As a core component of biomedical ultrasound systems, the performance of ultrasound transducers directly determines the effectiveness of diagnosis and treatment. In recent years, with the development of biomedical technology, there is an increasing demand for high-performance ultrasound transducers. This paper reviews the latest research progress of new ultrasound transducers in the biomedical field, analyzes the performance characteristics of different types of ultrasound transducers, such as small, air-coupled, transparent, flexible, high-frequency arrays, and low-power transducers, and discusses the application prospects in biomedical imaging, therapy, and neuromodulation, and summarizes the current challenges, including the selection of acoustic functional materials for the transducers, structural optimization, and diversification of application scenarios. application scenarios, etc. Future research should focus on the development of new acoustic materials, high-performance transducer design, and interdisciplinary cooperation to promote ultrasonic transducer technology to better serve biomedical applications.

Due to the high recurrence rate of chronic obstructive pulmonary disease (COPD), the issue of unplanned readmissions has become a significant challenge for patients. In this research, a framework and a methodology are proposed that integrate different structured data and multiple machine learning algorithms for risk prediction. A method is showed using genuine electronic medical information from approximately 10 000 COPD patients at a tertiary hospital in Guangzhou, China. To handle unstructured input, a Bidirectional Long Short-Term Memory-Conditional Random Field (BiLSTM-CRF) model known as named entity recognition is used. Furthermore, risk prediction models with Support Vector Machines (SVM), Random Forests (RF), Extreme Gradient Boosting (XGBoost), and Back Propagation (BP) Neural Network are developed. The results show that the XGBoost model performs best. The length of hospital stay, Charlson Comorbidity Index, disease duration, white blood cell count, and eosinophil count are also identified as the most relevant predictors for readmission. An understanding of chronic diseases is advanced by providing research insights and decision support tools for early detection, prompt diagnosis, and precise intervention.

The accurate segmentation of lung tumors plays a crucial role in tumor diagnosis and treatment. However, lung tumor segmentation is often challenged by several issues such as low contrast between lesions and surrounding tissues, tumor-normal tissue adhesion, and high background noise. To address these, this study introduces a lung tumor segmentation method based on Transformer and attention mechanisms. In the Transformer encoder stage, both global and local attention mechanisms are incorporated to enable the network to simultaneously focus on both global and local contextual information. In the skip connection stage, a channel-prior convolutional attention mechanism is utilized to enhance the spatial perception ability for complex lesions and reduce the channel dimension redundancy, such that the tumor segmentation accuracy can be improved. The experimental results on the private GDPH and public LUNG1 datasets demonstrate that the proposed method outperforms eight comparative methods in terms of the Dice metric by achieving approximately 90.96% and 88.18% on the two datasets, respectively. The proposed method can provide reliable assistance for clinical diagnosis and treatment.

Accurate diagnosis and subtyping of non-small cell lung cancer (NSCLC) are crucial for providing patient-specific precision treatment. However, the inherent tumor heterogeneity of NSCLC leads to significant morphological variations within the same subtype and similarities across different subtypes, presenting substantial challenges for pathologists. To address this issue, this study proposes a novel computer-aided diagnostic framework that integrates multi-scale feature extraction and fusion through multi-instance deep learning. The proposed method aims to effectively leverage the heterogeneous information presented in pathological whole-slide images (WSIs) to improve the accuracy of NSCLC subtype classification. Initially, the framework performs multi-scale sampling and feature extraction from WSIs at various levels, such as cellular and tissue levels, to capture both local and global contextual information. Subsequently, a vision transformer network is employed to model the complex dependencies among instances of varying granularity, facilitating end-to-end fusion of the extracted features for accurate classification. Furthermore, we introduce an attention-based instance loss function that adaptively weighs the contribution of each instance based on its discriminative power, providing additional supervision to enhance the classification performance of the model. We evaluat our method on a large public dataset containing 1 674 H&E-stained pathological slide images of NSCLC. The experimental results demonstrate that our multi-scale fusion method effectively leverages the rich information in multi-grained pathological data, significantly outperforming single-scale approaches in NSCLC subtype classification accuracy. Moreover, the method's attention heatmaps offer interpretability and allow for intuitive assessment of individual sample classification quality, serving as a quantitative analytical tool for further model refinement and validation. In conclusion, the proposed multi-scale multi-instance learning framework provides a powerful and interpretable solution for accurate NSCLC subtype classification, which has the potential to assist pathologists in making more reliable diagnostic decisions and ultimately improve patient care.

The circumferential fibers are the key areas of meniscus stress. The construction of three-dimensional microstructure of circumferential fibers is of great significance for the treatment of meniscus injury and the development of artificial meniscus. At present, the circumferential fibers of meniscus in micro computed tomography (MicroCT) images are segmented manually. Because of the complex microstructure of meniscus, manual segmentation has some problems such as low efficiency and inconsistent segmentation standards. To solve the problem of few sample images, an image amplification method is proposed based on the characteristics of MicroCT images. To solve the problem of large edge segmentation errors, an improved model based on TransUNet algorithm is proposed, image Relative Position Encoding (iRPE) is introduced, and the loss function is improved. The experimental results show that: (1) the improved model can accurately and completely segment the meniscus tissue, and the segmentation results can successfully complete the three-dimensional reconstruction of circumferential fibers. (2) The introduced iRPE algorithm improves the segmentation effect of model edge details, the improved loss function enables the model to better adapt to the situation of sample imbalance, and the proposed image amplification method solves the problem of insufficient data sets and comprehensively improves the performance of the model.The results show that the average precision of circumferential fiber segmentation is 98.66%. (3) In the three-dimensional model of circumferential fibers, it is found that fibers are divided into two parts, and a small amount of fibers are divided into three parts. The proposed method can segment the meniscus circumferential fibers in MicroCT images with high accuracy and efficiency, and can pave the way for the study of the force analysis of meniscus in three-dimensional space.

Currently, the main approach of Alzheimer's Disease (AD) diagnosis is realized by structural magnetic resonance imaging (sMRI) , and the existing deep learning-based AD diagnosis is mainly based on 2D (converting 3D sMRI to 2D slices) or 3D convolutional neural networks, which cannot effectively capture 3D sMRI global features. To address this, this paper improves Swin Transformer to realize 3D block division for global features extraction and constructs a Transformer predictive classification model. Due to the sensitivity of existing Alzheimer's patients' regions with atrophy to the transformation of dimensions, the existing deep learning models are not capable of localizing the lesion regions. To overcome this, we propose an adaptive feature extraction module (AFEM) to realize the deformable adaptive feature extraction, and extend the basic 3D Transformer model to construct the AFEM-Transformer deep learning model to further enhance the feature learning ability of the model and realize adaptive localization of the specific location of the pathological region, which can be used to assist clinical diagnosis and realize the classification and prediction of Alzheimer's disease. In this study, sMRI images of 2248 subjects provided by the Alzheimer's Disease Neuroimaging Initiative (ADNI) were selected as the experimental dataset. The proposed AFEM-Transformer model for Alzheimer's disease diagnosis and mild cognitive impairment (MCI) progression prediction tasks will be evaluated and compared with existing convolutional neural network-based models and basic Transformer models. The results show that the experimental results of accuracy, sensitivity, specificity, and area under curve (AUC) value of the proposed AFEM-Transformer model on the two tasks show significant performance improvement compared to the convolutional neural network-based models and basic Transformer model, demonstrating the effectiveness of the proposed AFEM module. The proposed AFEM-Transformer deep learning model is able to accurately diagnose Alzheimer's disease and predict the progression of MCI, and can automatically localize the lesion area, which can be used as an effective computer-aided method in the clinical diagnosis of Alzheimer's disease.

Sparse-view tomographic reconstruction is of significant importance for reducing radiation dose in clinical practice. In recent years, Implicit Neural Representation (INR) methods have been widely applied to medical image reconstruction in sparse-view scenario and have achieved competitive performance. However, traditional INR methods treat each sampling point individually as input, which neglect the inherent relations among neighboring sampling points, thus weakening the reconstruction performance. To address this, this paper proposes a novel INR method. The proposed method reorganizes neighboring sampling points on adjacent rays into multiple windows-of-interest, which are then fed into a Transformer query network equipped with a skip connection. By leveraging the self-attention mechanism of the Transformer network, the proposed method is able to capture the intrinsic relations among sampling points within each window-of-interest, thereby effectively enhancing the reconstructed image quality. This paper conducts extensive numerical experiments in two tomographic imaging modalities: Cone-Beam Computed Tomography (CBCT) and parallel-beam Single-Photon Emission Computed Tomography (SPECT) . The experimental results show that, compared to the advanced INR method Freq-NAF, the proposed method achieves superior performance in terms of reconstruction accuracy and image visualization under sparse-view conditions, particularly obtaining a 0.45 dB improvement in Peak Signal-to-Noise Ratio (PSNR) on the chest CBCT dataset.

Automatic medical report generation technology plays an important role in auxiliary diagnosis and can greatly reduce the workload of medical workers. As deep learning continues to develop in the medical field, automatic medical report generation technology has become one of the research hotspots. Currently, the main challenges in medical report generation are (1) the difficulty of capturing lesion regions in images by models, and (2) the large semantic gap between visual and language semantics, whose consistency problem is still not well solved. Therefore, in order to solve the above problems, a Cross-Modal Discrepancy Attention Network (CDAN) is proposed to bring closer the semantics between different modalities. The network includes a Reverse Attention (RA) module and a Semantic Consistency (SC) module: (1) the Reverse Attention module explores important areas in medical images more comprehensively, and (2) the Semantic Consistency module utilizes the features of the large language model as a reference to guide the visual features to continuously approach the reference language features, so that the visual semantics can be more accurately converted into language semantics. Experiments show that the Cross-Modal Discrepancy Attention Network is better than the previous model on both IU X-Ray and MIMIC-CXR public datasets, with BLEU4 scores reaching 17.9% and 10.9% respectively. Compared with the baseline model, improvement is significant in performance, which proves that the proposed model is capable of generating accurate and fluent medical reports.

In recent years, conductive hydrogels have attracted significant attention in the field of flexible wearable sensors. However, traditional conductive hydrogel-based sensors suffer from insufficient mechanical properties, severely limiting their application in flexible sensors. Therefore, enhancing the mechanical properties of conductive hydrogels is essential for flexible sensing applications. Poly(vinyl alcohol) /poly(ethyleneimine) -sodium sulfate (PVA/PEI-Na₂SO₄) hydrogels were successfully prepared using the directional freezing technique and salting-out effect, showing excellent mechanical properties. During the directional freezing process, the polymer chains in the PVA/PEI hydrogels were arranged orderly along the ice crystal growth direction. This ordered structure improved the mechanical properties of the hydrogels. By immersing the hydrogel in a sodium sulfate solution, the density of the hydrogel network structure increased through the salting-out effect, further enhancing its mechanical properties and endowing the hydrogel with ionic conductivity. The results demonstrated that the PVA/PEI-Na₂SO₄ hydrogel exhibited high compressive strength (5.98 MPa) and excellent force-electrical response properties, with stable electrical signals output during 100 external load-unload cycles. The PVA/PEI-Na₂SO₄ hydrogel developed in this study has potential applications in flexible sensing and finger muscle rehabilitation.

Ill-defined problems, with relatively defined goals and problem domains but expandable boundary conditions, are ubiquitous in life. It is of significant importance to provide directions for the intelligent expansion of knowledge under theoretical guidance, there by forming a comprehensive knowledge network for solving such open-ended problems. This research identifies three primary knowledge categories that constitute the foundation of complete knowledge network: descriptive knowledge of strategy generation materials, domain knowledge that delineates the initial goals and conditions of the problems, and methodological knowledge suitable for the tools employed in strategy generation. The completeness of knowledge network is validated through the achievement of the initial objectives. Finally, the feasibility of the proposed approach is demonstrated through a case study on the design of the ergonomic office chair. By adopting a hybrid approach that integrates formalization and quantification within the extension innovation methodology system, this research offers a novel pathway for establishing a comprehensive knowledge network for solving open-ended problems in an artificial intelligence context.

In order to solve the problem of lack of appropriate innovative design methods in low-carbon design of mechanical products, a low-carbon innovative design process of mechanical products based on resource expansion, which integrates TRIZ resource analysis and extension innovation method, is proposed. Firstly, the model of incompatibility problem is established for mechanical products with low carbon demand, and then the available resources are listed by resource analysis of the existing conditions of the product, and the available resources are described by the primitive model. Then, the multi-screen method is used to expand the resources of resource primitives to get the innovation path, and then the extension transformation is carried out on a number of innovation path to get product ideas. Finally, the superiority evaluation method is used to evaluate these product ideas and select the best idea, and the specific design scheme of the idea is obtained. Taking the traction household elevator as an example, the low-carbon innovative design process of mechanical products is applied to design a low-carbon rigid chain push-pull household elevator, which proves the feasibility of the design process.

Traditional offset tracking primarily relies on normalized cross correlation tracking method based on the regular matching window. However, in the analysis of optical remote sensing images, pixels representing disturbance factors such as cloud layers, water bodies, shadows are often present within the regular window. When applied to landslide monitoring, these disturbance factors may lead to errors in the offset estimation. In order to address this issue, an adaptive offset tracking algorithm is presented. Prior to the offset estimation, a pre-processing step is carried out to identify the locations of these disturbance factors in the study area and generate the corresponding masks. During offset estimation process, the disturbance factors of cross-correlation window can be found from its masks, then pixels representing disturbance factors within the cross-correlation window are excluded, thereby improving the accuracy and reliability of offset estimation experimental validation on the Baige landslide , which has demonstrated that this method can significantly enhance the accuracy and reliability of offset tracking.

For delay sensitive tasks in vehicular ad-hoc networks, vehicle to vehicle fog computing can effectively alleviate the heavy burden of computing tasks on roadside units. Existing studies generally assume that roadside units can obtain the global computing capability information of all vehicles in the network, and service vehicles can autonomously provide computation for service requesting vehicles. However, the high control cost to obtain global computing power information and vehicle selfishness have been overlooked. To addressthe vehicle selfishness, information asymmetry and service fairness problem in burden unloading of vehicle fog computing, we propose a service caching and task offloading integer linear programming model, aiming to maximize the minimum system’s service completion rate. By designing an efficient and lightweight incentive mechanism based on contract theory to incentivize vehicles to provide fog computing resources, roadside units do not need to obtain the global vehicle computing capability information, so as to be closer to the real runtime environment. Extensive simulation results demonstrate that the proposed CRA algorithmimproves the minimum service completion rate by approximately 73.16% and 48.72% over the benchmark algorithms, while the decrease in average total throughputs do not exceed 3.39% and 14.96%.

In recent years, the research on opacity of discrete event systems has attracted widespread attention due to the successful application of opacity in information security of cyber-physical systems. In this research, the enforcement issue of the current state opacity is investigated under the framework of fuzzy discrete event systems modeled by fuzzy automata. The notion of active opacity of fuzzy systems is formalized. For a fuzzy system that does not possess current state opacity, an active control strategy is introduced and then an approach to active opacity is proposed. By constructing a controllability verifier of fuzzy secret states, a controllable event sequence is identified for each secret state, which reinforces the opacity of the fuzzy system by allowing the system to enter a secret state undetected by intruders. Based on the controllability verifier, the necessary and sufficient condition for the active opacity of fuzzy systems is presented, and a corresponding verification algorithm is proposed.

In reality, investors are often influenced by mental account when they manage assets and liabilities at the same time. Therefore, an asset-liability portfolio optimization problem is considered with investors' mental accounts and debt-paying behavior in fuzzy environment. First, we assume that the return rates of assets and the growth rate of liability are LR-fuzzy numbers, with the objectives of maximizing the possibilistic mean of the net wealth and minimizing its lower semi-absolute deviation, a fuzzy asset-liability portfolio optimization model considering investors' mental accounts is proposed. Second, a novel hybrid intelligent algorithm is designed based on Particle Swarm Optimization and Simulated Annealing to solve it. Finally, based on real stock data, a numerical example is conducted to analyze the model and the solving algorithm. The results show that different mental accounts will have different investment strategies, the proposed model can describe investors' mental account characteristics and provide decision support for actual investment activities.