The intelligent fault diagnosis model has made a significant development, whose high-precision results rely on a large amount of labeled data. However, in the actual industrial environment, it is very difficult to obtain a large amount of labeled data. It will make it difficult for the fault diagnosis model to converge with limited labeled industrial data. To address this paradox, we propose a novel unsupervised domain adaptation framework (M-Net) for fault diagnosis of rotating machinery, which only requires unlabeled industrial data. The M-Net will be pretrained using the labeled data, which can be accessed through the labs. In this stage, we propose a multi-scale feature extractor that can extract and fuse multi-scale features. This operation will generalize the features further. Then, we will align the distribution of the labeled data and unlabeled industrial data using the generator model based on multi-kernel maximum mean discrepancy. This will reduce the distribution distance between the labeled data and the unlabeled industrial data. For now, the unsupervised domain adaptation problem has shifted to a semi-supervised domain adaptation problem. The results, obtained through experimental comparison, demonstrate that the M-Net can achieve an accuracy of over 99.99% with labeled data and a maximum transfer accuracy of over 99% with unlabeled industrial data.

In the burgeoning metaverse for consumer health (MCH), medical image segmentation methods with high accuracy and generalization capability are essential to drive personalized healthcare solutions and enhance the patient experience. To address the inherent challenges of capturing complex structures and features in medical image segmentation, we propose a convolutional neural network (CNN) and multi-layer-perceptron (MLP) mixed module named HCMM, which hierarchically incorporates local priors of CNN into fully-connected (FC) layers, ingeniously capturing specific details and a broader range of contextual information of the focused object from diverse perspectives. Then, we propose an MLP-based information fusion module (MIF) designed to dynamically merge feature maps of varying levels from different pathways, enhancing feature expression and discriminative power. Based on the above-proposed modules, we design a novel segmentation model, HCMMNet, which can adeptly capture feature maps from input medical images at different scales and perspectives. Through comparative experiments, we demonstrate the outstanding performance of the HCMMNet for medical image segmentation on three publicly available datasets and one self-organized dataset. Notably, our HCMMNet showcases remarkable efficacy while maintaining an extraordinarily lightweight profile, weighing in at a mere 3M, rendering it ideal for MCH application.

Physical oceanography models rely heavily on grid discretization. It is known that unstructured grids perform well in dealing with boundary fitting problems in complex nearshore regions. However, it is time-consuming to find a set of unstructured grids in specific ocean areas, particularly in the case of land areas that are frequently changed by human construction. In this work, an attempt was made to use machine learning for the optimization of the unstructured triangular meshes formed with Delaunay triangulation in the global ocean field, so that the triangles in the triangular mesh were closer to equilateral triangles, the long, narrow triangles in the triangular mesh were reduced, and the mesh quality was improved. Specifically, we used Delaunay triangulation to generate the unstructured grid, and then developed a K-means clustering-based algorithm to optimize the unstructured grid. With the proposed method, unstructured meshes were generated and optimized for global oceans, small sea areas, and the South China Sea estuary to carry out data experiments. The results suggested that the proportion of triangles with a triangle shape factor greater than 0.7 amounted to 77.80%, 79.78%, and 79.78%, respectively, in the unstructured mesh. Meanwhile, the proportion of long, narrow triangles in the unstructured mesh was decreased to 8.99%, 3.46%, and 4.12%, respectively.