A deep neural network prediction method for diabetes based on Kendall's correlation coefficient and attention mechanism.

Diabetes is a chronic disease, which is characterized by abnormally high blood sugar levels. It may affect various organs and tissues, and even lead to life-threatening complications. Accurate prediction of diabetes can significantly reduce its incidence. However, the current prediction methods struggle to accurately capture the essential characteristics of nonlinear data, and the black-box nature of these methods hampers its clinical application. To address these challenges, we propose KCCAM_DNN, a diabetes prediction method that integrates Kendall's correlation coefficient and an attention mechanism within a deep neural network. In the KCCAM_DNN, Kendall's correlation coefficient is initially employed for feature selection, which effectively filters out key features influencing diabetes prediction. For missing values in the data, polynomial regression is utilized for imputation, ensuring data completeness. Subsequently, we construct a deep neural network (KCCAM_DNN) based on the self-attention mechanism, which assigns greater weight to crucial features affecting diabetes and enhances the model's predictive performance. Finally, we employ the SHAP model to analyze the impact of each feature on diabetes prediction, augmenting the model's interpretability. Experimental results show that KCCAM_DNN exhibits superior performance on both PIMA Indian and LMCH diabetes datasets, achieving test accuracies of 99.090% and 99.333%, respectively, approximately 2% higher than the best existing method. These results suggest that KCCAM_DNN is proficient in diabetes prediction, providing a foundation for informed decision-making in the diagnosis and prevention of diabetes.

Integrative analysis of microbiome and metabolome revealed the effect of microbial inoculant on microbial community diversity and function in rhizospheric soil under tobacco monoculture.

Over-application of chemical fertilizers and continuous cropping obstacles seriously restrict the sustainable development of tobacco production. Localized fertilization of beneficial microbes has potential advantages in achieving higher productivity, but the underlying biological mechanisms of interactions between rhizospheric microorganisms and the related metabolic cycle remain poorly characterized. Here, an integrative analysis of microbiomes with non-targeted metabolomics was performed on 30 soil samples of rhizosphere, root surrounding, and bulk soils from flue-cured tobacco under continuous and non-continuous monocropping systems. The analysis was conducted using UPLC-MS/MS platforms and high-throughput amplicon sequencing targeting the bacterial 16S rRNA gene and fungal ITS gene. The microbial inoculant consisted of Bacillus subtilis, B. velezensis, and B. licheniformis at the ratio of 1:1:1 in effective microbial counts, improved the cured leaf yield and disease resistance of tobacco, and enhanced nicotine and nitrogen contents of tobacco leaves. The bacterial taxa Rhizobium, Pseudomonas, Sphingomonadaceae, and Burkholderiaceae of the phylum Proteobacteria accumulated in high relative abundance and were identified as biomarkers following the application of the microbial inoculant. Under continuous monocropping, metabolomics demonstrated that the application of the microbial inoculant significantly affected the soil metabolite spectrum, and the differential metabolites were significantly enriched to the synthesis and degradation of nicotine (nicotinate and nicotinamide metabolism and biosynthesis of alkaloids derived from nicotinic acid). In addition, microbes were closely related to the accumulation of metabolites through correlation analysis. The interactions between plant roots and rhizospheric microorganisms provide valuable information for understanding how these beneficial microbes affect complex biological processes and the adaption capacity of plants to environments.IMPORTANCEThis study elaborated on how the microbial fertilizer significantly changed overall community structures and metabolite spectrum of rhizospheric microbes, which provide insights into the process of rhizosphere microbial remolding in response to continuous monocropping. we verified the hypothesis that the application of the microbial inoculant in continuous cropping would lead to the selection of distinct microbiota communities by establishing models to correlate biomarkers. Through correlation analysis of the microbiome and metabolome, we proved that rhizospheric microbes were closely related to the accumulation of metabolites, including the synthesis and degradation of nicotine. The interactions between plant roots and rhizospheric microorganisms provide valuable information for understanding how these beneficial microbes affect complex biological processes and the adaption capacity of plants to environments.

Ultrasound image segmentation of renal tumors based on UNet++ with fusion of multiscale residuals and dual attention.

Objective. Laparoscopic renal unit-preserving resection is a routine and effective means of treating renal tumors. Image segmentation is an essential part before tumor resection. The current segmentation method mainly relies on doctors manual delineation, which is time-consuming, labor-intensive, and influenced by their personal experience and ability. And the image quality of segmentation is low, with problems such as blurred edges, unclear size and shape, which are not conducive to clinical diagnosis.Approach. To address these problems, we propose an automated segmentation method, i.e. the UNet++ algorithm fusing multiscale residuals and dual attention (MRDA_UNet++). It replaces two consecutive 3 × 3 convolutions in UNet++ with the 'MultiRes block' module, which incorporates coordinate attention to fuse features from different scales and suppress the impact of background noise. Furthermore, an attention gate is also added at the short connections to enhance the ability of the network to extract features from the target area.Main results. The experimental results show that MRDA_UNet++ achieves 93.18%, 92.87%, 93.66%, and 92.09% on the real-world dataset for MIoU, Dice, Precision, and Recall, respectively. Compared to the baseline model UNet++ on three public datasets, the MIoU, Dice, and Recall metrics improved by 6.00%, 7.90% and 18.09% respectively for BUSI, 0.39%, 0.27% and 1.03% for Dataset C, and 1.37%, 1.75% and 1.30% for DDTI.Significance. The proposed MRDA_UNet++ exhibits obvious advantages in feature extraction, which can not only significantly reduce the workload of doctors, but also further decrease the risk of misdiagnosis. It is of great value to assist doctors diagnosis in the clinic.

A Convolutional Neural Network Face Recognition Method Based on BiLSTM and Attention Mechanism.

Face recognition technology is a powerful means to capture biological facial features and match facial data in existing databases. With the advantages of noncontact and long-distance implementation, it is being used in more and more scenarios. Affected by factors such as light, posture, and background environment, the face images captured by the device are still insufficient in the recognition rate of existing face recognition models. We propose an AB-FR model, a convolutional neural network face recognition method based on BiLSTM and attention mechanism. By adding an attention mechanism to the CNN model structure, the information from different channels is integrated to enhance the robustness of the network, thereby enhancing the extraction of facial features. Then, the BiLSTM method is used to extract the timing characteristics of different angles or different time photos of the same person so that convolutional blocks can obtain more face detail information. Finally, we used the cross-entropy loss function to optimize the model and realize the correct face recognition. The experimental results show that the improved network model indicates better identification performance and stronger robustness on some public datasets (such as CASIA-FaceV5, LFW, MTFL, CNBC, and ORL). Besides, the accuracy rate is 99.35%, 96.46%, 97.04%, 97.19%, and 96.79%, respectively.

Symptoms and pathogens diversity of Corn Fusarium sheath rot in Sichuan Province, China.

To elucidate the symptoms and pathogens diversity of corn Fusarium sheath rot (CFSR), diseased samples were collected from 21 county-level regions in 12 prefecture-level districts of Sichuan Province from 2015 to 2018 in the present study. In the field, two symptom types appeared including small black spots with a linear distribution and wet blotches with a tawny or brown color. One hundred thirty-seven Fusarium isolates were identified based on morphological characteristics and phylogenetic analysis (EF1-α), and Koch's postulates were also assessed. The results identified the isolates as 8 species in the Fusarium genus, including F. verticillioides, F. proliferatum, F. fujikuroi, F. asiaticum, F. equiseti, F. meridionale, F. graminearum and F. oxysporum, with isolation frequencies of 30.00, 22.67, 15.33, 7.33, 6.00, 5.33, 3.33 and 1.33%, respectively. Fusarium verticillioides and F. proliferatum were the dominant and subdominant species, respectively. Two or more Fusarium species such as F. verticillioides and F. proliferatum were simultaneously identified at a mixed infection rate of 14.67% in the present study. The pathogenicity test results showed that F. proliferatum and F. fujikuroi exhibited the highest virulence, with average disease indices of 30.28 ± 2.87 and 28.06 ± 1.96, followed by F. equiseti and F. verticillioides, with disease indices of 21.48 ± 2.14 and 16.21 ± 1.84, respectively. Fusarium asiaticum, F. graminearum and F. meridonale showed lower virulence, with disease indices of 13.80 ± 2.07, 11.57 ± 2.40 and 13.89 ± 2.49, respectively. Finally, F. orysporum presented the lowest virulence in CFSR, with a disease index of 10.14 ± 1.20. To the best of our knowledge, this is the first report of F. fujikuroi, F. meridionale and F. asiaticum as CFSR pathogens in China.

Etiology and Symptoms of Maize Leaf Spot Caused by Bipolaris spp. in Sichuan, China.

Many species of the genus Bipolaris are important plant pathogens and often cause leaf spot, root rot, and seedling blight in an extremely wide range of hosts around the world. In recent years, maize leaf spot caused by Bipolaris species has frequently occurred with complex symptoms and is becoming increasingly serious in Sichuan Province of China. To investigate the population diversity of Bipolaris spp. and their corresponding symptoms in maize, 747 samples of maize leaf spot were collected from 132 sampling sites in 19 administrative districts of Sichuan Province from 2011 to 2018. Based on morphological characteristics, pathogenicity testing, and phylogenetic analysis of the rDNA internal transcribed spacer (ITS) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) genes, a total of 1186 Bipolaris isolates were identified as B. maydis, B. zeicola, B. cynodontis, B. oryzae, B. setariae, and B. saccharicola, among which B. maydis and B. zeicola were the dominant pathogenic species, accounting for 57.34% and 42.07% of the isolates, respectively. We found that B. zeicola isolates were mainly distributed in high altitude and cool mountainous areas, while B. maydis was more widely distributed in Sichuan Province. The typical symptoms caused by the Bipolaris species were clearly distinct in maize. The typical symptoms caused by B. maydis were elongated strip lesions, or fusiform, elliptical lesions, and those caused by B. zeicola were narrow linear lesions. Herein, B. saccharicola was first reported on maize and caused subrotund lesions. This study provides useful information for disease diagnosis and management for Bipolaris leaf spot in maize.

Terminating rice innate immunity induction requires a network of antagonistic and redox-responsive E3 ubiquitin ligases targeting a fungal sirtuin.

Fungal phytopathogens can suppress plant immune mechanisms in order to colonize living host cells. Identifying all the molecular components involved is critical for elaborating a detailed systems-level model of plant infection probing pathogen weaknesses; yet, the hierarchy of molecular events controlling fungal responses to the plant cell is not clear. Here we show how, in the blast fungus Magnaporthe oryzae, terminating rice innate immunity requires a dynamic network of redox-responsive E3 ubiquitin ligases targeting fungal sirtuin 2 (Sir2), an antioxidation regulator required for suppressing the host oxidative burst. Immunoblotting, immunopurification, mass spectrometry and gene functional analyses showed that Sir2 levels responded to oxidative stress via a mechanism involving ubiquitination and three antagonistic E3 ubiquitin ligases: Grr1 and Ptr1 maintained basal Sir2 levels in the absence of oxidative stress; Upl3 facilitated Sir2 accumulation in response to oxidative stress. Grr1 and Upl3 interacted directly with Sir2 in a manner that decreased and scaled with oxidative stress, respectively. Deleting UPL3 depleted Sir2 during growth in rice cells, triggering host immunity and preventing infection. Overexpressing SIR2 in the Δupl3 mutant remediated pathogenicity. Our work reveals how redox-responsive E3 ubiquitin ligases in M. oryzae mediate Sir2 accumulation-dependent antioxidation to modulate plant innate immunity and host susceptibility.

Latent Infection of Powdery Mildew on Volunteer Wheat in Sichuan Province, China.

Wheat powdery mildew, caused by the fungal pathogen Blumeria graminis f. sp. tritici, is one of the most destructive wheat diseases in China, especially in Sichuan Province. Successfully oversummered B. graminis f. sp. tritici can become a primary infection source for wheat seedlings in the fall. Determining the latent infection level of B. graminis f. sp. tritici in volunteer wheat and the oversummering areas of B. graminis f. sp. tritici is important for estimating potential B. graminis f. sp. tritici epidemics. In this study, we clarified the critical role of volunteer wheat in the B. graminis f. sp. tritici oversummering cycle and determined whether latent B. graminis f. sp. tritici infection was present in volunteer wheat by using real-time polymerase chain reaction (real-time PCR). The results indicated that volunteer wheat was mostly found in the northeast and middle regions of Sichuan, where lower temperatures and higher precipitation are common. A total of 13.2% of samples showed symptoms of B. graminis f. sp. tritici (spores) in the field, and 36.8% of samples were found to carry the B. graminis f. sp. tritici pathogen, even though no symptoms were observed. Volunteer wheat with B. graminis f. sp. tritici infection symptoms was found at an altitude of 536 m but volunteer wheat latently infected by B. graminis f. sp. tritici was identified at the lowest altitude of 323 m. Crop shade (e.g., corn and lima bean) provided suitable conditions for the survival of volunteer wheat in the summer. In addition, volunteer wheat played a key role in the B. graminis f. sp. tritici oversummering cycle. Moreover, B. graminis f. sp. tritici could oversummer by infecting generations of volunteer wheat in the summer, thereby becoming the primary infection source for autumn-sown wheat. The results showed that the latent infection of wheat diseases could be rapidly quantified by real-time PCR. Here, the primary disease center of autumn-sown wheat in Ya'an and Wenjiang were detected accurately based on this method. This study provides solid evidence for identifying the disease center, which offers guidance for wheat disease control and management.

Phenotypic and Genetic Characterization of Botrytis cinerea Population from Kiwifruit in Sichuan Province, China.

Botrytis cinerea (anamorph of Botryotinia fuckeliana) causes gray mold on numerous plants, including kiwifruit. The primary aim of this study was to investigate the phenotypic and genetic characteristics of the Botrytis cinerea population from kiwifruit in Sichuan Province, China. In all, 176 isolates were collected from kiwifruit orchards from eight geographic regions in Sichuan. All isolates were identified as B. cinerea sensu stricto based on the combined datasets, including morphological criteria, determination of the Bc-hch allele, and phylogenetic analysis of the genes RPB2, G3PDH, and HSP60. Three colony types (i.e., sclerotial, mycelial, and conidial) were observed on potato dextrose agar after 2 weeks, with sclerotial isolates, the predominant category, accounting for 40.91%. No obvious differences in microscopic characteristics were observed among the three types. Three genotypes of transposable elements were identified in the B. cinerea population: boty, flipper, and transposa types. The most prevalent genotype from different geographic populations of B. cinerea was transposa; in contrast, the flipper genotype accounted for only 3.98% of the total population, whereas the vacuma genotype was absent. According to MAT locus amplification, 87 and 89 isolates are MAT1-1 and MAT1-2 type, respectively, and the two mating types were found to be balanced overall in the population. Forty-eight representative isolates were all able to cause gray mold to some extent, and disease severities were significantly different between the cultivars Hongyang and Hort16A (P < 0.01). Disease severity was significantly greater on young leaves than on mature leaves (P < 0.01). No significant relationship was found between pathogenicity and geographical region, colony type, or transposon distribution. The results obtained in the present study suggest a relatively uniform species diversity of Botrytis but rich phenotypic and genetic differentiation within the B. cinerea population on kiwifruit in China. Utilizing resistant cultivars and rain-shelter cultivation instead of fungicides may be an effective approach to delaying pathogen variability.

A Feed-Forward Subnetwork Emerging from Integrated TOR- and cAMP/PKA-Signaling Architecture Reinforces Magnaporthe oryzae Appressorium Morphogenesis.

Appressoria are important mediators of plant-microbe interactions. In the devastating rice blast pathogen Magnaporthe oryzae, appressorial morphogenesis from germ tube tips requires activated cAMP/PKA signaling and inactivated TOR signaling (TORoff). TORoff temporarily arrests G2 at a metabolic checkpoint during the single round of mitosis that occurs following germination. G2 arrest induces autophagy and appressorium formation concomitantly, allowing reprogression of the cell cycle to G1/G0 quiescence and a single appressorial nucleus. Inappropriate TOR activation abrogates G2 arrest and inhibits cAMP/PKA signaling downstream of cPKA. This results in multiple rounds of germ tube mitosis and the loss of autophagy and appressoria formation. How cAMP/PKA signaling connects to cell cycle progression and autophagy is not known. To address this, we interrogated TOR and cAMP/PKA pathways using signaling mutants, different surface properties, and specific cell cycle inhibitors and discovered a feed-forward subnetwork arising from TOR- and cAMP/PKA-signaling integration. This adenylate cyclase-cAMP-TOR-adenylate cyclase subnetwork reinforces cAMP/PKA-dependent appressorium formation under favorable environmental conditions. Under unfavorable conditions, the subnetwork collapses, resulting in reversible cell cycle-mediated germ tube growth regardless of external nutrient status. Collectively, this work provides new molecular insights on germ tube morphogenetic decision-making in response to static and dynamic environmental conditions.

The optical lens coupled X-ray in-line phase contrast imaging system for the characterization of low Z materials.

Due to the high spatial resolution and contrast, the optical lens coupled X-ray in-line phase contrast imaging system with the secondary optical magnification is more suitable for the characterization of the low Z materials. The influence of the source to object distance and the object to scintillator distance on the image resolution and contrast is studied experimentally. A phase correlation algorithm is used for the image mosaic of a serial of X-ray phase contrast images acquired with high resolution, the resulting resolution is less than 1.0 µm, and the whole field of view is larger than 1.4 mm. Finally, the geometric morphology and the inner structure of various weakly absorbing samples and the evaporation of water in the plastic micro-shell are in situ characterized by the optical lens coupled X-ray in-line phase contrast imaging system.

Molecular and phenotypic characterization of Colletotrichum species associated with anthracnose disease in peppers from Sichuan Province, China.

The anthracnose caused by Colletotrichum species is an important disease that primarily causes fruit rot in pepper. Eighty-eight strains representing seven species of Colletotrichum were obtained from rotten pepper fruits in Sichuan Province, China, and characterized according to morphology and the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) sequence. Fifty-two strains were chosen for identification by phylogenetic analyses of multi-locus sequences, including the nuclear ribosomal internal transcribed spacer (ITS) region and the ß-tubulin (TUB2), actin (ACT), calmodulin (CAL) and GAPDH genes. Based on the combined datasets, the 88 strains were identified as Colletotrichum gloeosporioides, C. siamense, C. fructicola, C. truncatum, C. scovillei, and C. brevisporum, and one new species was detected, described as Colletotrichum sichuanensis. Notably, C. siamense and C. scovillei were recorded for the first time as the causes of anthracnose in peppers in China. In addition, with the exception of C. truncatum, this is the first report of all of the other Colletotrichum species studied in pepper from Sichuan. The fungal species were all non-host-specific, as the isolates were able to infect not only Capsicum spp. but also Pyrus pyrifolia in pathogenicity tests. These findings suggest that the fungal species associated with anthracnose in pepper may inoculate other hosts as initial inoculum.

Virulence Structure of Blumeria graminis f. sp. tritici and Its Genetic Diversity by ISSR and SRAP Profiling Analyses.

Blumeria graminis f. sp. tritici, which causes wheat powdery mildew, is an obligate biotrophic pathogen that can easily genetically adapt to its host plant. Understanding the virulence structure of and genetic variations in this pathogen is essential for disease control and for breeding resistance to wheat powdery mildew. This study investigated 17 pathogenic populations in Sichuan, China and classified 109 isolates into two distinct groups based on pathogenicity analysis: high virulence (HV, 92 isolates) and low virulence (LV, 17 isolates). Populations from Yibin (Southern region), Xichang (Western region), and Meishan (Middle region) showed lower virulence frequencies than populations from other regions. Many of the previously known resistance genes did not confer resistance in this study. The resistance gene Pm21 displayed an immune response to pathogenic challenge with all populations in Sichuan, and Pm13, Pm5b, Pm2+6, and PmXBD maintained resistance. AMOVA revealed significantly higher levels of variation within populations and lower levels of variation among populations within regions. High levels of gene flow were detected among populations in the four regions. Closely related populations within each region were distinguished by cluster analyses using ISSR and SRAP alleles. Both ISSR and SRAP allele profiling analyses revealed high levels of genetic diversity among pathogenic populations in Sichuan. Although ISSR and SRAP profiling analysis showed similar resolutions, the SRAP alleles appeared to be more informative. We did not detect any significant association between these alleles and the virulence or pathogenicity of the pathogen. Our results suggest that ISSR and SRAP alleles are more efficient for the characterization of small or closely related populations versus distantly related populations.

Treatment of pertrochanteric fractures with a proximal femur locking compression plate.

BACKGROUND: Pertrochanteric femoral fractures are one of the most common fractures in old patients. However, fixing pertrochanteric fractures properly is clinically challenging. There are also no routine treatments for this fracture. Here, we report the clinical trial of pertrochanteric fracture treatment with a proximal femur locking compression plate (PFLCP). By recording and analysing the radiographic and clinical results from patients treated with PFLCP, we found that PFLCP could provide three-dimensional fixation mechanical advantages compared with conventional treatments, even in the case of unstable fractures in the osteoporotic bone. METHODS: The report included a total of 110 patients (72 females and 38 males) with pertrochanteric femoral fractures who were subjected to PFLCP treatment. The mean age of the patients was 75 (48-93) years. Pertrochanteric fracture includes both intertrochanteric and subtrochanteric femoral fractures: intertrochanteric fractures were classified according to Jensen (1980), whereas subtrochanteric fractures were classified according to Zickel (1980). Detailed clinical conditions of all patients, including blood loss, drainage and length of incision, were recorded individually. The duration of image intensification was also monitored. Patients were revisited at 6 weeks, 3 months, 6 months and 1 year after the operation. The progress of healing, as well as the occurrence of complications, was recorded. RESULTS: Amongst the 110 patients, 108 (98%) were available for follow-up check-up at 6 weeks, 104 (95%) at 3 months, 100 (91%) at 6 months and 94 (85%) at 1 year. The other patients were lost to follow-up because of death. The union rate was 95% (99/104), 98% (98/100) and 100% (94/94) at the 3-month, 6-month and 1-year period during the follow-up check-up, respectively. The patients healed satisfactorily and had no complications, such as cut-out in most cases. However, there was one case of breakage of the implant and one case of non-union at the 3-month period during the follow-up check-up. Amongst all patients, 77 cases were successfully reduced with traction on a fracture table under fluoroscopy; the others were opened to correct the displacement. The average operation time was 35.5 min, and the mean bleeding amount was 150 ml (including operative blood loss and wound drainage). The mean image intensifier time was 5 min and the mean length of incision was 9 cm. CONCLUSION: The PFLCP can be a feasible alternative to the treatment of pertrochanteric fractures. Treatment with a PFLCP can provide good-to-excellent healing for pertrochanteric fractures, with a limited occurrence of complications.

[Minimally invasive percutaneous locking compression plate internal fixation in the treatment of tibial fractures].

OBJECTIVE: To summarize the clinical application of minimally invasive percutaneous locking compression plate (LCP) internal fixation in the treatment of tibial fractures and to evaluate its clinical effects. METHODS: From September 2005 to September 2007, 13 patients with tibial fractures were treated with indirect reduction and minimally invasive percutaneous LCP internal fixation, 8 males and 5 females, aged 18-35 years old (27 on average). Among them, the fractures were caused by traffic accidents in 3 cases, by falling in 5 cases, by falling from height in 4 cases and by bruise in 1 case. The fractures were located at 1/3 upper tibia in 2 cases, at 1/3 medium tibia in 6 cases and at 1/3 lower tibia in 5 cases. All fracture were closed ones. According to the AO classification, 4 cases were type A, 7 type B and 2 type C. The time between fractures and operation was from 3 hours to 5 days (2.5 days on average). RESULTS: All incisions obtained healing by first intention. All patients were followed up for 10-18 months (13 months on average). All fractures reached clinical healing, and the healing time was 12-20 weeks (16 weeks on average). There was no delayed fracture healing, nonunion, infection and internal fixation failure. No complications such as rotation, crispatura deformity and internal fixation loosening were found. According to the HSS scoring, the function of the knee joint was graded 85-95 (90 on average), and the range of motion was 100-130 degrees (120 degrees on average). According to the AOFAS Ankle Hindfoot Scoring, the function of the ankle joint was graded 80-95 (92.4 on average). Nine cases were excellent, 4 good, and the choiceness rate was 100%. CONCLUSION: Minimally invasive percutaneous LCP internal fixation is in accord with biological set principles and beneficial for tibial fracture healing and reconstruction of soft tissues.