It is vital to develop general models that can be shared across subjects and sessions in the real-world deployment of electroencephalogram (EEG) emotion recognition systems. Many prior studies have exploited domain adaptation algorithms to alleviate the inter-subject and inter-session discrepancies of EEG distributions. However, these methods only aligned the global domain divergence, but overlooked the local domain divergence with respect to each emotion category. This degenerates the emotion-discriminating ability of the domain invariant features. In this paper, we argue that aligning the EEG data within the same emotion categories is important for generalizable and discriminative features. Hence, we propose the dynamic domain adaptation (DDA) algorithm where the global and local divergences are disposed by minimizing the global domain discrepancy and local subdomain discrepancy , respectively. To tackle the absence of emotion labels in the target domain, we introduce a dynamic training strategy where the model focuses on optimizing the global domain discrepancy in the early training steps, and then gradually switches to the local subdomain discrepancy. The DDA algorithm is formally implemented as an unsupervised version and a semi-supervised version for different experimental settings. Based on the coarse-to-fine alignment, our model achieves the average peak accuracy of 91.08%, 92.89% on SEED, and 81.58%, 80.82% on SEED-IV in the cross-subject and cross-session scenarios, respectively. As open science, the code is available at https://github.com/JC-Journal-Club/DDA .

Emotion recognition is essential in the diagnosis and rehabilitation of various mental diseases. In the last decade, electroencephalogram (EEG)-based emotion recognition has been intensively investigated due to its prominative accuracy and reliability, and graph convolutional network (GCN) has become a mainstream model to decode emotions from EEG signals. However, the electrode relationship, especially long-range electrode dependencies across the scalp, may be underutilized by GCNs, although such relationships have been proven to be important in emotion recognition. The small receptive field makes shallow GCNs only aggregate local nodes. On the other hand, stacking too many layers leads to over-smoothing. To solve these problems, we propose the pyramidal graph convolutional network (PGCN), which aggregates features at three levels: local, mesoscopic, and global. First, we construct a vanilla GCN based on the 3D topological relationships of electrodes, which is used to integrate two-order local features; Second, we construct several mesoscopic brain regions based on priori knowledge and employ mesoscopic attention to sequentially calculate the virtual mesoscopic centers to focus on the functional connections of mesoscopic brain regions; Finally, we fuse the node features and their 3D positions to construct a numerical relationship adjacency matrix to integrate structural and functional connections from the global perspective. Experimental results on four public datasets indicate that PGCN enhances the relationship modelling across the scalp and achieves stateof-the-art performance in both subject-dependent and subjectindependent scenarios. Meanwhile, PGCN makes an effective trade-off between enhancing network depth and receptive fields while suppressing the ensuing over-smoothing. Our codes are publicly accessible at https://github.com/Jinminbox/PGCN .

An outbreak of coronavirus disease (COVID-19) in Wuhan, China caused by SARS-CoV-2 has led to a serious epidemic in China and other countries, resulting in worldwide concern. With active efforts of prevention and control, more and more patients are being discharged. However, how to manage these patients normatively is still challenging. This paper reports an asymptomatic discharged patient with COVID-19 who retested positive for SARS-CoV-2, which arouses concern regarding the present discharge standards of COVID-19.

Background and Aims: Anti‐tuberculosis drugs remain as an important cause of drug‐induced liver injury (DILI) worldwide. Adverse drug reactions reduce the effectiveness of treatment. We aimed to determine the incidence and risk factors associated with anti‐tuberculosis DILI (ATDILI).Methods: Using established criteria and causality assessment methods, risk factors for ATDILI were identified in a contemporary cohort and validated in another cohort prospectively. Independent determinants of ATDILI were identified using Cox regression analysis.Results: In the derivation cohort (n = 3155), 170 (5.4%) developed ATDILI of which 27 (15.9%) developed jaundice; 9(5.3%) developed acute liver failure (ALF) and 3 died. Among HBsAg positive patients, 11/27 (40.7%) of ATDILI developed after 3 months of starting treatment. In addition, of 218 (6.9%) who developed raised alanine transferase (ALT) levels ≥3 times upper limit normal, 193 (88.5%) resolved and 25 (11.4%) progressed to DILI. Age (HR = 1.014, 95% CI: 1.005‐1.023), baseline ALT (HR = 1.014, 95% CI: 1.003‐1.024), haemoglobin (HR = 1.011, 95% CI: 1.002‐1.020) and HBsAg positivity (HR = 1.516, 95% CI: 1.004‐2.290) were independent risk factors for DILI. In the second cohort (n = 1497) of which 85 (5.7%) developed ATDILI. Age (HR = 1.029, 95% CI: 1.003‐1.056), baseline AST (HR = 1.036, 95% CI: 1.010‐1.062), previous TB treatment (HR = 3.894, 95% CI: 1.304‐11.625) and active drinking (HR = 3.624, 95% CI: 1.147‐11.454) were risk factors for developing jaundice.Conclusion: Elevation of ALT of ≥3 × ULN during anti‐TB treatment resolves in the vast majority without developing serious consequences. In two cohorts involving 4652 patients, incidence of ALF and death because of ATDILI are low. Age, baseline ALT, haemoglobin and HBsAg positivity are risk factors for the development of DILI and these inform monitoring and management of these patients.

Layered double hydroxide (LDH) is a ‘sandwich’-like two-dimensional clay material that has been systematically investigated for biomedical application in the past two decades. LDH is an alum-similar adjuvant, which has a well-defined layered crystal structure and exhibits high adjuvanticity. The unique structure of LDH includes positively charged layers composed of divalent and trivalent cations and anion-exchangeable interlayer galleries. Among the many variants of LDH, MgAl-LDH (the cationic ions are Mg²⁺ and Al³⁺) has the highest affinity to antigens, bioadjuvants and drug molecules, and exhibits superior biosafety. Past research studies indicate that MgAl-LDH can simultaneously load antigens, bioadjuvants and molecular drugs to amplify the strength of immune responses, and induce broad-spectrum immune responses. Moreover, the size and dispersity of MgAl-LDH in biological environments can be well controlled to actively deliver antigens to the immune system, realizing the rapid induction and maintenance of durable immune responses. Furthermore, the functionalization of MgAl-LDH nanoadjuvants enables it to capture antigens in situ and induce personalized immune responses, thereby more effectively overcoming complex diseases. In this review, we comprehensively summarize the development and application of MgAl-LDH nanoparticles as a vaccine adjuvant, demonstrating that MgAl-LDH is the most potential adjuvant for clinical application.

Lung cancer is one of the leading causes of death over the world. Detecting and identifying malignant nodules on chest computed tomography (CT) plays an important role in the diagnosis and treatment of lung cancer. Computer-aided diagnosis (CAD) systems have been developed to identify lung nodules. However, the problem of a high false positive rate is still not well solved. In this paper, we propose a novel multi-task convolutional neural network (MT-CNN) framework to identify malignant nodules from benign nodules on chest CT scans. MT-CNN learns three-dimensional (3-D) lung nodule characteristics from nine two-dimensional (2-D) views, which are decomposed from different angles of each nodule. Each of 2-D MT-CNN model consists of two branches, one is the nodule classification branch (main task) and the other is the image reconstruction branch (auxiliary task). The motivation of the auxiliary task is to preserve more microscopic information in the hierarchical structure of CNN, which is beneficial to malignant nodule identification. The final classification result is obtained by integrating nine 2-D models. We test our method on the benchmark LUNA-16 and LIDC-IDRI datasets and compare it with state-of-the-art models. MT-CNN achieves the lowest false positive rate (3.2%) and highest AUC (97.3%) in LUNA-16 dataset and achieves an AUC of 95.59% in LIDC-IDRI. These results demonstrate the advantage of our method.