Search

Image data augmentation plays a crucial role in data augmentation (DA) by increasing the quantity and diversity of labeled training data. However, existing methods have limitations. Notably, techniques like image manipulation, erasing, and mixing can distort images, compromising data quality. Accurate representation of objects without confusion is a challenge in methods like auto augment and feature augmentation. Preserving fine details and spatial relationships also proves difficult in certain techniques, as seen in deep generative models. To address these limitations, we propose OFIDA, an object-focused image data augmentation algorithm. OFIDA implements one-to-many enhancements that not only preserve essential target regions but also elevate the authenticity of simulating real-world settings and data distributions. Specifically, OFIDA utilizes a graph-based structure and object detection to streamline augmentation. Specifically, by leveraging graph properties like connectivity and hierarchy, it captures object essence and context for improved comprehension in real-world scenarios. Then, we introduce DynamicFocusNet, a novel object detection algorithm built on the graph framework. DynamicFocusNet merges dynamic graph convolutions and attention mechanisms to flexibly adjust receptive fields. Finally, the detected target images are extracted to facilitate one-to-many data augmentation. Experimental results validate the superiority of our OFIDA method over state-of-the-art methods across six benchmark datasets.

[This corrects the article DOI: 10.1371/journal.pone.0050093.].

Based on the database data of Chinese industrial enterprises from 2000 to 2007 and the LP method, this paper measures the total factor productivity of enterprises and investigates the effect of different mixed ownership forms on enterprises' efficiency and the effect of heterogeneous ownership balance on the mixed ownership enterprises' efficiency. The state-owned enterprise and mixed ownership enterprise are identified by the enterprise's paid-up capital. The results show that, on the whole, for the mixed ownership enterprise, the higher the diversification degree of the shareholders is, the higher the efficiency becomes, and in different types of industries, the mixed forms of shareholders have different effects on the efficiency of enterprises. The heterogeneous ownership balance and the enterprise efficiency show nonlinear U-type relationships. Both the higher and lower heterogeneous ownership balance degrees will promote the enterprise's efficiency. However, when the ownership balance degree is in the range of [0.2 0.5], the increase in ownership balance will lead to the decline of enterprise efficiency. Therefore, when introducing non-state-owned capital, state-owned enterprises should take full account of their own characteristics by rationally controlling the shareholding ratio of non-state-owned capital and play the positive role of a mixed ownership structure in corporate governance with appropriate ownership balances.

The Eastern Mediterranean Region (EMR) is witnessing an increase in chronic disorders, including mental illness. With ongoing unrest, this is expected to rise. This is the first study to quantify the burden of mental disorders in the EMR. We used data from the Global Burden of Disease study (GBD) 2013. DALYs (disability-adjusted life years) allow assessment of both premature mortality (years of life lost-YLLs) and nonfatal outcomes (years lived with disability-YLDs). DALYs are computed by adding YLLs and YLDs for each age-sex-country group. In 2013, mental disorders contributed to 5.6% of the total disease burden in the EMR (1894 DALYS/100,000 population): 2519 DALYS/100,000 (2590/100,000 males, 2426/100,000 females) in high-income countries, 1884 DALYS/100,000 (1618/100,000 males, 2157/100,000 females) in middle-income countries, 1607 DALYS/100,000 (1500/100,000 males, 1717/100,000 females) in low-income countries. Females had a greater proportion of burden due to mental disorders than did males of equivalent ages, except for those under 15 years of age. The highest proportion of DALYs occurred in the 25-49 age group, with a peak in the 35-39 years age group (5344 DALYs/100,000). The burden of mental disorders in EMR increased from 1726 DALYs/100,000 in 1990 to 1912 DALYs/100,000 in 2013 (10.8% increase). Within the mental disorders group in EMR, depressive disorders accounted for most DALYs, followed by anxiety disorders. Among EMR countries, Palestine had the largest burden of mental disorders. Nearly all EMR countries had a higher mental disorder burden compared to the global level. Our findings call for EMR ministries of health to increase provision of mental health services and to address the stigma of mental illness. Moreover, our results showing the accelerating burden of mental health are alarming as the region is seeing an increased level of instability. Indeed, mental health problems, if not properly addressed, will lead to an increased burden of diseases in the region.

Mortality and morbidity are increased in patients with muscle atrophy resulting from catabolic diseases such as diabetes. At present there is no pharmacological treatment that successfully reverses muscle wasting from catabolic conditions. We hypothesized that acupuncture plus low frequency electric stimulation (Acu-LFES) would mimic the impact of exercise and prevent diabetes-induced muscle loss. Streptozotocin (STZ) was used to induce diabetes in mice. The mice were then treated with Acu-LFES for 15 minutes daily for 14 days. Acupuncture points were selected according to the WHO Standard Acupuncture Nomenclature guide. The needles were connected to an SDZ-II electronic acupuncture device delivering pulses at 20Hz and 1mA. Acu-LFES prevented soleus and EDL muscle weight loss and increased hind-limb muscle grip function in diabetic mice. Muscle regeneration capacity was significantly increased by Acu-LFES. The expression of Pax7, MyoD, myogenin and embryo myosin heavy chain (eMyHC) was significantly decreased in diabetic muscle vs. control muscle. The suppressed levels in diabetic muscle were reversed by Acu-LFES. The IGF-1 signaling pathway was also upregulated by Acu-LFES. Phosphorylation of Akt, mTOR and p70S6K were downregulated by diabetes leading to a decline in muscle mass, however, Acu-LFES countered the diabetes-induced decline. In addition, microRNA-1 and -206 were increased by Acu-LFES after 24 days of treatment. We conclude that Acu-LFES is effective in counteracting diabetes-induced skeletal muscle atrophy by increasing IGF-1 and its stimulation of muscle regeneration.

BackgroundDastarcus helophoroides is known as the most valuable natural enemy insect against many large-body longhorned beetles. The molecular mechanism of its long lifespan and reproduction makes it a unique resource for genomic research. However, molecular biological studies on this parasitic beetle are scarce, and genomic information for D. helophoroides is not currently available. Thus, transcriptome information for this species is an important resource that is required for a better understanding of the molecular mechanisms of D. helophoroides. In this study, we obtained transcriptome information of D. helophoroides using high-throughput RNA sequencing.ResultsUsing Illumina HiSeq 2000 sequencing, 27,543,746 clean reads corresponding to a total of 2.48 Gb nucleotides were obtained from a single run. These reads were assembled into 42,810 unigenes with a mean length of 683 bp. Using a sequence similarity search against the five public databases (NR, Swiss-Prot, GO, COG, KEGG) with a cut-off E-value of 10(-5) using Blastx, a total of 31,293 unigenes were annotated with gene description, gene ontology terms, or metabolic pathways.ConclusionsTo the best of our knowledge, this is the first study on the transcriptome information of D. helophoroides. The transcriptome data presented in this study provide comprehensive information for future studies in D. helophoroides, particularly for functional genomic studies in this parasitic beetle.

Variable number of tandem repeats (VNTRs) that are widely distributed in the genome of Yersinia pestis proved to be useful markers for the genotyping and source-tracing of this notorious pathogen. In this study, we probed into the features of VNTRs in the Y. pestis genome and developed a simple hierarchical genotyping system based on optimized VNTR loci.Capillary electrophoresis was used in this study for multi-locus VNTR analysis (MLVA) in 956 Y. pestis strains. The general features and genetic diversities of 88 VNTR loci in Y. pestis were analyzed with BioNumerics, and a "14+12" loci-based hierarchical genotyping system, which is compatible with single nucleotide polymorphism-based phylogenic analysis, was established.Appropriate selection of target loci reduces the impact of homoplasies caused by the rapid mutation rates of VNTR loci. The optimized "14+12" loci are highly discriminative in genotyping and source-tracing Y. pestis for molecular epidemiological or microbial forensic investigations with less time and lower cost. An MLVA genotyping datasets of representative strains will improve future research on the source-tracing and microevolution of Y. pestis.

A motif in a network is a connected graph that occurs significantly more frequently as an induced subgraph than would be expected in a similar randomized network. By virtue of being atypical, it is thought that motifs might play a more important role than arbitrary subgraphs. Recently, a flurry of advances in the study of network motifs has created demand for faster computational means for identifying motifs in increasingly larger networks. Motif detection is typically performed by enumerating subgraphs in an input network and in an ensemble of comparison networks; this poses a significant computational problem. Classifying the subgraphs encountered, for instance, is typically performed using a graph canonical labeling package, such as Nauty, and will typically be called billions of times. In this article, we describe an implementation of a network motif detection package, which we call NetMODE. NetMODE can only perform motif detection for [Formula: see text]-node subgraphs when [Formula: see text], but does so without the use of Nauty. To avoid using Nauty, NetMODE has an initial pretreatment phase, where [Formula: see text]-node graph data is stored in memory ([Formula: see text]). For [Formula: see text] we take a novel approach, which relates to the Reconstruction Conjecture for directed graphs. We find that NetMODE can perform up to around [Formula: see text] times faster than its predecessors when [Formula: see text] and up to around [Formula: see text] times faster when [Formula: see text] (the exact improvement varies considerably). NetMODE also (a) includes a method for generating comparison graphs uniformly at random, (b) can interface with external packages (e.g. R), and (c) can utilize multi-core architectures. NetMODE is available from netmode.sf.net.