Artificial Intelligence Review最新文献

{"title":"Intelligent strategy for severity scoring of skin diseases based on clinical decision-making thinking with lesion-aware transformer","authors":"Kai Huang,&nbsp;Kai Sun,&nbsp;Jiayi Li,&nbsp;Zhe Wu,&nbsp;Xian Wu,&nbsp;Yuping Duan,&nbsp;Xiang Chen,&nbsp;Shuang Zhao","doi":"10.1007/s10462-024-11083-9","DOIUrl":"10.1007/s10462-024-11083-9","url":null,"abstract":"<div><p>Skin diseases are numerous in types and high in incidence, posing a serious threat to human health. Accurately assessing the severity of skin diseases helps dermatologists in making personalized treatment decisions. However, focusing solely on the skin lesion itself and ignoring the true state of the surrounding skin can lead to distorted results. Assessing the severity of the condition should be a holistic process. Specifically, dermatologists need to compare the abnormal skin with surrounding skin to conduct the diagnosis. To imitate such diagnosis practice of dermatologists, we propose LSATrans, a Transformer based framework customized for severity scoring of skin diseases. Different from the Standard Self-Attention module, we propose the Lesion-aware Self-Attention (LSA) module. LSA can capture the visual features of both lesion and normal surrounding skin areas and include their relationship in modeling. In addition to LSA, the proposed LSATrans also introduces a contrastive learning strategy for further optimization. We first evaluated the performance of LSATrans in scar, atopic dermatitis, and psoriasis scoring tasks, and it achieved mean absolute errors of 0.5895, 0.5614, and 0.5416 respectively in these three tasks. Furthermore, we conducted additional validation of LSATrans’s performance in two distinct skin disease diagnosis tasks, where it demonstrated remarkable outcomes with AUCs of 0.9774 and 0.9801, respectively, in the classification of common skin diseases and subtypes of skin diseases. These results are better than existing methods, indicating that LSATrans is expected to become a universal, accurate and objective intelligent tool for scoring the severity of skin diseases.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11083-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109344","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":"Transfer learning in agriculture: a review","authors":"Md Ismail Hossen,&nbsp;Mohammad Awrangjeb,&nbsp;Shirui Pan,&nbsp;Abdullah Al Mamun","doi":"10.1007/s10462-024-11081-x","DOIUrl":"10.1007/s10462-024-11081-x","url":null,"abstract":"<div><p>The rapid growth of the global population has placed immense pressure on agriculture to enhance food production while addressing environmental and socioeconomic challenges such as biodiversity loss, water scarcity, and climate variability. Addressing these challenges requires adopting modern techniques and advancing agricultural research. Although some techniques, such as machine learning and deep learning, are increasingly used in agriculture, progress is constrained by the lack of large labelled datasets. This constraint arises because collecting data is often time-consuming, labour-intensive, and requires expert knowledge for data annotation. To mitigate data limitations, transfer learning (TL) offers a viable solution by allowing pre-trained models to be adapted for agricultural applications. Many researchers have demonstrated TL’s potential to advance agriculture. Despite its importance, there is a lack of a comprehensive review, which could be essential to guide researchers in this field. Given the significance and the lack of a review paper, this paper provides a review dedicated to TL in agriculture, offering three main contributions. First, we provide an in-depth background study on TL and its applications in agriculture. Second, we offer a comprehensive examination of TL-based agricultural applications, covering pre-trained models, dataset sources, input image types, implementation platforms, and TL approaches. Third, based on an exploration of the existing studies, we identify the challenges faced when applying TL in agriculture. Finally, to address the identified challenges, we recommend suggestions for future research directions.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11081-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109468","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":"HFA-Net: hybrid feature-aware network for large-scale point cloud semantic segmentation","authors":"Changji Wen,&nbsp;Long Zhang,&nbsp;Junfeng Ren,&nbsp;Rundong Hong,&nbsp;Chenshuang Li,&nbsp;Ce Yang,&nbsp;Yanfeng Lv,&nbsp;Hongbing Chen,&nbsp;Ning yang","doi":"10.1007/s10462-025-11111-2","DOIUrl":"10.1007/s10462-025-11111-2","url":null,"abstract":"<div><p>Semantic segmentation of large-scale point clouds in 3D computer vision is a challenging problem. Existing feature extraction modules often emphasize learning local geometry while not giving adequate consideration to the integration of color information. This limitation prevents the network from thoroughly learning local features, thereby impacting segmentation accuracy. In this study, we propose three modules for robust feature extraction and aggregation, forming a novel point cloud segmentation network (HFA-Net) for large-scale point cloud semantic segmentation. First, we introduce the Hybrid Feature Extraction Component (HFEC) and the Hybrid Bilateral Enhancement Component (HBAC) to comprehensively extract and enhance the geometric, color, and semantic information of point clouds. Second, we incorporate the Ternary-Distance Attention Pooling (TDAP) module, which leverages trilateral distances to further refine the network’s focus on various features, enabling it to emphasize both locally important features and broader local neighborhoods. These modules are stacked into dense residual components to expand the network’s receptive field. Our experiments on several large-scale benchmark datasets, including Semantic3D, Toronto3D, S3DIS and LASDU demonstrate the effectiveness of HFA-Net when compared to state-of-the-art networks.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11111-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109472","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":"Adaptive graph diffusion networks: compact and expressive GNNs with large receptive fields","authors":"Chuxiong Sun,&nbsp;Muhan Zhang,&nbsp;Jie Hu,&nbsp;Hongming Gu,&nbsp;Jinpeng Chen,&nbsp;Mingchuan Yang","doi":"10.1007/s10462-025-11114-z","DOIUrl":"10.1007/s10462-025-11114-z","url":null,"abstract":"<div><p>Graph neural networks (GNNs) are widely used in graph-based tasks, but deep GNNs often suffer from oversmoothing. Existing effective deep GNNs have various shortcomings including redundant complexity, oversimplified architecture, or predefined parameters. To address these issues, we propose adaptive graph diffusion networks (AGDNs), a class of compact and expressive GNNs that can effectively leverage deep neighborhood information. We introduce hopwise attention and hopwise convolution with positional embeddings for learning nodewise and channelwise hop weights, respectively, which overcomes oversmoothing and ensures a powerful ability to learn arbitrary filters in the spectral domain. Our experiments demonstrate that AGDNs can effectively learn various filters on images and exhibit superior performance on diverse and challenging open graph benchmark datasets for node classification and link prediction tasks while maintaining moderate complexity and fast running time.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11114-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109390","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":"Topological numbers in uniform intuitionistic fuzzy environment and their application in neural network","authors":"Siti Norziahidayu Amzee Zamri,&nbsp;Haseeb Ahmad,&nbsp;Muhammad Azeem,&nbsp;Bandar Almohsen","doi":"10.1007/s10462-024-10965-2","DOIUrl":"10.1007/s10462-024-10965-2","url":null,"abstract":"<div><p>Intuitionistic Fuzzy sets combine the ideas of uniformity, membership, and non-membership grades of the elements. Similarly, Intuitionistic Fuzzy Graphs are the generalization of simple fuzzy graphs. Depending on the uniformity of the fuzzy graphs (USIF) they can be categorized in different ways via membership values. From the idea of uniform fuzzy topological indices, we have developed the concepts of uniform intuitionistic fuzzy topological indices for uniform intuitionistic fuzzy graphs. This idea provides a more adaptable and nuanced representation of structural properties in graphs or networks. According to the theory of fuzzy topological indices, the importance of topological indices changes depending on the circumstance and the specific problem at hand. When the interactions between nodes are uncertain but not always hesitant, fuzzy graph theory and its adjusted topological indices are sufficient to capture and assess the underlying structure. In such cases where uncertainty is more complicated and hesitation is a major problem then there are better ways to address by intuitionistic fuzzy graph theory and the topological indices that go along with it. This article, developed the concept of uniform intuitionistic fuzzy graphs afresh and proposed Intuitionistic Fuzzy Topological Indices. We determine these indices using the topological indices and labeling of crisp graphs, rather than relying on the degrees of intuitionistic fuzzy graphs and edge portions. This approach is then applied to find intuitionistic fuzzy topological indices. Also, we have provided the MATLAB algorithm to illustrate the concept of IF labeling of cellular neural networks of any order. An example is given to explain the idea and approach towards one kind of uniform intuitionistic fuzzy graph represented by Cellular Neural Networks and graphical plots of the indices involved are also made.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-10965-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109469","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 in four-dimensional imaging for motion management in radiation therapy","authors":"Wang Yinghui,&nbsp;Xiao Haonan,&nbsp;Wang Jing,&nbsp;Wang Lu,&nbsp;Li Wen,&nbsp;Jiang Zhuoran,&nbsp;Ren Ge,&nbsp;Zhi Shaohua,&nbsp;Qian Josh,&nbsp;Dai Jianrong,&nbsp;Men Kuo,&nbsp;Ren Lei,&nbsp;Yang Xiaofeng,&nbsp;Li Tian,&nbsp;Cai Jing","doi":"10.1007/s10462-025-11109-w","DOIUrl":"10.1007/s10462-025-11109-w","url":null,"abstract":"<div><p>Four-dimensional imaging (4D-imaging) plays a critical role in achieving precise motion management in radiation therapy. However, challenges remain in 4D-imaging such as a long imaging time, suboptimal image quality, and inaccurate motion estimation. With the tremendous success of artificial intelligence (AI) in the image domain, particularly deep learning, there is great potential in overcoming these challenges and improving the accuracy and efficiency of 4D-imaging without the need for hardware modifications. In this review, we provide a comprehensive overview of how these AI-based methods could drive the evolution of 4D-imaging for motion management. We discuss the inherent issues associated with multiple 4D modalities and explore the current research progress of AI in 4D-imaging. Furthermore, we delve into the unresolved challenges and limitations in 4D-imaging and provide insights into the future direction of this field.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11109-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109393","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":"A systematic review of multilingual numeral recognition systems","authors":"Meenal Jabde,&nbsp;Chandrashekhar H. Patil,&nbsp;Amol D. Vibhute,&nbsp;Jatinderkumar R. Saini","doi":"10.1007/s10462-025-11105-0","DOIUrl":"10.1007/s10462-025-11105-0","url":null,"abstract":"<div><p>Multilingual numeral recognition systems in online-offline environments play an essential role in several applications like banking or financial transactions, educational sectors, hospitals, etc. Several approaches have been proposed and executed for multilingual numeral recognition for various languages. This study systematically reviews eighty-four articles on the current research on multilingual numeral recognition in offline and online environments. According to the screening criteria, 489 relevant studies were retrieved from standard databases, and only 84 studies were used for further analysis based on the insertion and elimination measures. Our study investigates and analyzes the earlier approaches, datasets developed and utilized, and machine and deep learning methods applied in multilingual numeral recognition across different languages and handwritings. It also provides possible applications and challenges for future studies. Our analysis shows that some datasets are available for scientific research, but comprehensive multilingual datasets and cross-lingual models for multilingual recognition systems are urgently needed. In addition, this review finds that convolutional neural networks (CNN) and support vector machines (SVM) are mainly applied methods in multilingual numeral recognition due to their high recognition accuracy. The findings of this review will provide valuable insights for researchers directing the development of multilingual datasets and robust and effective systems for offline and online multilingual numeral recognition for several multilingual applications.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11105-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109421","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":"Comprehensive exploration of diffusion models in image generation: a survey","authors":"Hang Chen,&nbsp;Qian Xiang,&nbsp;Jiaxin Hu,&nbsp;Meilin Ye,&nbsp;Chao Yu,&nbsp;Hao Cheng,&nbsp;Lei Zhang","doi":"10.1007/s10462-025-11110-3","DOIUrl":"10.1007/s10462-025-11110-3","url":null,"abstract":"<div><p>The rapid development of deep learning technology has led to the emergence of diffusion models as a promising generative model with diverse applications. These include image generation, audio and video synthesis, molecular design, and text generation. The distinctive generation mechanism and exceptional generation quality of diffusion models have made them a valuable tool in these diverse fields. However, with the extensive deployment of diffusion models in the domain of image generation, concerns pertaining to data privacy, data security, and artistic ethics have emerged with increasing prominence. Given the accelerated pace of development in the field of diffusion models, the majority of extant surveys are deficient in two respects: firstly, they fail to encompass the latest advances in diffusion-based image synthesis; and secondly, they seldom consider the potential social implications of diffusion models. In order to address these issues, this paper presents a comprehensive survey of the most recent applications of diffusion models in the field of image generation. Furthermore, it provides an in-depth analysis of the potential social impacts that may result from their use. Firstly, this paper presents a systematic survey of the background principles and theoretical foundations of diffusion models. Subsequently, this paper provides a detailed examination of the most recent applications of diffusion models across a range of image generation subfields, including style transfer, image completion, image editing, super-resolution, and beyond. Finally, we present a comprehensive examination of these social issues, addressing data privacy concerns, such as the potential for data leakage and the implementation of protective measures during model training. We also analyse the risk of malicious exploitation of the model and the defensive strategies employed to mitigate such risks. Additionally, we examine the implications of the authenticity and originality of generated images on artistic creativity and copyright protection.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11110-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109473","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":"A comprehensive review of deep learning-based hyperspectral image reconstruction for agri-food quality appraisal","authors":"Md. Toukir Ahmed,&nbsp;Ocean Monjur,&nbsp;Alin Khaliduzzaman,&nbsp;Mohammed Kamruzzaman","doi":"10.1007/s10462-024-11090-w","DOIUrl":"10.1007/s10462-024-11090-w","url":null,"abstract":"<div><p>Hyperspectral imaging (HSI) has recently emerged as a promising tool for various agricultural applications. However, high equipment cost, instrumentation complexity, and data-intensive nature have limited its widespread adoption. To overcome these challenges, reconstructing hyperspectral data from simple, cost-effective color or RGB (red-green-blue) images using advanced deep learning algorithms offers a practically attractive solution for a wide range of applications in food quality control and assurance. Through advanced deep learning algorithms, it is possible to capture and reconstruct spectral information from simple, cost-effective RGB imaging to create a reliable, efficient, and scalable system with accuracy comparable to dedicated, expensive HSI systems. This review provides a comprehensive overview of recent advances in deep learning techniques for HSI reconstruction and highlights the transformative impact of deep learning-based hyperspectral image reconstruction on agricultural and food products and anticipates a future where these innovations will lead to more advanced and widespread applications in the agri-food industry.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11090-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109504","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 in choroid through optical coherence tomography: a comprehensive review","authors":"Amrish Selvam,&nbsp;Matthew Driban,&nbsp;Joshua Ong,&nbsp;Sandeep Chandra Bollepalli,&nbsp;José-Alain Sahel,&nbsp;Jay Chhablani,&nbsp;Kiran Kumar Vupparaboina","doi":"10.1007/s10462-024-11067-9","DOIUrl":"10.1007/s10462-024-11067-9","url":null,"abstract":"<div><p>Vision-threatening conditions, such as age-related macular degeneration (AMD) and central serous chorioretinopathy (CSCR), arise from dysfunctions in the highly vascular choroid layer in the eye’s posterior segment. Optical coherence tomography (OCT) images play a crucial role in diagnosing choroidal structural changes in clinical practice. This review emphasizes the significant efforts in developing precise detection, quantification, and automated disease classification of choroidal biomarkers. The rapid progress of artificial intelligence (AI) has triggered transformative breakthroughs across sectors including medical image analysis. Recently, the integration of AI within the diagnosis and treatment of choroidal diseases has captured significant attention. Multiple studies highlight AI’s potential to enhance diagnostic precision and optimize clinical outcomes in this context. The review provides an extensive overview of AI’s current applications in choroidal analysis using OCT imaging. It encompasses a diverse array of algorithms and techniques employed for biomarker detection, such as thickness and vascularity index, and for identifying diseases like AMD and CSCR. The overarching goal of this review is to provide an updated and comprehensive exploration of AI’s impact on the choroid, highlighting its potential, challenges, and role in driving innovation in the field.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11067-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109467","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}