Explore the vast range of titles available.

The deployment of Convolutional Neural Networks (CNNs) on entry-level Edge FPGAs is severely constrained by the scarcity of Digital Signal Processing (DSP) blocks, a phenomenon termed the “DSP Wall”. To circumvent this bottleneck, this paper presents AEMAC, a Software–Hardware Co-Designed accelerator architecture that decouples arithmetic computation from DSP availability. The proposed methodology synergizes a software-level Dynamic Integer Scaling strategy with a hardware-level Adaptive Error-Compensated Multiply-Accumulate unit. By mapping floating-point activations to an optimal integer domain and employing a DSP-free, LUT-based tri-mode datapath, the architecture achieves extreme resource efficiency. To mitigate the precision loss inherent in logic-based truncation, a statistical bias compensation mechanism is integrated into the accumulator chain. Experimental validation on a Xilinx Zynq-7020 FPGA demonstrates a strictly zero-DSP implementation with minimal logic utilization (100 LUTs). Post-implementation timing simulations confirm a dynamic power of 0.490 W for a 64-core cluster under worst-case random workloads, yielding a verified energy efficiency of 26.1 GOPS/W. Micro-level analysis confirms a 16.7% reduction in arithmetic Mean Absolute Error (MAE) compared to naive truncation. Furthermore, macro-level evaluation on the CIFAR-10 dataset reveals that the co-design strategy recovers system accuracy to 64.74%, outperforming the uncompensated baseline by 0.55% and achieving statistical comparability to floating-point baselines. To ensure absolute internal consistency, all hardware metrics are strictly validated via SAIF-based post-implementation simulations. Based on a conservative full-chip projection that incorporates a routing derating model, these internally consistent results establish AEMAC as a highly scalable and reliable solution for breaking the DSP wall in resource-constrained edge intelligence.

This study assessed the cost-effectiveness of different diabetic retinopathy (DR) screening strategies in rural regions in China by using a Markov model to make health economic evaluations. In this study, we determined the structure of a Markov model according to the research objectives, which required parameters collected through field investigation and literature retrieval. After perfecting the model with parameters and assumptions, we developed a Markov decision analytic model according to the natural history of DR in TreeAge Pro 2011. For this model, we performed Markov cohort and cost-effectiveness analyses to simulate the probabilistic distributions of different developments in DR and the cumulative cost-effectiveness of artificial intelligence (AI)-based screening and ophthalmologist screening for DR in the rural population with diabetes mellitus (DM) in China. Additionally, a model-based health economic evaluation was performed by using quality-adjusted life years (QALYs) and incremental cost-effectiveness ratios. Last, one-way and probabilistic sensitivity analyses were performed to assess the stability of the results. From the perspective of the health system, compared with no screening, AI-based screening cost more (the incremental cost was 37,257.76 RMB (approximately 5,211.31 US dollars)), but the effect was better (the incremental utility was 0.33). Compared with AI-based screening, the cost of ophthalmologist screening was higher (the incremental cost was 14,886.76 RMB (approximately 2,070.19 US dollars)), and the effect was worse (the incremental utility was -0.31). Compared with no screening, the incremental cost-effectiveness ratio (ICER) of AI-based DR screening was 112,146.99 RMB (15,595.47 US dollars)/QALY, which was less than the threshold for the ICER (< 3 times the per capita gross domestic product (GDP), 217,341.00 RMB (30,224.03 US dollars)). Therefore, AI-based screening was cost-effective, which meant that the increased cost for each additional quality-adjusted life year was merited. Compared with no screening and ophthalmologist screening for DR, AI-based screening was the most cost-effective, which not only saved costs but also improved the quality of life of diabetes patients. Popularizing AI-based DR screening strategies in rural areas would be economically effective and feasible and can provide a scientific basis for the further formulation of early screening programs for diabetic retinopathy.

Faba bean ( Vicia faba L.) is an important food legume crop. Salinity soils severely constrain the production of faba bean, however, the seed germination of faba bean, which is a vital plant growth stage, is sensitive to salinity. Planting improved varieties of faba bean, which exhibit salt tolerance in seed germination stage, is an optimal strategy for faba bean product. To investigate the genes dynamics during the seed germination stage under salinity, RNA-seq method was used to investigate genome-wide transcription profiles of two faba bean varieties with contrast salt-tolerance during the seed germination. A total of 4,486 differentially expressed genes (DEGs) were identified among the comparison of salt-tolerant variety Y134 and salt-sensitive variety Y078 treated with salinity or not. Of these, 1,410 candidate DEGs were identified as salt-stress response genes. Furthermore, 623 DEGs were identified as variety-specific response gene during seed germination at 16 h or 24 h with salt treatment. Based on the pathway enrichment according to the Kyoto Encyclopedia of Genes and Genomes database (KEGG), these DEGs involving in cell wall loosening (e.g., xyloglucan endotransglucosylase/hydrolase, chitinase, and expansin), hormone metabolism (e.g., LEA genes, genes associated with ABA or ethylene signal pathway), chromatin remodeling (e.g., chromatin structure proteins, LHP1), small interfering RNA pathway, etc., were significantly up-regulated in salt-tolerance variety with salt treatment, indicating that they play critical roles in regulation of seed germination. The results indicated that a clearer mechanism of gene regulation that regulates the seed germination responding to salinity in faba bean. These findings are helpful to increase the understanding of the salt tolerance mechanism of crops during seed germination, and provide valuable genetic resource for the breeding of salt-tolerant faba bean varieties in future.

Double negative (DN) B cells, defined as CD27-IgD- B cells, are increased in older healthy individuals as well as in patients with certain diseases. However, the relationship between DN B cells and Sjögren’s disease (SjD) in newly diagnosed patients remains unclear. Seventy SjD patients and thirty-six healthy controls (HCs) were recruited in this study. Peripheral blood B cell subsets were analyzed using flow cytometry. The percentage and absolute count of DN B cells were significantly lower in SjD patients compared to HCs ( p = 0.000 and p = 0.007, respectively). Patients with medium to high disease activity, defined by an EULAR Sjögren Syndrome Disease Activity Index (ESSDAI) score ≥ 5, exhibited lower DN B cell levels compared to those with low disease activity (ESSDAI score < 5, p = 0.029 and p = 0.000, respectively). Moreover, lower DN B cell percentages were associated with glandular involvement ( p = 0.039) and neutropenia ( p = 0.034). Regression analysis revealed a negative correlation between DN B cells and SjD disease activity (B=-0.240, 95% CI:-0.468 to -0.013, p = 0.039). Additionally, DN B cells showed weak negative correlation with the percentage of total B cells ( r =-0.266, p = 0.026). This study demonstrates that DN B cells are decreased in SjD, particularly in patients with glandular involvement and high disease activity. Moreover, the inverse correlation between DN B cells and total B cells suggested that DN B cells may play a role in the pathogenesis of SjD.

China is the largest soybean-consuming country in the world, but its self-sufficiency rate (SSR) of 16% is very low and it therefore has to heavily rely on imports. To solve the soybean dilemma in China, it is necessary to examine the maximum amount of soybean that could be grown on the land currently used, how much land could reasonably be used to expand soybean acreage, and whether China could sustainably increase soybean self-sufficiency to reduce the risks of import reliance. To answer these questions, our paper presents a high-resolution spatial analysis of potential soybean production in China using primary data of weather and crop production practices that govern this potential. We employed a “bottom-up” scaling protocol to estimate gaps between potential yield with optimal management and current yields in three major soybean-planting regions, namely, Northeast China, Central China, and South China. We found that current soybean yield gap (Yg) in China is 49% and 45% of potential yield under irrigated and rainfed cropping systems, respectively. By closing the yield gap, Northeast China could provide additional soybean production equivalent to 32% of the current national total. Our results show that SSR could only be increased to 21–23% in 2030 by Yg closure alone but could be increased to a maximum of 52% by combining Yg closure and a reasonable area expansion. Even so, at least 61.08 million tons of soybean accounting for 38% of global soybean trade would still need to be imported to meet future domestic demand. We discuss strategies for soybean production increase based on Yg closure in the most valuable areas and cropland expansion in a sustainable manner in order to increase SSR as well as lessen the import pressure on the global market.

BackgroundGastric cancer (GC) is a highly heterogeneous tumor with different responses to immunotherapy. Identifying immune subtypes and landscape of GC could improve immunotherapeutic strategies.MethodsBased on the abundance of tumor-infiltrating immune cells in GC patients from The Cancer Genome Atlas, we used unsupervised consensus clustering algorithm to identify robust clusters of patients, and assessed their reproducibility in an independent cohort from Gene Expression Omnibus. We further confirmed the feasibility of our immune subtypes in five independent pan-cancer cohorts. Finally, functional enrichment analyses were provided, and a deep learning model studying the pathological images was constructed to identify the immune subtypes.ResultsWe identified and validated three reproducible immune subtypes presented with diverse components of tumor-infiltrating immune cells, molecular features, and clinical characteristics. An immune-inflamed subtype 3, with better prognosis and the highest immune score, had the highest abundance of CD8+ T cells, CD4+ T–activated cells, follicular helper T cells, M1 macrophages, and NK cells among three subtypes. By contrast, an immune-excluded subtype 1, with the worst prognosis and the highest stromal score, demonstrated the highest infiltration of CD4+ T resting cells, regulatory T cells, B cells, and dendritic cells, while an immune-desert subtype 2, with an intermediate prognosis and the lowest immune score, demonstrated the highest infiltration of M2 macrophages and mast cells, and the lowest infiltration of M1 macrophages. Besides, higher proportion of EVB and MSI of TCGA molecular subtyping, over expression of CTLA4, PD1, PDL1, and TP53, and low expression of JAK1 were observed in immune subtype 3, which consisted with the results from Gene Set Enrichment Analysis. These subtypes may suggest different immunotherapy strategies. Finally, deep learning can predict the immune subtypes well.ConclusionThis study offers a conceptual frame to better understand the tumor immune microenvironment of GC. Future work is required to estimate its reference value for the design of immune-related studies and immunotherapy selection.

Adzuki bean is an important legume crop due to its high-quality protein, fiber, vitamins, minerals as well as rich bioactive substances. However, it is vulnerable to drought at the germination stage. However, little information is available about the genetic control of drought tolerance during seed germination in adzuki bean. In this study, some differential expression proteins (DEPs) were identified during seed germination between the drought-tolerant variety 17235 and drought-sensitive variety 17033 in adzuki bean using iTRAQ method. A total of 2834 proteins were identified in the germinating seeds of these two adzuki beans. Compared with the variety 17033, 87 and 80 DEPs were increased and decreased accumulation in variety 17235 under drought, respectively. Meanwhile, in the control group, a few DEPs, including 9 up-regulated and 21 down-regulated proteins, were detected in variety 17235, respectively. GO, KEGG, and PPI analysis revealed that the DEPs related to carbohydrate metabolism and energy production were significantly increased in response to drought stresses. To validate the proteomic function, the ectopic overexpression of V-ATPase in tobacco was performed and the result showed that V-ATPase upregulation could enhance the drought tolerance of tobacco. The results provide valuable insights into genetic response to drought stress in adzuki bean, and the DEPs could be applied to develop biomarkers related to drought tolerant in adzuki bean breeding projects.

Although numerous studies have described the application of artificial intelligence (AI) in diabetic retinopathy (DR) screening among diabetic populations, studies among populations in rural areas are rare. The purpose of this study was to evaluate the application value of an AI-based diagnostic system for DR screening in rural areas of midwest China. In this diagnostic accuracy study, diabetes mellitus (DM) patients in the National Basic Public Health Information Systems of Licheng County and Lucheng County of Changzhi city from July to December 2020 were selected as the target population. A total of 7824 eyes of 3933 DM patients were enrolled in this screening; the patients included 1395 males and 2401 females, with an average age of 19-87 years (63±8.735 years). All fundus photographs were collected by a professional ophthalmologist under natural pupil conditions in a darkroom using the Zhiyuan Huitu fundus image AI analysis software EyeWisdom. The AI-based diagnostic system and ophthalmologists were tasked with diagnosing the photos independently, and the consistency rate, sensitivity and specificity of the two methods in diagnosing DR were calculated and compared. The prevalence rates of DR according to the ophthalmologist and AI diagnoses were 22.7% and 22.5%, respectively; the consistency rate was 81.6%. The sensitivity and specificity of the AI system relative to the ophthalmologists' grades were 81.2% (95% confidence interval [CI]: 80.3% 82.1%) and 94.3% (95% CI: 93.7% 94.8%), respectively. There was no significant difference in diagnostic outcomes between the methods (X2 = 0.329, P = 0.566, P>0.05), and the AI-based diagnostic system had high consistency with the ophthalmologists' diagnostic results ([kappa] = 0.752). Our research demonstrated that DR patients in rural area hospitals can be screened feasibly. Compared with that of the ophthalmologists, however, the accuracy of the AI system must be improved. The results of this study might lend support to the large-scale application of AI in DR screening among different populations.

Purpose To evaluate the relationship between vertebral bone marrow fat, blood flow perfusion, and IVD biochemical components in postmenopausal women using quantitative MRI. Materials and methods Postmenopausal women were categorized into three groups based on QCT: normal BMD ( n =  34), osteopenia ( n =  27), and osteoporosis ( n =  24). Lumbar vertebral bone marrow fat fraction (FF) was quantified using a 6-echo-based Dixon sequence. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) provided quantitative parameters for bone marrow blood perfusion ( K trans , K ep , V e ). T2 mapping imaging assessed lumbar disc degeneration, yielding T2 values for the nucleus pulposus (NP), anterior annulus fibrosus (AAF), and posterior annulus fibrosus (PAF). We analyzed Spearman correlation coefficients and identified independent factors influencing disc T2 values through multiple linear regression analysis. Results Significant differences were found in FF, K trans , K ep , and V e values across groups ( p <  0.001 , p <  0.001 , p <  0.001 , p =  0.025). FF showed a strong inverse correlation with adjacent NP T2 values at L1 ( r =  -0.760, adjusted p <  0.001), and moderate at L2 and L3 ( r =  -0.613, adjusted p <  0.001; r =  -0.455, adjusted p <  0.001). Positive correlations were found between K trans and adjacent NP T2 values at all levels ( r =  0.715 to 0.314, adjusted p <  0.01 to < 0.001). A multivariate linear regression analysis revealed that FF and K trans values were independent predictors of adjacent NP T2 values at L1 to L4 levels. Conclusion Both vertebral bone marrow FF and K trans are independent predictors of intervertebral disc health, as reflected by nucleus pulposus T2 values. Notably, these relationships exhibit a distinct craniocaudal gradient, with stronger vertebral-disc coupling in the upper lumbar spine. Multiparametric quantitative MRI provides deeper insights into early disc degeneration pathways beyond single-parameter assessments.

Idiopathic inflammatory myopathies (IIMs) are systemic autoimmune disorders with unknown etiology. Despite the established link between gut microbes and immunity, the roles of gut bacteriome, mycobiome, and virome in IIM are unexplored. We performed shotgun metagenomic sequencing on fecal samples from 34 IIM patients and 37 healthy controls to profile gut microbiota. Taxonomic, functional, network, and machine-learning analyses revealed microbial dysbiosis and its potential for discriminating IIM. All three microbial kingdoms were significantly altered in IIM. Several inflammation-associated bacterial taxa (e.g., Rothia mucilaginosa , Streptococcus parasanguinis , Trueperella pyogenes ) and opportunistic fungi (e.g., Aspergillus spp.) were enriched in IIM, while SCFA-producing bacteria and fungi were depleted. Virome analysis revealed substantial shifts, with higher abundance of Siphoviridae in IIM. Altered viral functional gene profiles suggesting enhanced phage-mediated genome integration, recombination, and bacterial stress adaptation. Multi-kingdom network analysis showed extensive rewiring in IIM, characterized by increased network connectivity and a shift toward fungi-centered ecological hubs, contrasting with bacteria/virus-dominated networks in controls. In machine-learning models, the virome demonstrated the strongest discriminatory power, and viral signatures dominated the combined multi-kingdom classifier (AUC = 0.997). This first comprehensive multi-kingdom gut microbiota analysis in IIM provides a foundation for developing diagnostic and therapeutic strategies.