Artificial Intelligence Review最新文献

{"title":"Survey: application and analysis of generative adversarial networks in medical images","authors":"Yang Heng,&nbsp;Ma Yinghua,&nbsp;Fiaz Gul Khan,&nbsp;Ahmad Khan,&nbsp;Farman Ali,&nbsp;Ahmad Ali AlZubi,&nbsp;Zeng Hui","doi":"10.1007/s10462-024-10992-z","DOIUrl":"10.1007/s10462-024-10992-z","url":null,"abstract":"<div><p>Generative Adversarial Networks (GANs) have shown promising prospects and achieved significant results in medical image analysis tasks. This article provides a comprehensive review of recent research on GANs and their variants in medical applications, including tasks such as image synthesis, segmentation, classification, detection, denoising, reconstruction, fusion, registration, and prediction. We summarize and analyze the reviewed literature, with a focus on model framework design,dataset sources, and performance evaluation. Our research findings are presented in the form of tables. In the end,article discusses open challenges and directions for future research.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-10992-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859522","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":"Genetic biomarkers and machine learning techniques for predicting diabetes: systematic review","authors":"Sulaiman Khan,&nbsp;Farida Mohsen,&nbsp;Zubair Shah","doi":"10.1007/s10462-024-11020-w","DOIUrl":"10.1007/s10462-024-11020-w","url":null,"abstract":"<div><p>Diabetes mellitus is a long-term metabolic condition marked by high blood sugar levels due to issues with insulin production, insulin effectiveness, or a combination of both. It stands as one of the fastest-growing diseases worldwide, projected to afflict 693 million adults by 2045. The escalating prevalence of diabetes and associated health complications (kidney disease, retinopathy, and neuropathy) underscore the imperative to devise predictive models for early diagnosis and intervention. These complications contribute to increased mortality rates, blindness, kidney failure, and an overall diminished quality of life in individuals living with diabetes. While clinical risk factors and glycemic control provide valuable insights, they alone cannot reliably predict the onset of vascular complications. Genetic biomarkers and machine learning techniques have emerged as promising tools for predicting diabetes development risk and associated complications. Despite the emergence of numerous smart AI models for diabetes prediction, there is still a need for a thorough review outlining their progress and challenges. To address this gap, this paper offers a systematic review of the literature on AI-based models for diabetes identification, following the PRISMA extension for scoping reviews guidelines. Our review revealed that multimodal diabetes prediction models outperformed unimodal models. Most studies focused on classical machine learning models, with SNPs being the most used data type, followed by gene expression profiles, while lipidomic and metabolomic data were the least utilized. Moreover, some studies focused on identifying genetic determinants of diabetes complications relied on familial linkage analysis, tailored for robust effect loci. However, these approaches had limitations, including susceptibility to false positives in candidate gene studies and underpowered AI models capabilities due to sample size constraints. The landscape shifted dramatically with the proliferation of genomic datasets, fueled by the emergence of biobanks and the amalgamation of global cohorts. This surge has led to a more than twofold increase in genetic discoveries related to both diabetes and its complications using AI. Our focus here is on these genetic breakthroughs, particularly those empowered by AI models. However, we also highlight the existing gaps in research and underscore the need for further advancements to propel genomic discovery to the next level.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11020-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859524","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":"Deep mining the textual gold in relation extraction","authors":"Tanvi Sharma,&nbsp;Frank Emmert-Streib","doi":"10.1007/s10462-024-11042-4","DOIUrl":"10.1007/s10462-024-11042-4","url":null,"abstract":"<div><p>Relation extraction (RE) is a fundamental task in natural language processing (NLP) that seeks to identify and categorize relationships among entities referenced in the text. Traditionally, RE has relied on rule-based systems. Still, recently, a variety of deep learning approaches have been employed, including recurrent neural networks (RNNs), convolutional neural networks (CNNs), and bidirectional encoder representations from transformers (BERT). This review aims to provide a comprehensive overview of relation extraction, focusing on deep learning models. Given the complexity of the RE problem, we will present it from a multi-dimensional perspective, covering model steps, relation types, method types, benchmark datasets, and applications. We will also highlight both historical and current research in the field, identifying promising research areas for further development and emerging directions. Specifically, we will focus on potential enhancements for relation extraction from poorly labeled data and provide a detailed assessment of current shortcomings in handling complex real-world situations.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11042-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142789273","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":"Generative AI model privacy: a survey","authors":"Yihao Liu,&nbsp;Jinhe Huang,&nbsp;Yanjie Li,&nbsp;Dong Wang,&nbsp;Bin Xiao","doi":"10.1007/s10462-024-11024-6","DOIUrl":"10.1007/s10462-024-11024-6","url":null,"abstract":"<div><p>The rapid progress of generative AI models has yielded substantial breakthroughs in AI, facilitating the generation of realistic synthetic data across various modalities. However, these advancements also introduce significant privacy risks, as the models may inadvertently expose sensitive information from their training data. Currently, there is no comprehensive survey work investigating privacy issues, e.g., attacking and defending privacy in generative AI models. We strive to identify existing attack techniques and mitigation strategies and to offer a summary of the current research landscape. Our survey encompasses a wide array of generative AI models, including language models, Generative Adversarial Networks, diffusion models, and their multi-modal counterparts. It indicates the critical need for continued research and development in privacy-preserving techniques for generative AI models. Furthermore, we offer insights into the challenges and discuss the open problems in the intersection of privacy and generative AI models.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11024-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142761983","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":"Data augmentation in predictive maintenance applicable to hydrogen combustion engines: a review","authors":"Alexander Schwarz,&nbsp;Jhonny Rodriguez Rahal,&nbsp;Benjamín Sahelices,&nbsp;Verónica Barroso-García,&nbsp;Ronny Weis,&nbsp;Simon Duque Antón","doi":"10.1007/s10462-024-11021-9","DOIUrl":"10.1007/s10462-024-11021-9","url":null,"abstract":"<div><p>Machine-learning-based predictive maintenance models, i.e. models that predict breakdowns of machines based on condition information, have a high potential to minimize maintenance costs in industrial applications by determining the best possible time to perform maintenance. Modern machines have sensors that can collect all relevant data of the operating condition and for legacy machines which are still widely used in the industry, retrofit sensors are readily, easily and inexpensively available. With the help of this data it is possible to train such a predictive maintenance model. The main problem is that most data is obtained from normal operating conditions, whereas only limited data are from failures. This leads to highly unbalanced data sets, which makes it very difficult, if not impossible, to train a predictive maintenance model that can detect faults reliably and timely. Another issue is the lack of available real data due to privacy concerns. To address these problems, a suitable data generation strategy is needed. In this work, a literature review is conducted to identify a solution approach for a suitable data augmentation strategy that can be applied to our specific use case of hydrogen combustion engines in the automotive field. This literature review shows that, among the different state-of-the-art proposals, the most promising for the generation of reliable synthetic data are the ones based on generative models. The analysis of the different metrics used in the state of the art allows to identify the most suitable ones to evaluate the quality of generated signals. Finally, an open problem in research in this area is identified and it is the need to validate the plausibility of the data generated. The generation of results in this area will contribute decisively to the development of predictive maintenance models.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11021-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142761693","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":"Scene reconstruction techniques for autonomous driving: a review of 3D Gaussian splatting","authors":"Huixin Zhu,&nbsp;Zhili Zhang,&nbsp;Junyang Zhao,&nbsp;Hui Duan,&nbsp;Yao Ding,&nbsp;Xiongwu Xiao,&nbsp;Junsong Yuan","doi":"10.1007/s10462-024-10955-4","DOIUrl":"10.1007/s10462-024-10955-4","url":null,"abstract":"<div><p>As the latest research result of the explicit radiated field technology, 3D Gaussian Splatting (3D GS) replaces the implicit expression represented by Neural Radiated Field (NeRF) and has become the hottest research direction in 3D scene reconstruction. Given the innovative work and vigorous development of 3D GS in autonomous driving, this paper comprehensively reviews and summarizes the existing related research to showcase the evolution of the 3D GS technology and possible future development directions. First, the overall research background of 3D GS is introduced based on two aspects 3D scene reconstruction methods and 3D GS research progress. Second, the relevant knowledge points of 3D GS and the core formulas to clarify the mathematical mechanism of 3D GS are presented. Third, the primary applications of the 3D scene reconstruction technology based on 3D GS in automatic driving are presented through new perspective synthesis, scene understanding, and simultaneous localization and map building (SLAM). Finally, the research frontier directions of 3D GS in autonomous driving are described, including structure optimization, 4D scene reconstruction, and cross-domain research. This paper may provide an effective and convenient pathway for researchers to understand, explore, apply this novel research method, and promote the development and application of 3D GS in automatic driving.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-10955-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142753923","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":"Optimizing seismic-based reservoir property prediction: a synthetic data-driven approach using convolutional neural networks and transfer learning with real data integration","authors":"Muhammad Ali,&nbsp;He Changxingyue,&nbsp;Ning Wei,&nbsp;Ren Jiang,&nbsp;Peimin Zhu,&nbsp;Zhang Hao,&nbsp;Wakeel Hussain,&nbsp;Umar Ashraf","doi":"10.1007/s10462-024-11030-8","DOIUrl":"10.1007/s10462-024-11030-8","url":null,"abstract":"<div><p>Reservoir characterization through seismic data analysis is essential for exploration and production in the petroleum industry. However, seismic-to-well tie discrepancies, limited availability of high-quality well data, and resolution constraints pose a reliability challenge. While previous studies offer valuable insights, they still struggle to achieve high-resolution predictions in a complex geologically environment given high reliance on well data. This study integrates synthetic data-driven techniques with real data, including convolutional neural networks (CNN) and transfer learning, to improve seismic reservoir characterization. We utilize nearby well statistics and a rock physics model (RPM) to simulate pseudo wells representing various geological scenarios. Synthetic seismic gathers are generated from these pseudo wells, which are based on RPM and local well control, to train the CNN. Transfer learning is then applied to adapt the CNN to better distinguish between real and synthetic data, enhancing reservoir predictions. A comparative analysis of P-impedance predictions from three methodologies: theory-driven Pre-Stack-Seismic-Inversion (TDSI), Deep-Neural-Network (DNN), and our CNN approach, showed that CNN achieved nearly 97% prediction accuracy with low error rates, compared to relatively lower prediction accuracy rates of DNN (86.2%) and TDSI (81.5%) with high error rates, according to robust metrics including R-square, RMSE, MSE, and MAE. These results indicate that CNN not only enhanced resolution but also closely aligned with well data and superior lateral continuity, even in blind well scenarios. This study effectively integrates synthetic data-driven techniques with CNNs and transfer learning to advance seismic reservoir property prediction, offering a robust solution to overcome limitations in traditional and DNN-based approaches.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11030-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142754328","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":"Transit search algorithm based on oscillation exploitation factor and Roche limit for wireless sensor network deployment optimization","authors":"Yu-Xuan Xing,&nbsp;Jie-Sheng Wang,&nbsp;Si-Wen Zhang,&nbsp;Shi-Hui Zhang,&nbsp;Xin-Ru Ma,&nbsp;Yun-Hao Zhang","doi":"10.1007/s10462-024-10951-8","DOIUrl":"10.1007/s10462-024-10951-8","url":null,"abstract":"<div><p>To optimize the deployment of nodes in Wireless Sensor Networks (WSN) and effectively control network node energy consumption, thereby improving the quality of perception services, a Transit search algorithm based on oscillation exploitation factor and Roche limit is proposed. The Roche limit-inspired approach enhances the stellar phase of the algorithm, accelerating the convergence rate in the mid-to-late stages of iteration while ensuring adequate exploration of the solution space. Subsequently, five weakening oscillation development factors are introduced to refine the algorithm’s exploitation phase and improve its fine-tuning accuracy. To validate the effectiveness of these strategies, various approaches are applied to optimize the coverage, waste rate and energy consumption in two models of WSN deployment, with connectivity recorded. The comparison reveals the optimal improved algorithm, SEROTS, which enhances coverage by 1.34% in the obstacle-free model compared to the original TS algorithm, with waste and energy consumption rates reduced by 2.05% and 0.00016%, respectively. In the obstacle model, coverage increases by 1.49%, while waste and energy consumption rates decrease by 6.96% and 0.0004%, respectively. To demonstrate the efficiency of the improved algorithm in optimizing WSN deployment, SEROTS is compared with four optimization algorithms: Egret Swarm Optimization Algorithm (ESOA), Honey Badger Algorithm (HBA), Sparrow Search Algorithm (SSA) and Differential Evolution (DE). Two models are selected, integrating the three objectives into a single objective function. Simulation results indicate that SEROTS performs best in both models, with an improvement of 0.53% and 0.79% over the second-best algorithm, respectively. Furthermore, the proposed strategies are compared with simulation results from five other studies, achieving higher coverage rates by 1.57%, 3.33%, 0.87%, 3.81% and 0.21%, respectively. Finally, experiments discuss the application in large-scale scenarios, verifying the feasibility and efficiency of the SEROTS algorithm in WSN deployment optimization.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-10951-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142714662","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":"Deep learning adversarial attacks and defenses in autonomous vehicles: a systematic literature review from a safety perspective","authors":"Ahmed Dawod Mohammed Ibrahum,&nbsp;Manzoor Hussain,&nbsp;Jang-Eui Hong","doi":"10.1007/s10462-024-11014-8","DOIUrl":"10.1007/s10462-024-11014-8","url":null,"abstract":"<div><p>The integration of Deep Learning (DL) algorithms in Autonomous Vehicles (AVs) has revolutionized their precision in navigating various driving scenarios, ranging from anti-fatigue safe driving to intelligent route planning. Despite their proven effectiveness, concerns regarding the safety and reliability of DL algorithms in AVs have emerged, particularly in light of the escalating threat of adversarial attacks, as emphasized by recent research. These digital or physical attacks present formidable challenges to AV safety, relying extensively on collecting and interpreting environmental data through integrated sensors and DL. This paper addresses this pressing issue through a systematic survey that meticulously explores robust adversarial attacks and defenses, specifically focusing on DL in AVs from a safety perspective. Going beyond a review of existing research papers on adversarial attacks and defenses, the paper introduces a safety scenarios taxonomy matrix Inspired by SOTIF designed to augment the safety of DL in AVs. This matrix categorizes safety scenarios into four distinct areas and classifies attacks into those areas in three scenarios, along with two defense scenarios. Furthermore, the paper investigates the testing and evaluation measurements critical for assessing attacks in the context of DL for AVs. It further explores the dynamic landscape of datasets and simulation platforms. This contribution significantly enriches the ongoing discourse surrounding the assurance of safety and reliability in autonomous vehicles, especially in the face of continually evolving adversarial challenges.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11014-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142714663","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":"Integrated MADM approach based on extended MABAC method with Aczel–Alsina generalized weighted Bonferroni mean operator","authors":"Kaushik Debnath,&nbsp;Sankar Kumar Roy,&nbsp;Muhammet Deveci,&nbsp;Hana Tomášková","doi":"10.1007/s10462-024-10980-3","DOIUrl":"10.1007/s10462-024-10980-3","url":null,"abstract":"<div><p>Currently, <i>q</i>-rung orthopair (<i>q</i>-ROF) set theory is one of the most effective set theories in dealing uncertainty associated with imprecise information. In complex decision-making problems, input variables can be described by <i>q</i>-ROF numbers to cope ambiguity. While, generalized weighted Bonferroni mean (GWBM) operator can reflect correlation among input arguments. Aczel–Alsina operations underline fair and accurate evaluation of decision-makers. Harnessing these benefits, a pioneering extension of the GWBM operator based on Aczel–Alsina operations is introduced. Simultaneously, a novel generalized distance measure is crafted, drawing inspiration from Dice and Jaccard similarities. Beside these, using stepwise weight assessment ratio analysis (SWARA) and multi-attribute border approximation area comparison (MABAC) methods, this study pioneers an integrated method, <i>q</i>-ROF-SWARA-MABAC for assessing and prioritizing factors and alternatives on <i>q</i>-ROF environment. Later, with the suggested model, a case study on high-speed rail corridor (HSRC) for India is solved, revealing Varanasi-Howrah HSRC as the most preferable choice.. Moving forward, detailed sensitive analysis of suggested model is performed to explore the pertinence and supremacy. Eventually, the outcomes manifest that novel framework is flexible, reliable, accurate and could be viable option to consider for future use.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 1","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-10980-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142714661","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}