Search

In federated learning,the local data distribution of users changes with the location and preferences of users,the data under the non-independent and identical distributed(Non-IID) data may lack data of some label categories,which significantly affects the update rate and the performance of the global model in federated aggregation.To solve this problem,a federated data augmentation based on conditional generative adversarial network(FDA-cGAN) algorithm is proposed,which can amplify data from participants with skewed data without compromising user privacy,and greatly improve the performance of the algorithm with Non-IID data.Experimental results show that,compared with the current mainstream federated average algorithm,under the Non-IID data setting,the prediction accuracy of MNIST and CIFAR-10 data sets improves by 1.18% and 14.6%,respectively,which demonstrates the effectiveness and practicability of the proposed algorithm for Non-IID data problems in federated learning.

Background: Leukemia is the tenth most common cause of cancer death worldwide and one of the most important causes of disability. To understand the current status and changing trends of the disease burden of leukemia at the global, regional, and national levels, and to provide a scientific basis for the development of leukemia prevention and treatment strategies. Methods: Based on open data from the Global Burden of Disease Study 2019 (GBD 2019), R software was used to calculate estimated annual percentage changes to estimate trends in the age‐standardized incidence (ASIR) and the age‐standardized disability‐adjusted life years (DALY) rate due to leukemia and its major subtypes from 1990 to 2019. Results: In 2019, globally, the number of incidences and DALYs of leukemia were 643.6 × 103 (587.0 × 103, 699.7 × 103) and 11,657.5 × 103 (10529.1 × 103, 12700.7 × 103), respectively. The ASIR (estimated annual percentage change (EAPC) = −0.37, 95%UI −0.46 to −0.28) and the age‐standardized DALY rate (EAPC = −1.72, 95%UI −1.80 to −1.65) of leukemia showed a decreasing trend from 1990 to 2019. The APC model analysis showed that the age effect of leukemia risk was a "U"‐shaped distribution of relative risk (RR) with increasing age from 1990 to 2019, globally. The time effect was an increase in incidence rate with increasing years but a decrease in DALY rate with increasing years. The cohort effects of both incidence and DALY rates tended to increase and then decrease with the development of the birth cohort. In 1990 and 2019, smoking, high body‐mass index, occupational exposure to benzene, and occupational exposure to formaldehyde were risk factors for DALY in leukemia, especially in areas with high SDI. Conclusions: From 1990 to 2019, the disease burden of leukemia showed a decreasing trend, but it is worth noting that its overall severity is still very high. The disease burden of leukemia varies greatly from region to region, and exclusive strategies for the prevention and treatment of leukemia should be developed according to the economic and cultural development of each region. [ABSTRACT FROM AUTHOR]

Background Major depressive disorder and bipolar disorder in adolescents are prevalent and are associated with cognitive impairment, executive dysfunction, and increased mortality. Early intervention in the initial stages of major depressive disorder and bipolar disorder can significantly improve personal health. Methods We collected 309 samples from the Adolescent Brain Cognitive Development study, including 116 adolescents with bipolar disorder, 64 adolescents with major depressive disorder, and 129 healthy adolescents, and employed a support vector machine to develop classification models for identification. We developed a multimodal model, which combined functional connectivity of resting-state functional magnetic resonance imaging and four anatomical measures of structural magnetic resonance imaging (cortical thickness, area, volume, and sulcal depth). We measured the performances of both multimodal and single modality classifiers. Results The multimodal classifiers showed outstanding performance compared with all five single modalities, and they are 100% for major depressive disorder versus healthy controls, 100% for bipolar disorder versus healthy control, 98.5% (95% CI: 95.4–100%) for major depressive disorder versus bipolar disorder, 100% for major depressive disorder versus depressed bipolar disorder and the leave-one-site-out analysis results are 77.4%, 63.3%, 79.4%, and 81.7%, separately. Conclusions The study shows that multimodal classifiers show high classification performances. Moreover, cuneus may be a potential biomarker to differentiate major depressive disorder, bipolar disorder, and healthy adolescents. Overall, this study can form multimodal diagnostic prediction workflows for clinically feasible to make more precise diagnose at the early stage and potentially reduce loss of personal pain and public society.

BackgroundMechanical ventilation (MV) is commonly used in anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis patients with serious conditions. However, little is known about the potential risk factors and long-term outcomes of anti-NMDAR encephalitis requiring MV, especially prolonged MV.MethodsThe data collected prospectively from 305 patients with anti-NMDAR encephalitis were retrospectively reviewed. The functional outcome was assessed using a modified Rankin scale (mRS) every 3 months.ResultsWe identified 62 (20.3%) patients who required MV. The most common reasons for MV were decreased consciousness and/or status epilepticus (SE). Among 60 patients analyzed, 27 patients required prolonged MV (>15 days). Prolonged MV primarily was based on the younger age, coma, tumor, and severe pneumonia. During the follow-up (median: 28 months, range: 3–87 months), 77% of patients required MV that exhibited a good outcome. In univariate analysis, prolonged MV, higher levels of C-reactive protein (CRP), and neutrophil-to-lymphocyte ratio (NLR) were found to be associated with poor neurological outcome at 6 months. Although the prolonged MV group exhibited a longer time to achieve a good outcome as compared to the short MV group (median duration 6 months vs. 3 months, p = 0.004), no significant difference was observed between the two groups about long-term outcomes.ConclusionIt is important to recognize that most anti-NMDAR encephalitis patients who required MV will achieve a favorable long-term outcomes, despite the longer duration of MV. Our results may help clinicians in the ventilator management of severe anti-NMDAR encephalitis patients.