Latent class analysis-derived classification improves the cancer-specific death stratification of lymphomas: A large retrospective cohort study.

Lymphomas have diverse etiologies, treatment approaches, and prognoses. Accurate survival estimation is challenging for lymphoma patients due to their heightened susceptibility to non-lymphoma-related mortality. To overcome this challenge, we propose a novel lymphoma classification system that utilizes latent class analysis (LCA) and incorporates demographic and clinicopathological factors as indicators. We conducted LCA using data from 221,812 primary lymphoma patients in the Surveillance, Epidemiology, and End Results (SEER) database and identified four distinct LCA-derived classes. The LCA-derived classification efficiently stratified patients, thereby adjusting the bias induced by competing risk events such as non-lymphoma-related death. This remains effective even in cases of limited availability of cause-of-death information, leading to an enhancement in the accuracy of lymphoma prognosis assessment. Additionally, we validated the LCA-derived classification model in an external cohort and observed its improved prognostic stratification of molecular subtypes. We further explored the molecular characteristics of the LCA subgroups and identified potential driver genes specific to each subgroup. In conclusion, our study introduces a novel LCA-based lymphoma classification system that provides improved prognostic prediction by accounting for competing risk events. The proposed classification system enhances the clinical relevance of molecular subtypes and offers insights into potential therapeutic targets.

Duration-sensitive association between air pollution exposure and changes in cardiometabolic biomarkers: Evidence from a predominantly African American cohort.

BACKGROUND: Ambient fine particulate matter (PM2.5) exposure has been related to cardiometabolic diseases, but the underlying biological pathways remain unclear at the population level. OBJECTIVE: To investigate the effect of PM2.5 exposure on changes in multiple cardiometabolic biomarkers across different exposure durations. METHOD: Data from a prospective cohort study were analyzed. Ten cardiometabolic biomarkers were measured, including ghrelin, resistin, leptin, C-peptide, creatine kinase myocardial band (CK-MB), monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor alpha (TNF-alpha), N-terminal pro B-type natriuretic peptide (NT-proBNP), troponin, and interleukin-6 (IL-6). PM2.5 levels across exposure durations from 1 to 36 months were assessed. Mixed effect model was used to estimate changes in biomarker levels against 1 µg/m3 increase in PM2.5 level across different exposure durations. RESULTS: Totally, 641 participants were included. The average PM2.5 exposure level was 9 µg/m3. PM2.5 exposure was inversely associated with ghrelin, and positively associated with all other biomarkers. The magnitudes of these associations were duration-sensitive and exhibited a U-shaped or inverted-U-shaped trend. For example, the association of resistin were ß = 0.05 (95% CI: 0.00, 0.09) for 1-month duration, strengthened to ß = 0.27 (95% CI: 0.14, 0.41) for 13-month duration, and weakened to ß = 0.12 (95% CI: -0.03, 0.26) for 24-month duration. Similar patterns were observed for other biomarkers except for CK-MB, of which the association direction switched from negative to positive as the duration increased. Resistin, leptin, MCP-1, TNF-alpha, and troponin had a sensitive exposure duration of nearly 12 months. Ghrelin and C-peptide were more sensitive to longer-term exposure (>18 months), while NT-proBNP and IL-6 were more sensitive to shorter-term exposure (<6 months). CONCLUSION: PM2.5 exposure was associated with elevated levels in cardiometabolic biomarkers related to insulin resistance, inflammation, and heart injury. The magnitudes of these associations depended on the exposure duration. The most sensitive exposure durations of different biomarkers varied.

Low-cost optimization of industrial textile landfill sludge re-dewatering using ferrous sulfate and blast furnace slag.

This study was based on an industrial sludge landfill with a scale of 1 million cubic meters, which had been filled for more than 10 years. It focused on the secondary dewatering of industrial textile landfill sludge (LS) with a total organic carbon (TOC) content greater than 50% and a volatile suspended solids to suspended solids (VSS/SS) ratio of 0.59. A response surface methodology (RSM) model was established using the coagulant ferrous sulfate (FeSO4) and conditioning agents such as hydrated magnesium oxide (MgO), blast furnace slag (BFS), and calcium oxide (CaO). By solving the RSM equations for the respective indicators, the optimal dosages of FeSO4, MgO, and BFS were determined to be 90 mg/g of dry sludge (DS), and for CaO 174.85 mg/g DS. Further examinations of the dewatering performance, apparent properties, extracellular polymeric substances (EPS) components, rheological characteristics, moisture distribution, and pollutant content variation led to the development of a green waste-based dewatering agent composed of FeSO4 and BFS. In small-scale diaphragm plate and frame filter press tests, the optimal water content (WC) was 69.11%. In the final production-scale experiments, it was 65.72%, with the actual application cost being only 13.07 $/ton DS. Additionally, when FeSO4 and BFS were used together, the combined action of Fe and Si could significantly reduce the biotoxicity of heavy metals (HMs), cut down 75.2% of the LS's TOC, and effectively reduced the leaching of organic substances from the leachate, which was beneficial for subsequent disposal. In conclusion, the combined use of FeSO4 and BFS for the secondary dewatering of industrial textile LS was economically efficient, effective in dewatering, and had significant harm reduction effects, making it a worthwhile for waste treatment.

Per- and polyfluoroalkyl substances and sleep health in U.S. adults, NHANES 2005-2014.

BACKGROUND: Per- and polyfluoroalkyl substances (PFAS) are synthetic chemicals that induce oxidative inflammatory responses and disrupt the endocrine and central nervous systems, all of which can influence sleep. OBJECTIVE: To investigate the association between PFAS exposure and sleep health measures in U.S. adults. METHODS: We analyzed serum concentration data of four PFAS [perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), perfluorohexane sulfonic acid (PFHxS), and perfluorononanoic acid (PFNA)] reported for 8913 adults in NHANES 2005-2014. Sleep outcomes, including trouble sleeping, having a diagnosis of sleep disorder, and recent daily sleep duration classified as insufficient or excessive sleep (<6 or >9 h/day) were examined. Weighted logistic regression was used to estimate the association between the sleep outcomes and each PFAS modeled continuously (log2) or in exposure tertiles. We applied quantile g-computation to estimate the effect of the four PFAS as a mixture on the sleep outcomes. We conducted a quantitative bias analysis to assess the potential influence of self-selection and uncontrolled confounding. RESULTS: We observed some inverse associations between serum PFAS and trouble sleeping or sleep disorder, which were more consistent for PFOS (e.g., per log2-PFOS (ng/ml) and trouble sleeping OR = 0.93, 95%CI: 0.89, 0.98; sleep disorder OR = 0.89, 95%CI: 0.83, 0.95). Per quartile increase of the PFAS mixture was inversely associated with trouble sleeping and sleep disorder. No consistent associations were found for sleep duration across analyses. Our bias analysis suggests that the finding on sleep disorder could be explained by a moderate level of self-selection and negative confounding effects. CONCLUSIONS: We found no evidence to suggest exposure to four legacy PFAS worsened self-reported sleep health among U.S. adults. While some inverse associations between specific PFAS and sleep disorder were observed, self-selection and uncontrolled confounding biases may play a role in these findings.

Breast Tumor Classification in Ultrasound Images by Fusion of Deep Convolutional Neural Network and Shallow LBP Feature.

Breast cancer is one of the most dangerous and common cancers in women which leads to a major research topic in medical science. To assist physicians in pre-screening for breast cancer to reduce unnecessary biopsies, breast ultrasound and computer-aided diagnosis (CAD) have been used to distinguish between benign and malignant tumors. In this study, we proposed a CAD system for tumor diagnosis using a multi-channel fusion method and feature extraction structure based on multi-feature fusion on breast ultrasound (BUS) images. In the pre-processing stage, the multi-channel fusion method completed the color conversion of the BUS image to make it contain richer information. In the feature extraction stage, the pre-trained ResNet50 network was selected as the basic network, and three levels of features were combined based on adaptive spatial feature fusion (ASFF), and finally, the shallow local binary pattern (LBP) texture features were fused. Support vector machine (SVM) was used for comparative analysis. A retrospective analysis was carried out, and 1615 breast tumor images (572 benign and 1043 malignant) confirmed by pathological examinations were collected. After data processing and augmentation, for an independent test set consisting of 874 breast ultrasound images (457 benign and 417 malignant), the accuracy, precision, recall, specificity, F1 score, and AUC of our method were 96.91%, 98.75%, 94.72%, 98.91%, 0.97, and 0.991, respectively. The results show that the integration of shallow LBP texture features and multi-level depth features can more effectively improve the comprehensive performance of breast tumor diagnosis, and has strong clinical application value. Compared with the past methods, our proposed method is expected to realize the automatic diagnosis of breast tumors and provide an auxiliary tool for radiologists to accurately diagnose breast diseases.

Predicting spatiotemporally-resolved mean air temperature over Sweden from satellite data using an ensemble model.

Mapping of air temperature (Ta) at high spatiotemporal resolution is critical to reducing exposure assessment errors in epidemiological studies on the health effects of air temperature. In this study, we applied a three-stage ensemble model to estimate daily mean Ta from satellite-based land surface temperature (Ts) over Sweden during 2001-2019 at a high spatial resolution of 1 × 1 km2. The ensemble model incorporated four base models, including a generalized additive model (GAM), a generalized additive mixed model (GAMM), and two machine learning models (random forest [RF] and extreme gradient boosting [XGBoost]), and allowed the weights for each model to vary over space, with the best-performing model for each grid cell assigned the highest weight. Various spatial predictors were included as adjustment variables in all the base models, including land cover type, normalized difference vegetation index (NDVI), and elevation. The ensemble model showed high performance with an overall R2 of 0.98 and a root mean square error of 1.38 °C in the ten-fold cross-validation, and outperformed each of the four base models. Although each base model performed well, the two machine learning models (RF [R2 = 0.97], XGBoost [R2 = 0.98]) had better performance than the two regression models (GAM [R2 = 0.95], GAMM [R2 = 0.96]). In the machine learning models, Ts was the dominant predictor of Ta, followed by day of year, NDVI, latitude, elevation, and longitude. The highly spatiotemporally-resolved Ta can improve temperature exposure assessment in future epidemiological studies.

Rolling Bearing Fault Diagnosis Based on WOA-VMD-MPE and MPSO-LSSVM.

In order to further improve the accuracy of fault identification of rolling bearings, a fault diagnosis method based on the modified particle swarm optimization (MPSO) algorithm optimized least square support vector machine (LSSVM), combining parameter optimization variational mode decomposition (VMD) and multi-scale permutation entropy (MPE), was proposed. Firstly, to solve the problem of insufficient decomposition and mode mixing caused by the improper selection of mode component K and penalty factor α in VMD algorithm, the whale optimization algorithm (WOA) was used to optimize the penalty factor and mode component number in the VMD algorithm, and the optimal parameter combination (K, α) was obtained. Secondly, the optimal parameter combination (K, α) was used for the VMD of the rolling bearing vibration signal to obtain several intrinsic mode functions (IMFs). According to the Pearson correlation coefficient (PCC) criterion, the optimal IMF component was selected, and its optimal multi-scale permutation entropy was calculated to form the feature set. Finally, K-fold cross-validation was used to train the MPSO-LSSVM model, and the test set was input into the trained model for identification. The experimental results show that compared with PSO-SVM, LSSVM, and PSO-LSSVM, the MPSO-LSSVM fault diagnosis model has higher recognition accuracy. At the same time, compared with VMD-SE, VMD-MPE, and PSO-VMD-MPE, WOA-VMD-MPE can extract more accurate features.

Magnetic cotton textile wastes pyrolyzed by ferric cerium oxide for degradation of p-nitrophenol by catalytic ozonation.

In this paper, magnetic cotton textile wastes pyrolyzed by ferric cerium oxide (FexCey oxide/PC) were synthesized for degradation of p-nitrophenol by catalytic ozonation, and the optimal Fe-Ce ratio was 10:1. Compared to Fe10Ce1 oxide, the Fe10Ce1 oxide/PC not only greatly improved the degradation efficiency of PNP, but also reduced the dosage of catalyst. Through the BET test, the Fe10Ce1 oxide/PC has a high specific surface area to absorb part of the pollutants. VSM test shows that the material is magnetic and easy to recycle. Response surface methodology (RSM) was applied to optimize the experimental condition, and the optimal removal rate was 90% when the initial pH was 9, the catalyst dosage was 0.4 g/L, and the ozone addition was 1.77 L/min (5.9 mg/L). Finally, the mechanism of PNP degradation was explored utilizing inhibitor and ESR free radical detection. The adsorption capacity of the material and electron-absorbing property of PNP jointly determined the high catalytic efficiency with Fe10Ce1 oxide/PC in catalytic ozonation.

The S100 family is a prognostic biomarker and correlated with immune cell infiltration in pan-cancer.

BACKGROUND: The S100 protein family is a group of small molecular EF-hand calcium-binding proteins that play critical roles in various biological processes, including promotion of growth, metastasis and immune evasion of tumor. However, the potential roles of S100 protein family expression in tumor microenvironment (TME) cell infiltration in pan-cancer remain elusive. METHODS: Herein, we conducted a comprehensive assessment of the expression patterns of the S100 protein family in pan-cancer, meticulously examining their correlation with characteristics of TME cell infiltration. The S100 score was constructed to quantify S100 family expression patterns of individual tumors. RESULTS: The S100 family was a potent risk factor in many cancers. Clustering analysis based on the transcriptome patterns of S100 protein family identified two cancer clusters with distinct immunophenotypes and clinical characteristics. Cluster A, with lower S100 expression, exhibited lower immune infiltration, whereas, Cluster B, with higher S100 expression, featured higher immune infiltration. Interestingly, Cluster B had a poorer prognosis, likely due to an immune-excluded phenotype resulting from stromal activation. The analysis revealed robust enrichment of the TGFb and EMT pathways in the cohort exhibiting high S100 score, alongside a positive correlation between the S100 score and Treg levels, suggesting the manifestation of an immune-excluded phenotype in this group. Moreover, S100 families were associated with the prognosis of 22 different cancers and a noteworthy association was observed between high S100 score and an unfavorable response to anti-PD-1/L1 immunotherapy. Consistent findings across two independent immunotherapy cohorts substantiated the advantageous therapeutic outcomes and clinical benefits in patients displaying lower S100score. CONCLUSION: Our analysis demonstrated the role of S100 family in formation of TME diversity and complexity, enabling deeper cognition of TME infiltration characterization and the development of personalized immunotherapy strategies targeting S100 family for unique tumor types.

Dynamics of N6-methyladenosine modification during Alzheimer's disease development.

N6-methyladenosine (m6A) modification is a common RNA modification in the central nervous system and has been linked to various neurological disorders, including Alzheimer's disease (AD). However, the dynamic of mRNA m6A modification and m6A enzymes during the development of AD are not well understood. Therefore, this study examined the expression profiles of m6A and its enzymes in the development of AD. The results showed that changes in the expression levels of m6A regulatory factors occur in the early stages of AD, indicating a potential role for m6A modification in the onset of the disease. Additionally, the analysis of mRNA m6A expression profiles using m6A-seq revealed significant differences in m6A modification between AD and control brains. The genes with differential methylation were found to be enriched in GO and KEGG terms related to processes such as inflammation response, immune system processes. And the differently expressed genes (DEGs) are negatively lryassociated with genes involved in microglia hemostasis, but positively associated with genes related to "disease-associated microglia" (DAM) associated genes. These findings suggest that dysregulation of mRNA m6A modification may contribute to the development of AD by affecting the function and gene expression of microglia.

Contextual Deprivation, Race and Ethnicity, and Income in Air Pollution and Cardiovascular Disease.

Importance: Socioeconomically disadvantaged subpopulations are more vulnerable to fine particulate matter (PM2.5) exposure. However, as prior studies focused on individual-level socioeconomic characteristics, how contextual deprivation modifies the association of PM2.5 exposure with cardiovascular health remains unclear. Objective: To assess disparities in PM2.5 exposure association with cardiovascular disease among subpopulations defined by different socioeconomic characteristics. Design, Setting, and Participants: This cohort study used longitudinal data on participants with electronic health records (EHRs) from the All of Us Research Program between calendar years 2016 and 2022. Statistical analysis was performed from September 25, 2023, through February 23, 2024. Exposure: Satellite-derived 5-year mean PM2.5 exposure at the 3-digit zip code level according to participants' residential address. Main Outcome and Measures: Incident myocardial infarction (MI) and stroke were obtained from the EHRs. Stratified Cox proportional hazards regression models were used to estimate the hazard ratio (HR) between PM2.5 exposure and incident MI or stroke. We evaluated subpopulations defined by 3 socioeconomic characteristics: contextual deprivation (less deprived, more deprived), annual household income (≥$50 000, <$50 000), and race and ethnicity (non-Hispanic Black, non-Hispanic White). We calculated the ratio of HRs (RHR) to quantify disparities between these subpopulations. Results: A total of 210 554 participants were analyzed (40% age >60 years; 59.4% female; 16.7% Hispanic, 19.4% Non-Hispanic Black, 56.1% Non-Hispanic White, 7.9% other [American Indian, Asian, more than 1 race and ethnicity]), among whom 954 MI and 1407 stroke cases were identified. Higher PM2.5 levels were associated with higher MI and stroke risks. However, disadvantaged groups (more deprived, income <$50 000 per year, Black race) were more vulnerable to high PM2.5 levels. The disparities were most pronounced between groups defined by contextual deprivation. For instance, increasing PM2.5 from 6 to 10 µg/m3, the HR for stroke was 1.13 (95% CI, 0.85-1.51) in the less-deprived vs 2.57 (95% CI, 2.06-3.21) in the more-deprived cohort; 1.46 (95% CI, 1.07-2.01) in the $50 000 or more per year vs 2.27 (95% CI, 1.73-2.97) in the under $50 000 per year cohort; and 1.70 (95% CI, 1.35-2.16) in White individuals vs 2.76 (95% CI, 1.89-4.02) in Black individuals. The RHR was highest for contextual deprivation (2.27; 95% CI, 1.59-3.24), compared with income (1.55; 95% CI, 1.05-2.29) and race and ethnicity (1.62; 95% CI, 1.02-2.58). Conclusions and Relevance: In this cohort study, while individual race and ethnicity and income remained crucial in the adverse association of PM2.5 with cardiovascular risks, contextual deprivation was a more robust socioeconomic characteristic modifying the association of PM2.5 exposure.

Differing associations of PM2.5 exposure with systolic and diastolic blood pressures across exposure durations in a predominantly non-Hispanic Black cohort.

Environmental health research has suggested that fine particulate matter (PM2.5) exposure can lead to high blood pressures, but it is unclear whether the impacts remain the same for systolic and diastolic blood pressures (SBP and DBP). This study aimed to examine whether the effects of PM2.5 exposure on SBP and DBP differ using data from a predominantly non-Hispanic Black cohort collected between 2013 and 2019 in the US. PM2.5 exposure was assessed based on a satellite-derived model across exposure durations from 1 to 36 months. The average PM2.5 exposure level was between 9.5 and 9.8 µg/m3 from 1 through 36 months. Mixed effects models were used to estimate the association of PM2.5 with SBP, DBP, and related hypertension types, adjusted for potential confounders. A total of 6381 participants were included. PM2.5 exposure was positively associated with both SBP and DBP. The association magnitudes depended on exposure durations. The association with SBP was null at the 1-month duration (ß = 0.05, 95% CI: - 0.23, 0.33), strengthened as duration increased, and plateaued at the 24-month duration (ß = 1.14, 95% CI: 0.54, 1.73). The association with DBP started with ß = 0.29 (95% CI: 0.11, 0.47) at the 1-month duration, and plateaued at the 12-month duration (ß = 1.61, 95% CI: 1.23, 1.99). PM2.5 was associated with isolated diastolic hypertension (12-month duration: odds ratio = 1.20, 95% CI: 1.07, 1.34) and systolic-diastolic hypertension (12-month duration: odds ratio = 1.18, 95% CI: 1.10, 1.26), but not with isolated systolic hypertension. The findings suggest DBP is more sensitive to PM2.5 exposure and support differing effects of PM2.5 exposure on SBP and DBP. As elevation of SBP and DBP differentially predict CVD outcomes, this finding is relevant for prevention and treatment.

Impact of Warehouse Expansion on Ambient PM2.5 and Elemental Carbon Levels in Southern California's Disadvantaged Communities: A Two-Decade Analysis.

Over the past two decades, the surge in warehouse construction near seaports and in economically lower-cost land areas has intensified product transportation and e-commerce activities, particularly affecting air quality and health in nearby socially disadvantaged communities. This study, spanning from 2000 to 2019 in Southern California, investigated the relationship between ambient concentrations of PM2.5 and elemental carbon (EC) and the proliferation of warehouses. Utilizing satellite-driven estimates of annual mean ambient pollution levels at the ZIP code level and linear mixed effect models, positive associations were found between warehouse characteristics such as rentable building area (RBA), number of loading docks (LD), and parking spaces (PS), and increases in PM2.5 and EC concentrations. After adjusting for demographic covariates, an Interquartile Range increase of the RBA, LD, and PS were associated with a 0.16 µg/m³ (95% CI = [0.13, 0.19], p < 0.001), 0.10 µg/m³ (95% CI = [0.08, 0.12], p < 0.001), and 0.21 µg/m³ (95% CI = [0.18, 0.24], p < 0.001) increase in PM2.5, respectively. For EC concentrations, an IQR increase of RBA, LD, and PS were each associated with a 0.021 µg/m³ (95% CI = [0.019, 0.024], p < 0.001), 0.014 µg/m³ (95% CI = [0.012, 0.015], p < 0.001), and 0.021 µg/m³ (95% CI = [0.019, 0.024], p < 0.001) increase. The study also highlighted that disadvantaged populations, including racial/ethnic minorities, individuals with lower education levels, and lower-income earners, were disproportionately affected by higher pollution levels.

Cabrol procedure and its modifications: a systematic review and meta-analysis.

BACKGROUND: The Cabrol procedure has undergone various modifications and developments since its invention. However, there is a notable gap in the literature regarding meta-analyses assessing it. METHODS: A systematic review and meta-analysis was conducted to evaluate the effectiveness and long-term outcomes of the Cabrol procedure and its modifications. Pooling was conducted using random effects model. Outcome events were reported as linearized occurrence rates (percentage per patient-year) with 95% confidence intervals. RESULTS: A total of 14 studies involving 833 patients (mean age: 50.8 years; 68.0% male) were included in this meta-analysis. The pooled all-cause early mortality was 9.0% (66 patients), and the combined rate of reoperation due to bleeding was 4.9% (17 patients). During the average 4.4-year follow-up (3,727.3 patient-years), the annual occurrence rates (linearized) for complications were as follows: 3.63% (2.79-4.73) for late mortality, 0.64% (0.35-1.16) for aortic root reoperation, 0.57% (0.25-1.31) for hemorrhage events, 0.66% (0.16-2.74) for thromboembolism, 0.60% (0.29-1.26) for endocarditis, 2.32% (1.04-5.16) for major valve-related adverse events, and 0.58% (0.34-1.00) for Cabrol-related coronary graft complications. CONCLUSION: This systematic review provides evidence that the outcomes of the Cabrol procedure and its modifications are acceptable in terms of mortality, reoperation, anticoagulation, and valve-related complications, especially in Cabrol-related coronary graft complications. Notably, the majority of Cabrol procedures were performed in reoperations and complex cases. Furthermore, the design and anastomosis of the Dacron interposition graft for coronary reimplantation, considering natural anatomy and physiological hemodynamics, may promise future advancements in this field.

Self-Supervised 3D Behavior Representation Learning Based on Homotopic Hyperbolic Embedding.

Behavior sequences are generated by a series of spatio-temporal interactions and have a high-dimensional nonlinear manifold structure. Therefore, it is difficult to learn 3D behavior representations without relying on supervised signals. To this end, self-supervised learning methods can be used to explore the rich information contained in the data itself. Context-context contrastive self-supervised methods construct the manifold embedded in Euclidean space by learning the distance relationship between data, and find the geometric distribution of data. However, traditional Euclidean space is difficult to express context joint features. In order to obtain an effective global representation from the relationship between data under unlabeled conditions, this paper adopts contrastive learning to compare global feature, and proposes a self-supervised learning method based on hyperbolic embedding to mine the nonlinear relationship of behavior trajectories. This method adopts the framework of discarding negative samples, which overcomes the shortcomings of the paradigm based on positive and negative samples that pull similar data away in the feature space. Meanwhile, the output of the network is embedded in a hyperbolic space, and a multi-layer perceptron is added to convert the entire module into a homotopic mapping by using the geometric properties of operations in the hyperbolic space, so as to obtain homotopy invariant knowledge. The proposed method combines the geometric properties of hyperbolic manifolds and the equivariance of homotopy groups to promote better supervised signals for the network, which improves the performance of unsupervised learning.

Learning from pseudo-lesion: a self-supervised framework for COVID-19 diagnosis.

The Coronavirus disease 2019 (COVID-19) has rapidly spread all over the world since its first report in December 2019, and thoracic computed tomography (CT) has become one of the main tools for its diagnosis. In recent years, deep learning-based approaches have shown impressive performance in myriad image recognition tasks. However, they usually require a large number of annotated data for training. Inspired by ground glass opacity, a common finding in COIVD-19 patient's CT scans, we proposed in this paper a novel self-supervised pretraining method based on pseudo-lesion generation and restoration for COVID-19 diagnosis. We used Perlin noise, a gradient noise based mathematical model, to generate lesion-like patterns, which were then randomly pasted to the lung regions of normal CT images to generate pseudo-COVID-19 images. The pairs of normal and pseudo-COVID-19 images were then used to train an encoder-decoder architecture-based U-Net for image restoration, which does not require any labeled data. The pretrained encoder was then fine-tuned using labeled data for COVID-19 diagnosis task. Two public COVID-19 diagnosis datasets made up of CT images were employed for evaluation. Comprehensive experimental results demonstrated that the proposed self-supervised learning approach could extract better feature representation for COVID-19 diagnosis, and the accuracy of the proposed method outperformed the supervised model pretrained on large-scale images by 6.57% and 3.03% on SARS-CoV-2 dataset and Jinan COVID-19 dataset, respectively.

Influence of Low-Temperature Cap Layer Thickness on Luminescence Characteristics of Green InGaN/GaN Quantum Wells.

GaN cap layer with different thicknesses was grown on each InGaN well layer during MOCVD growth for InGaN/GaN multiple quantum well (MQW) samples to study the influence of the cap layer on the photoluminescence (PL) characteristics of MQWs. Through the temperature-dependent (TD) PL spectra, it was found that when the cap layer was too thick, the localized states of the quantum wells were relatively non-uniform. The thicker the well layer, the worse the uniformity of the localized states. Furthermore, through micro-area fluorescence imaging tests, it was found that when the cap layer was too thick, the luminescence quality of the quantum well was worse. In summary, the uniformity of the localized states in the quantum wells and the luminescence characteristics of the quantum wells could be improved when a relatively thin cap layer of the quantum well was prepared during the growth. These results could facilitate high efficiency QW preparation, especially for green LEDs.

Application of multi-angle spaceborne observations in characterizing the long-term particulate organic carbon pollution in China.

Ambient PM2.5 pollution is recognized as a leading environmental risk factor, causing significant mortality and morbidity in China. However, the specific contributions of individual PM2.5 constituents remain unclear, primarily due to the lack of a comprehensive ground monitoring network for constituents. This issue is particularly critical for carbonaceous species such as organic carbon (OC) and elemental carbon (EC), which are known for their significant health impacts, and understanding the OC/EC ratio is crucial for identifying pollution sources. To address this, we developed a Super Learner model integrating Multi-angle Imaging SpectroRadiometer (MISR) retrievals to predict daily OC concentrations across China from 2003 to 2019 at a 10-km spatial resolution. Our model demonstrates robust predictive accuracy, as evidenced by a random cross-validation R2 of 0.84 and an RMSE of 4.9 µg/m3, at the daily level. Although MISR is a polar-orbiting instrument, its fractional aerosol data make a significant contribution to the OC exposure model. We then use the model to explore the spatiotemporal distributions of OC and further calculate the EC/OC ratio in China. We compared regional pollution discrepancies and source contributions of carbonaceous pollution over three selected regions: Beijing-Tianjin-Hebei, Fenwei Plain, and Yunnan Province. Our model observes that OC levels are elevated in Northern China due to industrial operations and central heating during the heating season, while in Yunnan, OC pollution is mainly contributed by local forest fires during fire seasons. Additionally, we found that OC pollution in China is likely influenced by climate phenomena such as the El Niño-Southern Oscillation. Considering that climate change is increasing the severity of OC concentrations with more frequent fire events, and its influence on OC formation and dispersion, we suggest emphasizing the role of climate change in future OC pollution control policies. We believe this study will contribute to future epidemiological studies on OC, aiding in refining public health guidelines and enhancing air quality management in China.

Surface mechanism and optimization of catalytic ozonation with CoxFe oxides as catalyst for degradation of sodium p-toluenesulfonate in water.

In this study, the removals of sodium p-toluenesulfonate (NaTSA) by catalytic ozonation with two different cobalt-iron compounds, CoxFe oxides prepared by co-precipitation/calcination (CPO) and CoxFe oxides prepared by direct calcination (DCO), as the catalysts, had a difference of about 12%. It was found that the CPO surface contained active type c water, which was generally adsorbed on the oxygen vacancy. The test of oxygen temperature-programmed desorption (O2-TPD) showed that the surface of CPO was rich in oxygen vacancy. Through the electrochemical oxygen evolution reaction (OER) detection, a pair of Co valence redox peaks were detected from the CV curves, and the results of XPS test showed the replacement of octahedral Co3+ with Fe3 + in the Co3O4 during preparation of CPO. The enriched oxygen vacancy could be used as active sites for ozone adsorption and improve the charge transfer capacity. The number of hydroxyl radicals was detected by electron spin resonance (EPR) and it indicated that CPO contained more hydroxyl radicals, so it had higher effect in catalytic ozonation for organic pollutant degradation. In this paper, the relationship between oxygen vacancy and reactive center in the microstructure of the catalysts was established to discuss their working mechanism. The influence of the initial pH value, catalyst dosage, and ozone concentration on the removal of NaTSA was investigated by response surface design, and the optimal experimental conditions were predicted and verified.