techniques such as SHAP (Shapley Additive Explanations) and Grad-CAM (Gradient-weighted Class Activation Mapping) can pinpoint which spectral wavelengths or image regions contribute most to a prediction, enhancing transparency and aiding quality control inspectors in verifying AI-generated assessments. This survey presents a taxonomy for classifying food quality research using XAI techniques, organized by data types and explanation methods, to guide researchers in choosing suitable approaches. We also highlight trends, challenges, and opportunities to encourage the adoption of XAI in Food Engineering.</p>\u0000 </div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"59 8","pages":""},"PeriodicalIF":18.8,"publicationDate":"2026-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-026-11553-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148466429","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}