Applied Intelligence最新文献

{"title":"A non-adaptive segmentation algorithm for particle images in controlled environments with uniform backgrounds based on two-round superpixel segmentation and ensemble learning","authors":"Shiming Zhang,&nbsp;Zhikang Ma,&nbsp;Yan Ma","doi":"10.1007/s10489-025-06792-9","DOIUrl":"10.1007/s10489-025-06792-9","url":null,"abstract":"<div><p>Particle image segmentation under controlled environments with uniform backgrounds remains a challenging task due to issues such as particle adhesion, low contrast, and uneven illumination. Existing methods often suffer from over-segmentation or under-segmentation, especially when applied to microscopic or industrial particles. To address these problems, this paper proposes a non-adaptive segmentation algorithm called TS-EL (Two-round Superpixel Segmentation and Ensemble Learning), which is specifically designed for particle images captured in controlled settings with homogeneous backgrounds. The TS-EL framework performs coarse-to-fine superpixel segmentation and easy-to-hard classification. It introduces a gradient distance-based superpixel segmentation algorithm (GradSE) to improve boundary alignment between superpixels and particle contours. A Gaussian model and dual-factor classification criteria are employed to categorize high-confidence superpixels into foreground and background, while low-confidence regions are refined using a second-round segmentation based on minimum bounding boxes. The final classification of ambiguous regions is achieved via the LogitBoost ensemble learning algorithm. Experimental results on three types of particle images (grain, color masterbatch, and cell images) demonstrate that the proposed method outperforms seven state-of-the-art comparative algorithms in terms of segmentation accuracy and boundary adherence. The method is non-adaptive and relies on empirically set parameters, making it well-suited for batch processing in controlled environments but less generalizable to natural or complex scenes.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144891508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"3D ME-Net: multi-scale and edge-guided enhancement network for intracranial aneurysm segmentation","authors":"Jiaqi Wang,&nbsp;Juntong Liu,&nbsp;Jun Li,&nbsp;Aiping Wu,&nbsp;Yunfeng Zhou,&nbsp;Mingquan Ye","doi":"10.1007/s10489-025-06779-6","DOIUrl":"10.1007/s10489-025-06779-6","url":null,"abstract":"<div><p>Intracranial aneurysms are relatively common and life-threatening conditions, making precise segmentation during early diagnosis crucial. However, the challenges of poor imaging quality and high noise levels often result in unclear aneurysm edges. Additionally, the varying sizes of aneurysms further complicate accurate segmentation. To address these issues, we propose a <b>M</b>ultiscale and <b>E</b>dge-guided enhanced 3D deep learning model. First, the asymmetrically larger network with enhanced hierarchical feature representation effectively captures subtle image features, thereby improving the localization of anatomical structures. Second, the multi-scale feature fusion mechanism within the encoder improves feature diversity and edge information, enhancing segmentation precision for aneurysms of different sizes. Finally, the edge-guided attention technique within the decoder combines local features with predicted heatmaps to extract comprehensive edge information. The experimental results demonstrate that the model outperforms general models in five key metrics on the internal dataset. External dataset testing confirms its adaptability and robustness across data from different acquisition protocols and hardware configurations. Clinical trials have further validated its practicality, assisting radiologists in more accurate intracranial aneurysm diagnosis.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144891509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Memristive Bi-Path Wavelet Transformer for low-light image enhancement","authors":"Dirui Xie,&nbsp;Qi Cheng,&nbsp;Yue Zhou,&nbsp;Xiaofang Hu","doi":"10.1007/s10489-025-06771-0","DOIUrl":"10.1007/s10489-025-06771-0","url":null,"abstract":"<div><p>Images captured under low-light conditions are characterized by poor quality and insufficient exposure, which adversely affects the performance of downstream tasks, such as autonomous driving and nighttime surveillance. Recently, Transformer-based methods have achieved notable success in low-light image enhancement. However, these methods exhibit limited local information modeling capabilities and encounter issues with outliers due to insufficient dynamic range, which curtail their performance in low-light image enhancement. Additionally, the quadratic computational complexity of their Softmax-based self-attention mechanisms renders these methods challenging to deploy on edge devices. To address these issues, we propose a memristor-based Bi-Path Wavelet Transformer (BWT) with linear computational complexity. Specifically, we design a novel Dual-path Wavelet Linear Attention (BWLA) to replace the Softmax-based self-attention, enabling efficient local and global information extraction and aggregation at linear complexity. We propose a hardware implementation scheme of BWT based on memristors, which reduces deployment complexity and offers an effective solution for deploying low-light enhancement algorithms on edge devices. Experiments on multiple low-light enhancement benchmark datasets demonstrate that our method outperforms multiple state-of-the-art (SOTA) methods.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144868746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TrajDiffRefine: refinement of spatio-temporal stochastic trajectory prediction via diffusion","authors":"Xiangyun Tan,&nbsp;Qi Zou","doi":"10.1007/s10489-025-06805-7","DOIUrl":"10.1007/s10489-025-06805-7","url":null,"abstract":"<div><p>To effectively respond to sudden events in dynamic and complex environments, trajectory prediction systems must have rapid inference capabilities and low error. This is challenging because it requires using low-complexity models to achieve high-precision predictions, which means having an appropriate balance between inference speed and prediction error. To address this challenge, we present a trajectory prediction model based on diffusion for optimizing predicted trajectories — TrajDiffRefine. The core of the proposed TrajDiffRefine is to construct a simple network for initial predictions, followed by diffusion which progressively refines the predictions. This approach significantly accelerates the inference process while ensuring the precision of the final predictions. Moreover, Initial Estimator accounts for the stochasticity and multi-modal nature of human behavior, including variability in individual decision-making, interaction dynamics, and environmental influences. The introduction of indeterminacy effectively improves prediction performance. Experiments on three real-world datasets—NBA, SDD, and ETH-UCY—show that the proposed method outperforms others in terms of both prediction error and efficiency.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144868888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-feature fusion-based evolutionary algorithm for large-scale sparse multi-objective optimization problems","authors":"Liping Wang,&nbsp;Bangjin Che,&nbsp;Qicang Qiu,&nbsp;Yuyan Gao,&nbsp;Peipei Zhao","doi":"10.1007/s10489-025-06772-z","DOIUrl":"10.1007/s10489-025-06772-z","url":null,"abstract":"<div><p>Large-scale sparse multi-objective optimization problems are common and present significant challenges in scientific research and engineering practice. The primary characteristics of these problems include the high dimensionality of decision variables and the sparsity of the solution set, which greatly increase the problem’s difficulty. During the algorithmic solution process, the interference of non-critical variables reduces the algorithm’s solving efficiency and negatively impacts the quality of the solution set. Therefore, this paper proposes a large-scale sparse multi-objective evolutionary algorithm based on multi-feature fusion, comprehensively considering the importance of decision variables from multiple aspects. First, we introduce a reference point perturbation clustering method. By evenly distributing reference points in the decision space, we control the perturbation of decision variables. The perturbed decision variables are clustered, and an activation function is used to transform the clustering results into contribution values that assess the importance of the decision variables. Second, we propose a sparse feature detection method to mine sparse features from the sparse information of the decision variables, evaluating the informational content of the decision variables. This information is used to filter decision variables to reduce the search space. Finally, the filtered decision variables are competitively optimized using contribution values. Experiments on eight benchmark problems and three real-world applications demonstrate that the algorithm surpasses current state-of-the-art large-scale sparse multi-objective evolutionary algorithms in terms of convergence speed and solution set quality.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144868629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Incomplete multiview clustering with bipartite tensors","authors":"Jiaquan Luo,&nbsp;Changming Zhu","doi":"10.1007/s10489-025-06814-6","DOIUrl":"10.1007/s10489-025-06814-6","url":null,"abstract":"<div><p>Incomplete Multi-view Clustering (IMC) serves as a pivotal approach in multi-view learning, as it effectively captures latent representations from incomplete multi-view data. This capability significantly enhances intelligent systems’ fault tolerance, reduces data acquisition costs, decreases dependency on data completeness in engineering applications, and improves overall robustness. However, existing incomplete multi-view clustering methods suffer from at least one of the following limitations: 1) they fail to fully explore the clustering structure of incomplete multi-view data; 2) they are sensitive to high missing ratios; 3) they treat different views equally, neglecting the inherent differences among views. This results in certain limitations for existing methods in practical applications, as they still rely on specific data completeness requirements. In this paper, we propose a novel tensor low-rank graph learning framework. First, we introduce a similarity matrix fitting module to construct independent low-dimensional representation matrices for different views under low-rank constraints and connectivity constraints. This method can effectively capture the clustering structure of the data. Furthermore, we introduce the tensor Schatten p-norm to reduce the sensitivity of the proposed method to high missing ratios. Then, we stack these low-dimensional representation matrices into a third-order tensor and leverage the advantages of rotation tensors in encoding higher-order correlations and complementary information between views to learn a low-dimensional consensus representation matrix for these low-dimensional representations. Additionally, we introduce an adaptive strategy to maximize the contribution of each view. Extensive experimental results indicate that IMCBT delivers superior performance in clustering tasks compared to various existing incomplete multi-view methods.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144868891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A deep feature refinement self-supervised learning algorithm for medical image annotation","authors":"Jiyong Zhang,&nbsp;Deguang Li,&nbsp;Yan Wu,&nbsp;Zhengwei Zhao,&nbsp;Yanlei Wang,&nbsp;Yang Li,&nbsp;Binqing Zhang","doi":"10.1007/s10489-025-06737-2","DOIUrl":"10.1007/s10489-025-06737-2","url":null,"abstract":"&lt;div&gt;&lt;p&gt;Biomedical image segmentation models heavily rely on large-scale annotated data for training, yet manual annotation notoriously labor-intensive, error-prone, and cost-prohibitive, especially in medical domains requiring expert knowledge. To address this limitation, we propose a self-supervised learning (SSL) framework that leverages unlabeled data to automatically extract discrim- inative features, thereby reducing dependence on human annotations. In this study, self-supervising refers to a learning paradigm where supervisory signals are generated directly from the data itself (e.g., spatial context, channel corre- lations) without external labels. Our goal is to design an SSL method tailored for medical image annotation tasks, enabling robust feature representation even with limited labeled data. This paper introduces a novel self-supervised learning algorithm for refining deep features in the context of medical image annotation tasks. By leveraging the self-supervised ability to learn from unlabeled data, the pro- posed approach aims to enhance feature representation. With the help of spatial and channel attention blocks, our method focuses on intricate feature details within medical images. The spatial attention component enables the network to selectively attend to relevant regions, while the channel attention mecha- nism fine-tunes feature maps for improved annotation accuracy. Both strengthen the model’s ability to capture intricate details and fine-grained information in medical images. To verify the effectiveness of the proposed model, we conducted exten- sive research on four benchmark datasets. The experimental results show that our approach achieves competitive performance compared with other state-of-the-art annotation methods. On the KDSB18(20%) dataset, the values of Precision, Dice and mIoU are 0.964, 0.888, 0.880 (without Barlow Twins Strategy), and 0.965, 0.888, 0.880 (with Barlow Twins Strategy). On the BUSIS dataset with 20% labeled data, the proposed framework achieves a Dice score of 0.861 and mIoU of 0.869, surpassing the baseline U-Net by 36.5% and 23.4%, respectively. For BraTS18 brain tumor segmentation under 10% supervision, our method attains a boundary localization accuracy (Dice) of 0.853, outperforming state-of-the-art models (e.g., RCA-IUNet) by 3.7%. This study develops a novel model that integrates spatial and channel attention, spatial information compression, and dilated convolutions. By leveraging a self-supervised pre-training network with BT strategy, the model optimizes its parameters for improved accuracy and stability on testing data. Experimental results on four datasets demonstrate that our framework consistently improves Dice scores by 12.8–29.8% compared to vanilla self-supervised methods (e.g., Barlow Twins) on medical image segmentation tasks with ≤ 20% annotations. The proposed lesion-aware contrastive loss reduces false positives by 18.5% (from 0.23 to 0.19) in small lesion detection, as va","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144869037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Seeking fixed-time practical consensus tracking of networked nonlinear agent systems with saturation via improved extended state observer","authors":"Chenglin Han,&nbsp;Mengji Shi,&nbsp;Meng Li,&nbsp;Boxian Lin,&nbsp;Weihao Li,&nbsp;Kaiyu Qin","doi":"10.1007/s10489-025-06794-7","DOIUrl":"10.1007/s10489-025-06794-7","url":null,"abstract":"<div><p>This paper addresses the adaptive fixed-time practical consensus tracking control problem of networked systems subject to unknown dynamics, external disturbances, and input saturations. At first, an Improved Extended State Observer (IESO) is developed to estimate state and external disturbances of the leader model accurately. Subsequently, neural networks are utilized to approximate the lumped uncertainties, which include the unknown dynamics and external disturbances of Euler-Lagrange Systems (ELSs), in real-time. Adaptive update laws are formulated to ensure the boundedness of the neural network estimation error. Additionally, an Auxiliary Dynamic System (ADS) is introduced to mitigate the effects of input saturation. A novel adaptive fixed-time controller is proposed and coupled with the ADS, ensuring that the tracking error converges to a predefined residual set. Through the fine-tuning of parameters within the observer and controller, the convergence time of the system can be precisely controlled. The fixed-time convergence of the proposed control scheme is rigorously demonstrated using Lyapunov stability theory. The efficacy of the proposed control strategy is substantiated through simulation examples.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144843201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EGNet: explainable graph neural network with similarity explanation for medication recommendation","authors":"Minh-Van Nguyen,&nbsp;Duy-Thinh Nguyen,&nbsp;Bac Le","doi":"10.1007/s10489-025-06806-6","DOIUrl":"10.1007/s10489-025-06806-6","url":null,"abstract":"<div><p>Giving medication recommendations is a crucial step in improving patient well-being and reducing adverse events. However, existing methods usually fail to capture the complex and dynamic relationships between patient health records, medication efficacy, safety, and drug-drug interactions (DDI), yielding inexplicable outcomes. In this study, we propose an innovative approach that uses graph convolution networks (GCN) with extra external knowledge graphs, attention modules, and an explanation to support prescription recommendations. While the attention system can determine the patient depiction in extended data, GCN can efficiently integrate the external information with the DDI graph into a low-dimensional embedding. We then evaluate our approach using the MIMIC-III and MIMIC-IV datasets, demonstrating that it outperforms several benchmarks in recommendation precision and Drug-Drug Interaction (DDI) prevention. Additionally, we include an explanation stage to illustrate the results and their significant potential impact on industrial applications. The findings confirm that EANet can deliver unparalleled performance while requiring less computational resources and providing enhanced interpretability.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144843200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Don’t get me wrong: How to apply deep visual interpretations to time series","authors":"Christoffer Löffler,&nbsp;Wei-Cheng Lai,&nbsp;Dario Zanca,&nbsp;Lukas Schmidt,&nbsp;Björn M. Eskofier,&nbsp;Christopher Mutschler","doi":"10.1007/s10489-025-06798-3","DOIUrl":"10.1007/s10489-025-06798-3","url":null,"abstract":"<div><p>The correct interpretation of convolutional models is a hard problem for time series data. While saliency methods promise visual validation of predictions for image and language processing, they fall short when applied to time series. These tend to be less intuitive and represent highly diverse data, such as the tool-use time series dataset. Furthermore, saliency methods often generate varied, conflicting explanations, complicating the reliability of these methods. Consequently, a rigorous objective assessment is necessary to establish trust in them. This paper investigates saliency methods on time series data to formulate recommendations for interpreting convolutional models and implements them on the tool-use time series problem. To achieve this, we first employ nine gradient-, propagation-, or perturbation-based post-hoc saliency methods across six varied and complex real-world datasets. Next, we evaluate these methods using five independent metrics to generate recommendations. Subsequently, we implement a case study focusing on tool-use time series using convolutional classification models. Our results validate our recommendations that indicate that none of the saliency methods consistently outperforms others on all metrics, while some are sometimes ahead. Our insights and step-by-step guidelines allow experts to choose suitable saliency methods for a given model and dataset.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 13","pages":""},"PeriodicalIF":3.5,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144832257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}