The increasing number of maritime transport vessels has elevated ship distress accidents, emphasizing the imperative for effective Search and Rescue (SAR) operations. In this study, we introduce a coverage path planning framework for multiple Unmanned Aerial Vehicles (UAVs) to optimize maritime SAR missions by explicitly accounting for regional priorities. The proposed framework hierarchically incorporates these priorities and derives coverage paths using a heuristic algorithm based on Ant Colony Optimization (ACO) for each priority region. The algorithm consists of two key phases: the Region Allocation Phase (RAP), which utilizes a greedy approach guided by regional importance, and the Dynamic Refinement Phase (DRP), which employs ACO to ensure balanced resource usage. Comprehensive numerical experiments, utilizing actual sea coordinates, are conducted to validate the effectiveness of the proposed methodology. The proposed method outperforms both a MILP model and a reinforcement learning-based approach, particularly in large-scale instances. It demonstrates computational robustness and scalability, consistently generating high-quality solutions within practical computation times, even in complex environments. The methodology is further validated through experiments on irregular environments, confirming its adaptability and effectiveness.

Diagnosing malignant skin tumors accurately at an early stage can be challenging due to ambiguous and even confusing visual characteristics displayed by various categories of skin tumors. To improve diagnosis precision, all available clinical data from multiple sources, particularly clinical images, dermoscopy images, and medical history, could be considered. Aligning with clinical practice, we propose a novel Transformer model, named RemixFormer++ that consists of a clinical image branch, a dermoscopy image branch, and a metadata branch. Given the unique characteristics inherent in clinical and dermoscopy images, specialized attention strategies are adopted for each type. Clinical images are processed through a top-down architecture, capturing both localized lesion details and global contextual information. Conversely, dermoscopy images undergo a bottom-up processing with two-level hierarchical encoders, designed to pinpoint fine-grained structural and textural features. A dedicated metadata branch seamlessly integrates non-visual information by encoding relevant patient data. Fusing features from three branches substantially boosts disease classification accuracy. RemixFormer++ demonstrates exceptional performance on four single-modality datasets (PAD-UFES-20, ISIC 2017/2018/2019). Compared with the previous best method using a public multi-modal Derm7pt dataset, we achieved an absolute 5.3% increase in averaged F1 and 1.2% in accuracy for the classification of five skin tumors. Furthermore, using a large-scale in-house dataset of 10,351 patients with the twelve most common skin tumors, our method obtained an overall classification accuracy of 92.6%. These promising results, on par or better with the performance of 191 dermatologists through a comprehensive reader study, evidently imply the potential clinical usability of our method.

Conventional one-shot segmentation methods typically employ either registration methods for label propagation from a reference atlas or utilize synthetically labeled data to augment the training of segmentation networks. However, these approaches often fail to accurately capture anatomical structural information from real images, resulting in suboptimal segmentation performance and limited generalizability across different medical imaging modalities. In this paper, we propose a robust one-shot brain tissue segmentation framework, which requires only a single labeled image and a few unlabeled ones. Our approach features a novel synthesis module based on patch-wise contrastive learning for generating realistic, well-labeled training samples, followed by a dual-head Variational AutoEncoder (VAE) module for the joint learning of reconstruction and segmentation. During the final inference, the well-trained VAE is capable of precisely segmenting new, unseen images. Our method demonstrates versatility across various imaging modalities. Evaluations on two public MRI datasets and one in-house CT dataset reveal that our method achieves superior performance and enhanced generalization capabilities compared to existing state-of-the-art methods.

Web Application Firewalls (WAFs) are widely deployed to protect key web applications against multiple security threats, so it is important to test WAFs regularly to prevent attackers from bypassing them easily. Machine-learning-based black-box WAF testing is gaining more attention, though existing learning-based approaches have strict requirements on the source and scale of payload data and suffer from the local optimum problem, limiting their effectiveness and practical application. We propose GPTFuzzer, apracticalandeffectivegeneration-based approach to test WAFs by generating attack payloads token-by-token. Specifically, we fine-tune a Generative Pre-trained Transformer language model with reinforcement learning to make GPTFuzzer have the least restrictions on payload data and thus more applicable in practice, and we use reward modeling and KL-divergence penalty to improve the effectiveness of our approach and mitigate the local optimum issue. We implement GPTFuzzer and evaluate it on two well-known open-source WAFs against three kinds of common attacks. Experimental results show that GPTFuzzer significantly outperforms state-of-the-art approaches,i.e.ML-Driven and RAT, finding up to 7.8× (3.2× on average) more bypassing payloads within 1,250,000 requests, or finding out all bypassing payloads using up to 8.1× (3.3× on average) fewer requests.

We present an innovative physics-informed numerical flux within the framework of the discontinuous Galerkin (DG) method for solving elastic wave equations in 2-D heterogeneous media that include interfaces between elastic materials. The first-order velocity–stress equations are used, which can be written in the formulation of hyperbolic system and can be easily incorporated into the framework of DG method. Numerical flux is carefully proposed to maintain the physical laws on both sides of the interface where material parameters are discontinuous. We compare the newly suggested numerical flux with both the classic local Lax–Friedrichs flux and the exact upwind flux. The latter is derived from solving the Riemann problem and adheres to the Rankine–Hugoniot condition. In the comparison, we find that our flux is formally similar to the classic local Lax–Friedrichs flux, but the numerical behavior is different from it; its numerical performance is similar to the exact upwind flux. The biggest advantage of the numerical flux we propose is that it does not need to accurately solve the Riemann problem, but it can maintain the physical continuity conditions near the interface with material discontinuities. We present three numerical examples of elastic wave propagation in heterogeneous media, including horizontal and inclined interfaces, and the Marmousi model. The numerical results demonstrate the effectiveness of this flux.

Extracting invariant representations in unlabeled electrocardiogram (ECG) signals is a challenge for deep neural networks (DNNs). Contrastive learning is a promising method for unsupervised learning. However, it should improve its robustness to noise and learn the spatiotemporal and semantic representations of categories, just like cardiologists. This article proposes a patient-level adversarial spatiotemporal contrastive learning (ASTCL) framework, which includes ECG augmentations, an adversarial module, and a spatiotemporal contrastive module. Based on the ECG noise attributes, two distinct but effective ECG augmentations, ECG noise enhancement, and ECG noise denoising, are introduced. These methods are beneficial for ASTCL to enhance the robustness of the DNN to noise. This article proposes a self-supervised task to increase the antiperturbation ability. This task is represented as a game between the discriminator and encoder in the adversarial module, which pulls the extracted representations into the shared distribution between the positive pairs to discard the perturbation representations and learn the invariant representations. The spatiotemporal contrastive module combines spatiotemporal prediction and patient discrimination to learn the spatiotemporal and semantic representations of categories. To learn category representations effectively, this article only uses patient-level positive pairs and alternately uses the predictor and the stop-gradient to avoid model collapse. To verify the effectiveness of the proposed method, various groups of experiments are conducted on four ECG benchmark datasets and one clinical dataset compared with the state-of-the-art methods. Experimental results showed that the proposed method outperforms the state-of-the-art methods.