Explore the vast range of titles available.

In electroencephalography (EEG)-based emotion recognition systems, the distribution between the training samples and the testing samples may be mismatched if they are sampled from different experimental sessions or subjects because of user fatigue, different electrode placements, varying impedances, etc. Therefore, it is difficult to directly classify the EEG patterns with a conventional classifier. The domain adaptation method, which is aimed at obtaining a common representation across training and test domains, is an effective method for reducing the distribution discrepancy. However, the existing domain adaptation strategies either employ a linear transformation or learn the nonlinearity mapping without a consistency constraint; they are not sufficiently powerful to obtain a similar distribution from highly non-stationary EEG signals. To address this problem, in this paper, a novel component, called the subspace alignment auto-encoder (SAAE), is proposed. Taking advantage of both nonlinear transformation and a consistency constraint, we combine an auto-encoder network and a subspace alignment solution in a unified framework. As a result, the source domain can be aligned with the target domain together with its class label, and any supervised method can be applied to the new source domain to train a classifier for classification in the target domain, as the aligned source domain follows a distribution similar to that of the target domain. We compared our SAAE method with six typical approaches using a public EEG dataset containing three affective states: positive, neutral, and negative. Subject-to-subject and session-to-session evaluations were performed. The subject-to-subject experimental results demonstrate that our component achieves a mean accuracy of 77.88% in comparison with a state-of-the-art method, TCA, which achieves 73.82% on average. In addition, the average classification accuracy of SAAE in the session-to-session evaluation for all the 15 subjects in a dataset is 81.81%, an improvement of up to 1.62% on average as compared to the best baseline TCA. The experimental results show the effectiveness of the proposed method relative to state-of-the-art methods. It can be concluded that SAAE is a useful and effective tool for decreasing domain discrepancy and reducing performance degradation across subjects and sessions in the EEG-based emotion recognition field. •Deep auto-encoder network and subspace alignment solution are combined to constrain the distribution discrepancy.•Demonstration of the improved classification accuracy compared with several state-of-the-art domain adaptation techniques.•Illustration of the suitability of the SAAE for non-stationary EEG signal classification.

Hereditary hemorrhagic telangiectasia (HHT) coexisting with moyamoya disease (MMD) is exceptionally rare. We report the first case of a 45-year-old female harboring two genetic variants implicated in vascular disease: a pathogenic mutation in ACVRL1 (c.1231C>T, p.Arg411Trp) and a novel variant of uncertain significance in RNF213 (c.13685C>T, p.Pro4562Leu). This case is remarkable for the concurrent manifestation of HHT-associated peripheral telangiectasia and MMD-characteristic intracranial arterial stenosis, suggesting a possible synergistic interaction between variants affecting distinct vascular signaling pathways. These findings offer new insights into the genetic mechanisms underlying complex hereditary vascular disorders and emphasize the importance of comprehensive genetic testing in diagnosing atypical vascular phenotypes.

This paper introduces a novel unsupervised domain adaptation (UDA) method, MeTa Discriminative Class-Wise MMD (MCWMMD), which combines meta-learning with a Class-Wise Maximum Mean Discrepancy (MMD) approach to enhance domain adaptation. Traditional MMD methods align overall distributions but struggle with class-wise alignment, reducing feature distinguishability. MCWMMD incorporates a meta-module to dynamically learn a deep kernel for MMD, improving alignment accuracy and model adaptability. This meta-learning technique enhances the model’s ability to generalize across tasks by ensuring domain-invariant and class-discriminative feature representations. Despite the complexity of the method, including the need for meta-module training, it presents a significant advancement in UDA. Future work will explore scalability in diverse real-world scenarios and further optimize the meta-learning framework. MCWMMD offers a promising solution to the persistent challenge of domain adaptation, paving the way for more adaptable and generalizable deep learning models.

This paper explores an alternative way for analyzing static Functional Connectivity (FC) in functional Magnetic Resonance Imaging (fMRI) data across multiple timescales using a class of adaptive frequency-based methods referred to as Multivariate Mode Decomposition (MMD). The proposed method decomposes fMRI into their intrinsic multivariate oscillatory components through a fully data-driven approach, and enables the isolation of intrinsic neurophysiological activation patterns across multiple frequency bands from other interfering components. Unlike other methods, this approach is inherently equipped to handle the multivariate nature of fMRI data by aligning frequency information across multiple regions of interest. The proposed method was validated using three fMRI experiments: resting-state, motor and gambling experiments. Results demonstrate the capability of the methodology to extract reliable and reproducible FC patterns across individuals while uncovering unique connectivity features at different times scales. In addition, the results evidence the effect of the different task on the spectral organization of FC patterns, highlighting the importance of multiscale analysis for understanding functional interactions.

To solve the problem of tendon adhesion after an operation, an anti‐adhesion membrane is designed, which can inhibit the exogenous healing of tendons and promote endogenous healing. Here, poly[3(S)‐methyl‐morpholine‐2,5‐dione‐co‐lactic acid] P(MMD‐co‐LA) containing alanine units is obtained by melt ring‐opening polymerization (ROP). It can effectively reduce the production of acidic degradation products that cause aseptic inflammation. Then a kind of oriented P(MMD‐co‐LA)/deferoxamine (DFO) anti‐adhesion membrane with good mechanical properties, cell adhesion properties, and induced macrophage polarization properties is constructed by electrospinning. In vitro and in vivo studies have shown that oriented fibrous membranes can down‐regulate the expression of inflammatory cytokines. In addition, the fibrous membrane can up‐regulate the expression of angiogenesis‐related genes, namely HIF1‐α, SDF‐1α, and VEGF, by releasing DFO in situ, thus promoting the revascularization of tendon defects and providing nutritional support for endogenous tendon healing. This method provides a new barrier strategy to reduce tendon adhesion through anti‐inflammation combined with vascularization to inhibit exogenous tendon healing while promoting tendon endogenous healing. The results provide essential insights into the corresponding regulation of the tendon healing microenvironment. Poly[3(S)‐methyl‐morpholine‐2,5‐dione‐co‐lactic acid] P(MMD‐co‐LA) containing alanine units is obtained by melt ring‐opening polymerization (ROP). The oriented P(MMD‐co‐LA)/deferoxamine (DFO) anti‐adhesion membrane is constructed by electrospinning. This method provides a new barrier strategy to reduce tendon adhesion by inhibiting exogenous tendon healing and promoting endogenous tendon healing, which provides essential insights for the corresponding regulation of the tendon healing microenvironment.

In machinery fault diagnosis, labeled data are always difficult or even impossible to obtain. Transfer learning can leverage related fault diagnosis knowledge from fully labeled source domain to enhance the fault diagnosis performance in sparsely labeled or unlabeled target domain, which has been widely used for cross domain fault diagnosis. However, existing methods focus on either marginal distribution adaptation (MDA) or conditional distribution adaptation (CDA). In practice, marginal and conditional distributions discrepancies both have significant but different influences on the domain divergence. In this paper, a dynamic distribution adaptation based transfer network (DDATN) is proposed for cross domain bearing fault diagnosis. DDATN utilizes the proposed instance-weighted dynamic maximum mean discrepancy (IDMMD) for dynamic distribution adaptation (DDA), which can dynamically estimate the influences of marginal and conditional distribution and adapt target domain with source domain. The experimental evaluation on cross domain bearing fault diagnosis demonstrates that DDATN can outperformance the state-of-the-art cross domain fault diagnosis methods.

Introduction Increasing access to multi‐month dispensing (MMD) of antiretroviral therapy (ART) supports treatment continuity and viral load suppression for people living with HIV (PLHIV) and reduces burden on health facilities. During the COVID‐19 response, PEPFAR worked with ministries of health to scale up MMD and expand eligibility to new groups of PLHIV, including children and pregnant/breastfeeding women. We analysed PEPFAR program data to understand the impact of the policy changes on actual practice. Methods We conducted a desk review in 21 PEPFAR‐supported countries to identify and collect official documentation released between March and June 2020 addressing changes to MMD guidance during the COVID‐19 response. MMD coverage, the proportion of all ART clients on MMD, was assessed in the calendar quarters preceding the COVID‐19 response (Q4 2019, October–December 2019; and Q1, January–March 2020) and the quarters following the start of the response (Q2 2020, April–June 2020; Q3 2020, July–September, 2020; Q4 2020, October–December 2020). We used the two‐proportion Z‐test to test for differences in MMD coverage pre‐COVID‐19 (Q4 2019) and during implementation of COVID‐19 policy adaptations (Q2 2020). Results and discussion As of June 2020, 16 of the 21 PEPFAR‐supported countries analysed adapted MMD policy or promoted intensified scale‐up of MMD in response to COVID‐19. MMD coverage for all clients on ART grew from 49% in Q4 2019 pre‐COVID‐19 to 72% in Q2 2020 during COVID‐19; among paediatric clients (< 15), MMD coverage increased from 27% to 51% in the same period. Adaptations to MMD policy were associated with a significantly accelerated growth in the proportion of clients on MMD (p < 0.001) for all populations, irrespective of age and dispensing interval. Conclusions Access to MMD markedly expanded during the COVID‐19 pandemic, supporting treatment continuity while mitigating exposure to COVID‐19 at health facilities. This model is beneficial in public health emergencies and during disruptions to the healthcare system. Outside emergency contexts, expanded MMD eligibility extends client‐centred care to previously excluded populations. The success in expanding MMD access during COVID‐19 should motivate countries to recommend broader MMD access as a new standard of care.

Moyamoya disease (MMD) is a unique cerebrovascular occlusive disease with abnormal vascular hyperplasia, which causes cerebrovascular accidents like intracranial arteriosclerosis. This study aimed to explore whether plasma apelin levels are related to good collateral circulation in ischemic diseases, which may be higher in patients with MMD than middle cerebral artery (MCA) occlusion or healthy controls, and may have a connection with the MMD grades. We recruited 68 MMD patients and 25 MCA occlusion patients diagnosed by angiography, including 29 patients without cerebrovascular problems as controls. We examined the plasma apelin, serum nitric oxide (NO), and vascular endothelial growth factor (VEGF) levels of all subjects by ELISA kit. We compared the relationship between apelin, NO, and VEGF in the blood of three groups, to explore the relationship. We also investigated whether the plasma apelin-13, apelin-17, and apelin-36 levels correlate with the MMD classification. Univariate analyses indicated that the MMD group had the higher plasma apelin-13, apelin-17, apelin-36, and serum NO levels than the MCA occlusion and healthy control groups. Binary logistic regression analyses further showed that the apelin-13 level was substantially higher in MMD patients than in MCA occlusion patients. Patients with MMD were significantly younger than patients with MCA occlusion by their mean ages. Linear regression analyses were performed to compare apelin levels between different grades of the patients with MMD. Apelin-13, apelin-17, and apelin-36 levels increased with the gradual increase of compensation grades level independent of NO and VEGF. Apelin-13 and apelin-36 showed a positive effect on the compensation scores in MMD. Our study demonstrated that apelin-13 was significantly increased in patients with MMD than patients with MCA occlusion independent of NO and VEGF. Moreover, plasma apelin-13, apelin-17, and apelin-36 levels increase with the grades of MMD.

As industrial development increases, electric machine systems are more widely used in industrial production. Rolling bearings play a key role in machine systems and so the prevention of faults in rolling bearings is more important than ever before. Recently, with the development of artificial intelligence, neural networks have been used to monitor the remaining useful life of rolling bearings. However, there are two problems with this technique. First, a network trained by data for a single operating condition (source domain) cannot predict the remaining useful life of bearings under a different operating condition (target domain), such as a different load or speed. Second, a large number of labeled data are needed for network training, but the acquisition of labeled data for different operating conditions is a challenging task. To address these problems, this paper proposes a domain-adaptive adversarial network, in which a transfer learning strategy and maximum mean discrepancy algorithm are used for network optimization, so that remaining useful life can be predicted without labeled data in target domain training. Our results confirm that a model trained by source domain data alone cannot predict the remaining useful life of bearings under different conditions, but the domain-adaptive adversarial network can accurately predict remaining useful life for varying operating conditions. The method proposed also exhibits good performance even if there are noises in the signals.

Background: The impact of gene mutations typically associated with myelodysplastic syndrome (MDS) in acute myeloid leukemia (AML) with NPM1 mutation is unclear. Methods: Using a cohort of 107 patients with NPM1-mutated AML treated with risk-adapted therapy, we compared survival outcomes of patients without MDS-related gene mutations (group A) with those carrying concurrent FLT3-ITD (group B) or with MDS-related gene mutations (group C). Minimal measurable disease (MMD) status assessed by multiparameter flow cytometry (MFC), polymerase chain reaction (PCR), and/or next-generation sequencing (NGS) were reviewed. Results: Among the 69 patients treated intensively, group C showed significantly inferior progression-free survival (PFS, p < 0.0001) but not overall survival (OS, p = 0.055) compared to group A. Though groups A and C had a similar MMD rate, group C patients had a higher relapse rate (p = 0.016). Relapse correlated with MMD status at the end of cycle 2 induction (p = 0.023). Survival of group C patients was similar to that of group B. Conclusion: MDS-related gene mutations are associated with an inferior survival in NPM1-mutated AML.