A document by Liu Jiasen. Click on the document to view its contents.

Hyperspectral images contain rich spatial and spectral information, which provides a strong basis for distinguishing different land-cover objects. Therefore, hyperspectral image classification has been a hot research topic. With the advent of deep learning, convolutional neural networks (CNNs) have become a popular method for hyperspectral image classification. However, CNN has strong local feature extraction ability but cannot deal with long-distance dependence well. Vision Transformer (ViT) is a recent development that can address this limitation, but it is not effective in extracting local features and has low computational efficiency. To overcome these drawbacks, we propose a hybrid classification network that combines the strengths of both CNN and ViT, names Spatial-Spectral Former(SSF). The shallow layer employs 3D convolution to extract local features and reduce data dimensions. The deep layer employs a spectral-spatial transformer module for global feature extraction and information enhancement in spectral and spatial dimensions. Our proposed model achieves promising results on widely used public HSI datasets compared to other deep learning methods, including CNN, ViT, and hybrid models.

Multivariate time series data classification has a wide range of applications in reality. With rapid development of deep learning, convolutional networks are widely used in this task and have achieved the current best performance. However, due to high difficulty and cost of collecting this type of data, labeled data is still scarce. In some tasks, the model shows overfitting, resulting in relatively poor classification performance. In order to improve the classification effect under such situation, this paper proposes a novel classification method based on transfer learning - similar domain knowledge transfer (call SDKT for short). Firstly, we designed a multivariate time series domain distance calulation method (call MTSDDC for short), which helped selecting the source domain that are most similar to target domain; Secondly, we used ResNet as a pre-trained classifier, transfered the parameters of the similar domain network to the target domain network and continued to fine-tune the parameters. To verify our method, we conducted experiments on several public datasets. Our study has also shown that the transfer effect from the source domain to the target domain is highly negatively correlated with the distance between them, with an average pearson coefficient of -0.78. For the transfer of most similar source domain, compared to the ResNet model without transfer and the current best model, the average accuracy improvement on the datasets we used is 4.01% and 1.46% respectively.

Personalized health monitoring and prediction have become essential for improving health- care delivery, especially with the growing prevalence of chronic diseases and an aging population. Deep learning (DL) has emerged as a promising approach for developing personalized health mon- itoring systems that can predict health outcomes accurately and efficiently. With the increasing availability of personal health data, DL-based methods have emerged as a promising approach to improve healthcare delivery by providing accurate and timely predictions of health outcomes. This article provides a comprehensive review of the recent developments in the application of DL for personalized health monitoring and prediction. It summarizes various DL architectures and their applications for personalized health monitoring, including wearable devices, electronic health records, and social media data. Furthermore, the article also explores the challenges and future directions for the application of DL in personalized health monitoring. valuable insights into the potential of DL for personalized health monitoring and prediction.