More appropriate DenseNetBL classifier for small sample tree species classification using UAV-based RGB imagery.

To effectively classify tree species within datasets characterized by limited samples, we introduced a novel approach named DenseNetBL, founded upon the fusion of the DenseNet architecture and a pivotal bottleneck layer. This bottleneck layer, encompassing a compact convolutional component, played a central role in our methodology. The evaluation of DenseNetBL was conducted under varying conditions, encompassing small-sample tree species data, extensive remote sensing datasets, and state-of-the-art classifiers. Furthermore, a quantitative assessment was executed to extract tree species areas. This was achieved by quantifying pixel areas within manually delineated tree species maps and classifier-generated counterparts. The findings of our study indicated that, in scenarios devoid of pre-trained weights, DenseNetBL consistently outperformed its DenseNet counterpart with equivalent layer numbers. In the realm of small-sample situations, both the Swin Transformer and Vision Transformer exhibited inferior performance when juxtaposed with DenseNet and DenseNetBL. Remarkably, among the shallow architectures, DenseNet33BL showcased superior aptitude for small-sample tree species classification, culminating in the most commendable results (Overall Accuracy (OA) = 0.901, Kappa = 0.892). Conversely, the Vision Transformer yielded the least favorable classification outcomes (OA = 0.767, Kappa = 0.708). The amalgamation of DenseNet33BL and simple linear iterative clustering emerged as the optimal strategy for attaining robust tree species area extraction results across two prototypical forests. In contrast, DenseNet121 exhibited suboptimal performance in the same forests, attaining the least satisfactory tree species area extraction results. These comprehensive findings underscore the efficacy of our DenseNetBL approach in addressing the challenges associated with small-sample tree species classification and accurate tree species area extraction.

A general approach for automatic segmentation of pneumonia, pulmonary nodule, and tuberculosis in CT images.

Proposing a general segmentation approach for lung lesions, including pulmonary nodules, pneumonia, and tuberculosis, in CT images will improve efficiency in radiology. However, the performance of generative adversarial networks is hampered by the limited availability of annotated samples and the catastrophic forgetting of the discriminator, whereas the universality of traditional morphology-based methods is insufficient for segmenting diverse lung lesions. A cascaded dual-attention network with a context-aware pyramid feature extraction module was designed to address these challenges. A self-supervised rotation loss was designed to mitigate discriminator forgetting. The proposed model achieved Dice coefficients of 70.92, 73.55, and 68.52% on multi-center pneumonia, lung nodule, and tuberculosis test datasets, respectively. No significant decrease in accuracy was observed (p > 0.10) when a small training sample size was used. The cyclic training of the discriminator was reduced with self-supervised rotation loss (p < 0.01). The proposed approach is promising for segmenting multiple lung lesion types in CT images.

Machine Learning Customized Novel Material for Energy-Efficient 4D Printing.

Existing commercial powders for laser additive manufacturing (LAM) are designed for traditional manufacturing methods requiring post heat treatments (PHT). LAM's unique cyclic thermal history induces intrinsic heat treatment (IHT) on materials during deposition, which offers an opportunity to develop LAM-customized new materials. This work customized a novel Fe-Ni-Ti-Al maraging steel assisted by machine learning to leverage the IHT effect for in situ forming massive precipitates during LAM without PHT. Fast precipitation kinetics in steel, tailored intermittent deposition strategy, and the IHT effect facilitate the in situ Ni3 Ti precipitation in the martensitic matrix via heterogeneous nucleation on high-density dislocations. The as-built steel achieves a tensile strength of 1538 MPa and a uniform elongation of 8.1%, which is superior to a wide range of as-LAM-processed high-strength steel. In the current mainstream ex situ 4D printing, the time-dependent evolutions (i.e., property or functionality changes) of a 3D printed structure occur after part formation. This work highlights in situ 4D printing via the synchronous integration of time-dependent precipitation hardening with 3D geometry shaping, which shows high energy efficiency and sustainability. The findings provide insight into developing LAM-customized materials by understanding and utilizing the IHT-materials interaction.

Exportin 4 depletion leads to nuclear accumulation of a subset of circular RNAs.

Numerous RNAs are exported from the nucleus, abnormalities of which lead to cellular complications and diseases. How thousands of circular RNAs (circRNAs) are exported from the nucleus remains elusive. Here, we provide lines of evidence to demonstrate a link between the conserved Exportin 4 (XPO4) and nuclear export of a subset of circRNAs in metazoans. Exonic circRNAs (ecircRNAs) with higher expression levels, larger length, and lower GC content are more sensitive to XPO4 deficiency. Cellular insufficiency of XPO4 leads to nuclear circRNA accumulation, circRNA:DNA (ciR-loop) formation, linear RNA:DNA (liR-loop) buildup, and DNA damage. DDX39 known to modulate circRNA export can resolve ciR-loop, and splicing factors involved in the biogenesis of circRNAs can also affect the levels of ciR-loop. Testis and brain are two organs with high abundance of circRNAs, and insufficient XPO4 levels are detrimental, as Xpo4 heterozygous mice display male infertility and neural phenotypes. Increased levels of ciR-loop, R-loop, and DNA damage along with decreased cell numbers are observed in testis and hippocampus of Xpo4 heterozygotes. This study sheds light on the understandings of mechanism of circRNA export and reveals the significance of efficient nuclear export of circRNAs in cellular physiology.

Development and validation of a deep learning model for survival prognosis of transcatheter arterial chemoembolization in patients with intermediate-stage hepatocellular carcinoma.

PURPOSE: We aimed to develop a deep learning-based approach to evaluate both time-to-progression (TTP) and overall survival (OS) prognosis of transcatheter arterial chemoembolization (TACE) in treatment-naïve patients with intermediate-stage hepatocellular carcinoma (HCC) and compare the approach's performance with those of radiomics and clinical models. METHODS: EfficientNetV2 was used to build a prognosis model for treatment-naïve patients with HCC. Data of 414 intermediate-stage HCC patients from one participant center were collected to construct the training and validation datasets (70%:30%) for TTP prognosis, while data of 129 intermediate-stage HCC patients from another participant center were collected as the test dataset for both TTP and OS prognosis. Three radiomics and three clinical models were then constructed for comparison. RESULTS: Patients with EfficientNetV2-based model score ≤ 0.5 had better TTP than those with higher scores (hazard ratio [HR]: 0.32, 95%CI: 0.22-0.46, P < 0.0001; HR: 0.28, 95%CI: 0.20-0.41, P < 0.0001; and HR: 0.55, 95%CI: 0.36-0.88, P = 0.005 in the training, validation, and test datasets, respectively). Patients with model score ≤ 0.5 had better OS (38.8 months vs 20.9 months, HR: 0.58, 95%CI: 0.37-0.90, P = 0.008). Compared with the radiomics (intra-tumoral and peri-tumoral) and three clinical models, the EfficientNetV2-based model showed better survival prognosis for TACE (P < 0.05) in the test dataset. CONCLUSIONS: The EfficientNetV2-based model enables assessment of both TTP and OS prognosis of TACE in treatment-naïve, intermediate-stage HCC. Patients with lower scores will benefit from TACE. The model can potentially be used by clinicians to improve decision making regarding TACE treatment choices.

A deep learning-based system for survival benefit prediction of tyrosine kinase inhibitors and immune checkpoint inhibitors in stage IV non-small cell lung cancer patients: A multicenter, prognostic study.

Background: For clinical decision making, it is crucial to identify patients with stage IV non-small cell lung cancer (NSCLC) who may benefit from tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors (ICIs). In this study, a deep learning-based system was designed and validated using pre-therapy computed tomography (CT) images to predict the survival benefits of EGFR-TKIs and ICIs in stage IV NSCLC patients. Methods: This retrospective study collected data from 570 patients with stage IV EGFR-mutant NSCLC treated with EGFR-TKIs at five institutions between 2010 and 2021 (data of 314 patients were from a previously registered study), and 129 patients with stage IV NSCLC treated with ICIs at three institutions between 2017 and 2021 to build the ICI test dataset. Five-fold cross-validation was applied to divide the EGFR-TKI-treated patients from four institutions into training and internal validation datasets randomly in a ratio of 80%:20%, and the data from another institution was used as an external test dataset. An EfficientNetV2-based survival benefit prognosis (ESBP) system was developed with pre-therapy CT images as the input and the probability score as the output to identify which patients would receive additional survival benefit longer than the median PFS. Its prognostic performance was validated on the ICI test dataset. For diagnosing which patient would receive additional survival benefit, the accuracy of ESBP was compared with the estimations of three radiologists and three oncologists with varying degrees of expertise (two, five, and ten years). Improvements in the clinicians' diagnostic accuracy with ESBP assistance were then quantified. Findings: ESBP achieved positive predictive values of 80·40%, 75·40%, and 77·43% for additional EGFR-TKI survival benefit prediction using the probability score of 0·2 as the threshold on the training, internal validation, and external test datasets, respectively. The higher ESBP score (>0·2) indicated a better prognosis for progression-free survival (hazard ratio: 0·36, 95% CI: 0·19-0·68, p<0·0001) in patients on the external test dataset. Patients with scores >0·2 in the ICI test dataset also showed better survival benefit (hazard ratio: 0·33, 95% CI: 0·18-0·55, p<0·0001). This suggests the potential of ESBP to identify the two subgroups of benefiting patients by decoding the commonalities from pre-therapy CT images (stage IV EGFR-mutant NSCLC patients receiving additional survival benefit from EGFR-TKIs and stage IV NSCLC patients receiving additional survival benefit from ICIs). ESBP assistance improved the diagnostic accuracy of the clinicians with two years of experience from 47·91% to 66·32%, and the clinicians with five years of experience from 53·12% to 61·41%. Interpretation: This study developed and externally validated a preoperative CT image-based deep learning model to predict the survival benefits of EGFR-TKI and ICI therapies in stage IV NSCLC patients, which will facilitate optimized and individualized treatment strategies. Funding: This study received funding from the National Natural Science Foundation of China (82001904, 81930053, and 62027901), and Key-Area Research and Development Program of Guangdong Province (2021B0101420005).

Z-Score Reference Ranges for Umbilical Vein Diameter and Blood Flow Volume in Normal Fetuses.

OBJECTIVES: To establish Z-scores for the diameter and blood flow volume of the umbilical vein (UV) in normal fetuses. METHODS: This was a prospective study involving 907 normal fetuses. We measured the diameter (Duv) of two different segments of the UV (FUV: the free loop of the UV; FIUV: the fetal intra-abdominal UV). Next, we calculated the blood flow volume (Quv). Z-scores were created for both Duv and Quv using gestational age, femur length, and biparietal diameter as independent variables. RESULTS: We successfully acquired 858 (94.6%) normal fetal measurements. Between 20 and 39 weeks, the Duv of the FUV and FIUV increased from 0.38 to 0.80 cm and from 0.33 to 0.70 cm, respectively. The Quv of the FUV and FIUV increased from 32.66 to 381.88 ml/min and from 31.50 to 360.15 ml/min, respectively. Linear or quadratic regression models were best fitted between the parameters of UV and the independent variables. Z-scores were successfully determined for both the Duv and Quv. CONCLUSIONS: The calculation of Z-scores for the Duv and Quv is simple by applying standard statistical methods. These Z-scores may be useful to evaluate placental circulation and provide a rationale for monitoring and evaluating the prognosis of fetuses.

Large-Scale Genome Scanning within Exonic Regions Revealed the Contributions of Selective Sweep Prone Genes to Host Divergence and Adaptation in Magnaporthe oryzae Species Complex.

Magnaporthe oryzae, one of the most notorious plant pathogens in the agronomic ecosystem, causes a destructive rice blast disease around the world. The blast fungus infects wide arrays of cultivated and non-cultivated plants within the Poaceae. Studies have shown that host speciation exerts selection pressure that drives the evolution and divergence of the M. oryzae population. Population genetic relationship deducted by genome-wide single nucleotide polymorphisms showed that M. oryzae differentiation is highly consistent with the host speciation process. In particular, the rice-infecting population of M. oryzae is distinct from populations from other hosts. However, how genome regions prone to host-mediated selection pressures associated with speciation in M. oryzae, especially at a large-scale population level, has not been extensively characterized. Here, we detected strong evidence of sweep selection throughout the genomes of rice and non-rice pathotypes of M. oryzae population using integrated haplotype score (iHS), cross population extended haplotype homozygosity (XPEHH), and cross population composite likelihood ratio (XPCLR) tests. Functional annotation analyses of the genes associated with host-mediated selection pressure showed that 14 pathogenicity-related genes are under positive selection pressure. Additionally, we showed that 17 candidate effector proteins are under positive and divergent selection among the blast fungus population through sweep selection analysis. Specifically, we find that a divergent selective gene, MGG_13871, is experiencing host-directed mutation in two amino acid residues in rice and non-rice infecting populations. These results provide a crucial insight into the impact of selective sweeping on the differentiation of M. oryzae populations and the dynamic influences of genomic regions in promoting host adaptation and speciation among M. oryzae species.

Spatiotemporal variation in near-surface CH4 concentrations in China over the last two decades.

Methane is one of the main greenhouse trace gases and seriously affects the radiation balance of Earth systems due to its strong heat absorption capacity and long atmospheric retention time. Based on the methane stratification data simulated by the community atmospheric model with chemistry (CAM-chem), near-surface methane concentrations were estimated by utilizing the Gaussian function, and the spatiotemporal variation in the near-surface methane concentration in China from 2001 to 2019 was discussed in this research. The results show that (1) based on the methane stratification concentration data simulated by the atmospheric chemical model, the near-surface CH4 concentration estimated by Gaussian function model is reliable, which provides a new method to estimate the near-surface CH4 concentration over China; (2) from 2001 to 2019, the near-surface methane concentration in China showed an increasing trend with an annual growth rate of 7.20±0.23 ppb·a-1. The annual maximum near-surface methane concentration was measured in winter, and the minimum was measured in summer; (3) the spatial distribution differences are obvious: the methane concentration in the east was higher than that in the west, and the methane concentration in the north was higher than that in the south. Moreover, the distributions of methane in the east and west are consistent with the division of Hu Huanyong population line.

End-to-end automatic differentiation of the coronavirus disease 2019 (COVID-19) from viral pneumonia based on chest CT.

PURPOSE: In the absence of a virus nucleic acid real-time reverse transcriptase-polymerase chain reaction (RT-PCR) test and experienced radiologists, clinical diagnosis is challenging for viral pneumonia with clinical symptoms and CT signs similar to that of coronavirus disease 2019 (COVID-19). We developed an end-to-end automatic differentiation method based on CT images to identify COVID-19 pneumonia patients in real time. METHODS: From January 18 to February 23, 2020, we conducted a retrospective study and enrolled 201 patients from two hospitals in China who underwent chest CT and RT-PCR tests, of which 98 patients tested positive for COVID-19 (118 males and 83 females, with an average age of 42 years). Patient CT images from one hospital were divided among training, validation and test datasets with an 80%:10%:10% ratio. An end-to-end representation learning method using a large-scale bi-directional generative adversarial network (BigBiGAN) architecture was designed to extract semantic features from the CT images. The semantic feature matrix was input for linear classifier construction. Patients from the other hospital were used for external validation. Differentiation accuracy was evaluated using a receiver operating characteristic curve. RESULTS: Based on the 120-dimensional semantic features extracted by BigBiGAN from each image, the linear classifier results indicated that the area under the curve (AUC) in the training, validation and test datasets were 0.979, 0.968 and 0.972, respectively, with an average sensitivity of 92% and specificity of 91%. The AUC for external validation was 0.850, with a sensitivity of 80% and specificity of 75%. Publicly available architecture and computing resources were used throughout the study to ensure reproducibility. CONCLUSION: This study provides an efficient recognition method for coronavirus disease 2019 pneumonia, using an end-to-end design to implement targeted and effective isolation for the containment of this communicable disease.

[A multicenter survey of antibiotic use in very and extremely low birth weight infants in Hunan Province].

OBJECTIVE: To investigate the current status of antibiotic use for very and extremely low birth weight (VLBW/ELBW) infants in neonatal intensive care units (NICUs) of Hunan Province. METHODS: The use of antibiotics was investigated in multiple level 3 NICUs of Hunan Province for VLBW and ELBW infants born between January, 2017 and December, 2017. RESULTS: The clinical data of 1 442 VLBW/ELBW infants were collected from 24 NICUs in 2017. The median antibiotic use duration was 17 days (range: 0-86 days), accounting for 53.0% of the total length of hospital stay. The highest duration of antibiotic use was up to 91.4% of the total length of hospital stay, with the lowest at 14.6%. In 16 out of 24 NICUs, the antibiotic use duration was accounted for more than 50.0% of the hospitalization days. There were 113 cases with positive bacterial culture grown in blood or cerebrospinal fluid, making the positive rate of overall bacterial culture as 7.84%. The positive rate of bacterial culture in different NICUs was significantly different from 0% to 14.9%. The common isolated bacterial pathogens Klebsiella pneumoniae was 29 cases (25.7%); Escherichia coli 12 cases (10.6%); Staphylococcus aureus 3 cases (2.7%). The most commonly used antibiotics were third-generation of cephalosporins, accounting for 41.00% of the total antibiotics, followed by penicillins, accounting for 32.10%, and followed by carbapenems, accounting for 13.15%. The proportion of antibiotic use time was negatively correlated with birth weight Z-score and the change in weight Z-score between birth and hospital discharge (rs=-0.095, -0.151 respectively, P<0.01), positively correlated with death/withdrawal of care (rs=0.196, P<0.01). CONCLUSIONS: Antibiotics used for VLBW/ELBW infants in NICUs of Hunan Province are obviously prolonged in many NICUs. The proportion of routine use of third-generation of cephalosporins and carbapenems antibiotics is high among the NICUs.

Rapid adaptation of the Irish potato famine pathogen Phytophthora infestans to changing temperature.

Temperature plays a multidimensional role in host-pathogen interactions. As an important element of climate change, elevated world temperature resulting from global warming presents new challenges to sustainable disease management. Knowledge of pathogen adaptation to global warming is needed to predict future disease epidemiology and formulate mitigating strategies. In this study, 21 Phytophthora infestans isolates originating from seven thermal environments were acclimated for 200 days under stepwise increase or decrease of experimental temperatures and evolutionary responses of the isolates to the thermal changes were evaluated. We found temperature acclimation significantly increased the fitness and genetic adaptation of P. infestans isolates at both low and high temperatures. Low-temperature acclimation enforced the countergradient adaptation of the pathogen to its past selection and enhanced the positive association between the pathogen's intrinsic growth rate and aggressiveness. At high temperatures, we found that pathogen growth collapsed near the maximum temperature for growth, suggesting a thermal niche boundary may exist in the evolutionary adaptation of P. infestans. These results indicate that pathogens can quickly adapt to temperature shifts in global warming. If this is associated with environmental conditions favoring pathogen spread, it will threaten future food security and human health and require the establishment of mitigating actions.

Improvement of learning domains of nursing students with the use of authentic assessment pedagogy in clinical practice.

With evolving healthcare demands, nursing educators need to constantly review their teaching methodologies in order to enhance learners' knowledge and competency of skills in the clinical settings. Learning is an active process in which meaning is accomplished on the basis of experience and that authentic assessment pedagogy will enable nursing students to play an active part in their learning. The study was conducted with an aim to examine nursing students' learning domains through the introduction of the authentic assessment pedagogy during their clinical practice. A quasi-experimental study (n = 54) was conducted over a period of 10 weeks at a local tertiary hospital. The experimental group was exposed to the authentic assessment pedagogy and were taught to use the assessment rubrics as an instrument to help enhance their learning. Students were assessed and scored according to the assessment rubrics, which were categorized into four domains; cognitive, psychomotor, affective and critical thinking abilities. The findings indicated that an overall score for the four domains between the experimental and control groups were significant, with p value of <0.05. Critical thinking scores were indicative of consistent improvement within the experimental group. The findings confirmed that learning outcomes of the nursing students were enhanced through the early introduction of the authentic assessment pedagogy in the clinical setting.

Roll-to-roll embossing of optical linear Fresnel lens polymer film for solar concentration.

Roll-to-roll manufacturing has been proven to be a high-throughput and low-cost technology for continuous fabrication of functional optical polymer films. In this paper, we have firstly studied a complete manufacturing cycle of linear Fresnel lens polymer film for solar concentration in the aspects of ultra-precision diamond machining of metal roller mold, roll-to-roll embossing, and measurement on film profile and functionality. A metal roller mold patterned with linear Fresnel lenses is obtained using single point diamond turning technique. The roller mold is installed onto a self-developed roll-to-roll UV embossing system to realize continuous manufacturing of linear Fresnel lens film. Profile measurement of the machined roller mold and the embossed polymer film, which is conducted using a stylus profilometer, shows good agreement between measured facet angles with designed ones. Functionality test is conducted on a solar simulation system with a reference solar cell, and results show that strong light concentration is realized.

Profile uniformity of overlapped oxide dots induced by atomic force microscopy.

The technique for assembling a uniform oxide line by overlapping a series of nanosized oxide dots induced by atomic force microscopy is analytically and experimentally investigated. In addition to the normal continuous (static) pulses, the oxide growth rates under various discontinuous (modulated) pluses are studied to quantify the overlapping effect under multiple pulses used by the assembling technique. In the analysis of the assembling technique, the superposition principle is used to predict the assembled profiles and to define the uniformity criteria. Experiments have been performed to demonstrate the analytical prediction, including the threshold or minimum pitch for forming uniform lines, and the onset pitch for the overlapping effect to be considered. Indeed, by following the uniformity criteria developed, uniform and reliable oxide lines can be obtained by overlapping oxide dots.

[Investigation of birth state and disease spectrum in 1,434 hospitalized neonates].

OBJECTIVE: To investigate the birth state of neonates and the disease spectrum of hospitalized neonates from a primary hospital, and compare with the national data of the same period. METHODS: A retrospective investigation was carried out in 1,434 neonates born or hospitalized in this hospital from January 2005 to December 2005. RESULTS: During the investigation period, there were 1,100 neonates born in the department of obstetrics. The incidence of premature birth was 2.3%. The caesarean birth accounted for 54.2%, significantly higher than the national average (49.2%, p<0.01). The neonatal mortality was 0.2%. The incidences of antepartum hemorrhage, threatened abortion, and pregnancy infection in preterm infant' s mothers were significantly higher than those in full-term infant' s mothers. A total of 344 neonates were admitted to the department of pediatrics during the investigation period. Preterm infants accounted for 38.0% which was higher than the national average (26.2%; p<0.01). Beside preterm infants, asphyxia, respiratory distress syndrome (RDS), sepsis and intracranial hemorrhage were shown to have a significantly higher proportion than the national averages. The mortality of hospitalized neonates was 0.9%. CONCLUSIONS: The higher cesarean section rate should be controlled in our hospital. Prenatal health care and fetal monitoring should be strengthened to decrease the incidence of premature birth, RDS, sepsis and intracranial hemorrhage, thus reducing the mortality of neonates.