Recently, the recurrence of positive SARS-CoV-2 viral RNA in recovered COVID-19 patients is receiving more attention. Herein we report a cohort study on the follow-up of 182 recovered patients under medical isolation observation. Twenty (10.99%) patients out of the 182 were detected to be SARS-CoV-2 RNA positive (re-positives), although none showed any clinical symptomatic recurrence, indicating that COVID-19 responds well to treatment. Patients aged under 18 years had higher re-positive rates than average, and none of the severely ill patients re-tested positive. There were no significant differences in sex between re-positives and non-re-positives. Notably, most of the re-positives turned negative in the following tests, and all of them carried antibodies against SARS-CoV-2. This indicates that they might not be infectious, although it is still important to perform regular SARS-CoV-2 RNA testing and follow-up for assessment of infectivity. The findings of this study provide information for improving the management of recovered patients, and for differentiating the follow-up of recovered patients with different risk levels.

In the detection of surface floating garbage, the existence of complex backgrounds and the small target sizes make the surface floating garbage easy to mis-detect. Existing approaches cannot yet provide a solution to the aforementioned problems and they are typically limited to addressing specific issues. This paper proposes a PC-Net algorithm for floating garbage detection. First, a pyramid anchor generation approach is proposed, which makes the anchor to be generated centrally near the target and reduces the interference of background information in the anchor generation. Then, in the RoI Pooling feature map import stage, the classification map is used as the feature map. This approach generates feature maps with a higher resolution and more distinct features, thereby enhancing the feature information of small targets and enhancing the classification accuracy. Experimental results on floating garbage dataset indicate that the average detection accuracy of the proposed approach is 86.4%. Compared with existing detection approaches, such as Faster R-CNN, YOLOv3, YOLOX, and Dynamic R-CNN, the average accuracy of detection is increased by 4.1%, 3.6%, and 2.8%, respectively.

In recent years, deep learning has witnessed astonishing success in the field of remote sensing in images. Generally, deep learning requires a large amount of labeled training data. Nevertheless, in remote sensing, sufficient labeled data are scarce because labeled data are often difficult, expensive, or time-consuming to obtain. To address these problems, we propose a deep curriculum learning semi-supervised framework (DCLSSF) for remote sensing image scene classification. This framework employs a multimodal deep curriculum learning method which can realize the classification of images on a range of easy–difficult. Specifically, by utilizing multiple pretrained networks to extract multiple deep features of images as their multimodal feature representations, it can comprehensively mine the information from labeled and unlabeled images from diverse perspectives. Subsequently, a feature fusion method is used on deep features of different modalities to obtain deep fusion features with a strong discrimination ability and low dimensionality. Finally, the multimodal deep features are fed into multimodal curriculum learning methods for classification. Multimodal curriculum learning can integrate the easy curricula recommended by each modal according to the order of the samples of each modal and then learn step by step. Experiments on three publicly available datasets (UC Merced, AID, and NWPU-RESISC45) show that the semi-supervised classification framework achieves high accuracy rates (99.14%, 97.95%, and 93.01%), even surpassing those of the most supervised classification methods. The DCLSSF method can not only fully exploit the rich features extracted by the multimodal deep learning network but can also perform the semi-supervised classification of unlabeled samples in a range of easy–difficult.

The advancement in satellite image sensors has enabled the acquisition of high-resolution remote sensing (HRRS) images. However, interpreting these images accurately and obtaining the computational power needed to do so is challenging due to the complexity involved. This manuscript proposed a multi-stream convolutional neural network (CNN) fusion framework that involves multi-scale and multi-CNN integration for HRRS image recognition. The pre-trained CNNs were used to learn and extract semantic features from multi-scale HRRS images. Feature extraction using pre-trained CNNs is more efficient than training a CNN from scratch or fine-tuning a CNN. Discriminative canonical correlation analysis (DCCA) was used to fuse deep features extracted across CNNs and image scales. DCCA reduced the dimension of the features extracted from CNNs while providing a discriminative representation by maximizing the within-class correlation and minimizing the between-class correlation. The proposed model has been evaluated on NWPU-RESISC45 and UC Merced datasets. The accuracy associated with DCCA was 10% and 6% higher than discriminant correlation analysis (DCA) in the NWPU-RESISC45 and UC Merced datasets. The advantage of DCCA was better demonstrated in the NWPU-RESISC45 dataset due to the incorporation of richer within-class variability in this dataset. While both DCA and DCCA minimize between-class correlation, only DCCA maximizes the within-class correlation and, therefore, attains better accuracy. The proposed framework achieved higher accuracy than all state-of-the-art frameworks involving unsupervised learning and pre-trained CNNs and 2–3% higher than the majority of fine-tuned CNNs. The proposed framework offers computational time advantages, requiring only 13 s for training in NWPU-RESISC45, compared to a day for fine-tuning the existing CNNs. Thus, the proposed framework achieves a favourable balance between efficiency and accuracy in HRRS image recognition.