A newly emerged coronavirus (COVID-19) seriously threatens human life and health worldwide. In coping and fighting against COVID-19, the most critical step is to effectively screen and diagnose infected patients. Among them, chest X-ray imaging technology is a valuable imaging diagnosis method. The use of computer-aided diagnosis to screen X-ray images of COVID-19 cases can provide experts with auxiliary diagnosis suggestions, which can reduce the burden of experts to a certain extent. In this study, we first used conventional transfer learning methods, using five pre-trained deep learning models, which the Xception model showed a relatively ideal effect, and the diagnostic accuracy reached 96.75%. In order to further improve the diagnostic accuracy, we propose an efficient diagnostic method that uses a combination of deep features and machine learning classification. It implements an end-to-end diagnostic model. The proposed method was tested on two datasets and performed exceptionally well on both of them. We first evaluated the model on 1102 chest X-ray images. The experimental results show that the diagnostic accuracy of Xception + SVM is as high as 99.33%. Compared with the baseline Xception model, the diagnostic accuracy is improved by 2.58%. The sensitivity, specificity and AUC of this model reached 99.27%, 99.38% and 99.32%, respectively. To further illustrate the robustness of our method, we also tested our proposed model on another dataset. Finally also achieved good results. Compared with related research, our proposed method has higher classification accuracy and efficient diagnostic performance. Overall, the proposed method substantially advances the current radiology based methodology, it can be very helpful tool for clinical practitioners and radiologists to aid them in diagnosis and follow-up of COVID-19 cases.
By trying to solve the issue of identifying multiple types of intrusion vibration signals collected by distributed vibrating fiber optic sensors, this study investigates the signal identification and feature extraction of intrusion signals, and proposes an optical fiber vibration signal (OFVS) identification method based on deep learning. The external vibration signal is collected by the Sagnac fiber optic interferometer, and then denoised by spectral subtraction. Endpoint detection is carried out by combining the short-time logarithmic energy method and the spectral entropy method. Finally, the equal-length signal containing valid information is intercepted and the corresponding preprocessing is carried out. The method for feature processing incorporates the strong feature learning capability of the long-short-term memory (LSTM) and the great short-term feature extraction capability of the convolutional neural network (CNN). At the same time, to further enhance the signal feature identification, a convolutional block attention module (CBAM) is introduced to perform adaptive feature refinement on the signal. In summary, a network model combining CNN, LSTM, and CBAM is proposed to process the signal features, and finally, the multi-layer perceptron (MLP) is used to complete the task of classification and recognition of multi-type intrusion signals. The experimental findings indicate that the OFVS method of CNN-CBAM-LSTM can effectively identify four kinds of OFVS, and the overall average recognition accuracy reaches 97.9%. Walking and knocking signals among them are recognized with over 99% accuracy.
In recent years, the application of federated learning to medical image classification has received much attention and achieved some results in the study of semi-supervised problems, but there are problems such as the lack of thorough study of labeled data, and serious model degradation in the case of small batches in the face of the data category imbalance problem. In this paper, we propose a federated learning method using a combination of regularization constraints and pseudo-label construction, where the federated learning framework consists of a central server and local clients containing only unlabeled data, and labeled data are passed from the central server to each local client to take part in semi-supervised training. We first extracted the class imbalance factors from the labeled data to participate in the training to achieve label constraints, and secondly fused the labeled data with the unlabeled data at the local client to construct augmented samples, looped through to generate pseudo-labels. The purpose of combining these two methods is to select fewer classes with higher probability, thus providing an effective solution to the class imbalance problem and improving the sensitivity of the network to unlabeled data. We experimentally validated our method on a publicly available medical image classification data set consisting of 10,015 images with small batches of data. Our method improved the AUC by 7.35% and the average class sensitivity by 1.34% compared to the state-of-the-art methods, which indicates that our method maintains a strong learning capability even with an unbalanced data set with fewer batches of trained models.