An abnormal metabolism-related gene, ALG3, is a potential diagnostic and prognostic biomarker for lung adenocarcinoma.

BACKGROUND: To explore the abnormal metabolism-related genes that affect the prognosis of patients with lung adenocarcinoma (LUAD), and analyze the relationship with immune infiltration and competing endogenous RNA (ceRNA) network. METHODS: Transcriptome data of LUAD were downloaded from the Cancer Genome Atlas database. Abnormal metabolism-related differentially expressed genes in LUAD were screened by the R language. Cox analysis was used to construct LUAD prognostic risk model. Kaplan-Meier test, ROC curve and nomograms were used to evaluate the predictive ability of metabolic related gene prognostic model. CIBERSORT algorithm was used to analyze the relationship between risk score and immune infiltration. The starBase database constructed a regulatory network consistent with the ceRNA hypothesis. IHC experiments were performed to verify the differential expression of ALG3 in LUAD and paracancerous samples. RESULTS: In this study, 42 abnormal metabolism-related differential genes were screened. After survival analysis, the final 5 metabolism-related genes were used as the construction of prognosis model, including ALG3, COL7A1, KL, MST1, and SLC52A1. In the model, the survival rate of LUAD patients in the high-risk subgroup was lower than that in the low-risk group. In addition, the risk score of the constructed LUAD prognostic model can be used as an independent prognostic factor for patients. According to the analysis of CIBERSORT algorithm, the risk score is related to the infiltration of multiple immune cells. The potential ceRNA network of model genes in LUAD was constructed through the starBase database. IHC experiments revealed that ALG3 expression was upregulated in LUAD. CONCLUSION: The prognostic model of LUAD reveals the relationship between metabolism and prognosis of LUAD, and provides a novel perspective for diagnosis and research of LUAD.

Construction and validation of a prognostic model for esophageal cancer based on prognostic-related RNA-binding protein.

BACKGROUND: Construction of a prognostic model for esophageal cancer (ESCA) based on prognostic RNA-binding proteins (RBPs) and preliminary evaluation of RBP function. METHODS: RNA-seq data of ESCA was downloaded from The Cancer Genome Atlas database and mRNA was extracted to screen differentially expressed genes using R. After screening RBPs in differentially expressed genes, R packages clusterProfiler and pathview were used to analyze the RBPs for Gene Ontology enrichment and Kyoto Encyclopedia of Genes and Genomes pathway. Based on the prognosis-related RBPs, COX regression was used to establish the prognostic risk model of ESCA. Risk model predictive ability was assessed using calibration analysis, receiver operating characteristic curves, Kaplan-Meier curves, decision curve analysis, and Harrell consistency index (C-index). A nomogram was established by combining the risk model with clinicopathological features. RESULTS: A total of 105 RBPs were screened from ESCA. A prognostic risk model consisting of 6 prognostic RBPs (ARHGEF28, BOLL, CIRBP, DKC1, SNRPB, and TRIT1) was constructed by COX regression analysis. The prognosis was worse in the high-risk group, and the receiver operating characteristic curve showed (area under the curve = 0.90) that the model better predicted patients' 5-year survival. In addition, 6 prognostic RBPs had good diagnostic power for ESCA. In addition, a total of 39 mRNAs were identified as predicted target molecules for DKC1. CONCLUSION: ARHGEF28, BOLL, CIRBP, DKC1, SNRPB, and TRIT1, as RBPs, are associated with the prognosis of ESCA, which may provide new ideas for targeted therapy of ESCA.

Lipopolysaccharide-induced bacterial infection model: microRNA-370-3p participates in the anti-infection response by targeting the macrophage TLR4-NLRP3 caspase-1 cellular pyroptosis pathway.

OBJECTIVE: To explore the effect of miR-370-3p on LPS triggering, in particular its involvement in disease progression by targeting the TLR4-NLRP3-caspase-1 cellular pyroptosis pathway in macrophages. METHODS: Human macrophage RAW264.7 was divided into 6 groups: control, LPS, LPS + inhibitor-NC, LPS + miR-370-3p inhibitor, LPS + mimics-NC and LPS + miR-370-3p mimics. RT-qPCR was used to detect the expression level of miR-370-3p and analyzed comparatively. CCK-8 and flow cytometry assays were used to detect cell viability and apoptosis. ELISA assay was used to detect the levels of IL-1ß and TNF-α in the supernatant of the cells. The WB assay was used to detect TLR4, NLRP3, Caspase-1 and GSDMD levels. RESULTS: After LPS induction, macrophage miR-370-3p levels decreased, cell viability decreased, and apoptosis increased. At the same time, the levels of TLR4, NLRP3, Caspase-1 and GSDMD increased in the cells, and the levels of IL-1ß and TNF-α increased in the cell supernatant. Compared with the LPS group, the significantly higher expression level of miR-370-3p in the cells of the LPS + miR-370-3p mimics group was accompanied by significantly higher cell viability, significantly lower apoptosis rate, significantly lower levels of TLR4, NLRP3, Caspase-1, and GSDMD in the cells, and significantly lower levels of IL-1ß and TNF-α in the cell supernatant. CONCLUSION: MiR-370-3p may be involved in anti-infective immune responses by targeting and inhibiting the macrophage TLR4-NLRP3-caspase-1 cellular pyroptosis pathway.

A deep learning radiomics model based on CT images for predicting the biological activity of hepatic cystic echinococcosis.

Introduction: Hepatic cystic echinococcosis (HCE) is a widely seen parasitic infection. Biological activity is crucial for treatment planning. This work aims to explore the potential applications of a deep learning radiomics (DLR) model, based on CT images, in predicting the biological activity grading of hepatic cystic echinococcosis. Methods: A retrospective analysis of 160 patients with hepatic echinococcosis was performed (127 and 33 in training and validation sets). Volume of interests (VOIs) were drawn, and radiomics features and deep neural network features were extracted. Feature selection was performed on the training set, and radiomics score (Rad Score) and deep learning score (Deep Score) were calculated. Seven diagnostics models (based on logistic regression algorithm) for the biological activity grading were constructed using the selected radiomics features and two deep model features respectively. All models were evaluated using the receiver operating characteristic curve, and the area under the curve (AUC) was calculated. A nomogram was constructed using the combined model, and its calibration, discriminatory ability, and clinical utility were assessed. Results: 12, 6 and 10 optimal radiomics features, deep learning features were selected from two deep learning network (DLN) features, respectively. For biological activity grading of hepatic cystic echinococcosis, the combined model demonstrated strong diagnostic performance, with an AUC value of 0.888 (95% CI: 0.837-0.936) in the training set and 0.876 (0.761-0.964) in the validation set. The clinical decision analysis curve indicated promising results, while the calibration curve revealed that the nomogram's prediction result was highly compatible with the actual result. Conclusion: The DLR model can be used for predicting the biological activity grading of hepatic echinococcosis.

Predicting hospitalization costs for pulmonary tuberculosis patients based on machine learning.

BACKGROUND: Pulmonary tuberculosis (PTB) is a prevalent chronic disease associated with a significant economic burden on patients. Using machine learning to predict hospitalization costs can allocate medical resources effectively and optimize the cost structure rationally, so as to control the hospitalization costs of patients better. METHODS: This research analyzed data (2020-2022) from a Kashgar pulmonary hospital's information system, involving 9570 eligible PTB patients. SPSS 26.0 was used for multiple regression analysis, while Python 3.7 was used for random forest regression (RFR) and MLP. The training set included data from 2020 and 2021, while the test set included data from 2022. The models predicted seven various costs related to PTB patients, including diagnostic cost, medical service cost, material cost, treatment cost, drug cost, other cost, and total hospitalization cost. The model's predictive performance was evaluated using R-square (R2), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) metrics. RESULTS: Among the 9570 PTB patients included in the study, the median and quartile of total hospitalization cost were 13,150.45 (9891.34, 19,648.48) yuan. Nine factors, including age, marital status, admission condition, length of hospital stay, initial treatment, presence of other diseases, transfer, drug resistance, and admission department, significantly influenced hospitalization costs for PTB patients. Overall, MLP demonstrated superior performance in most cost predictions, outperforming RFR and multiple regression; The performance of RFR is between MLP and multiple regression; The predictive performance of multiple regression is the lowest, but it shows the best results for Other costs. CONCLUSION: The MLP can effectively leverage patient information and accurately predict various hospitalization costs, achieving a rationalized structure of hospitalization costs by adjusting higher-cost inpatient items and balancing different cost categories. The insights of this predictive model also hold relevance for research in other medical conditions.

Credibility of the evidence on green space and human health: an overview of meta-analyses using evidence grading approaches.

BACKGROUND: Green space is an important part of the human living environment, with many epidemiological studies estimating its impact on human health. However, no study has quantitatively assessed the credibility of the existing evidence, impeding their translations into policy decisions and hindering researchers from identifying new research gaps. This overview aims to evaluate and rank such evidence credibility. METHODS: Following the PRISMA guideline, we systematically searched PubMed, Web of Science, and Embase databases for systematic reviews with meta-analyses concerning green spaces and health outcomes published up to January 15, 2024. We categorized the credibility of meta-analytical evidence from interventional studies into four levels (i.e., high, moderate, low, and very low) using the Grading of Recommendation, Assessment, Development and Evaluations framework, based on five domains including risk of bias, inconsistency, indirectness, imprecision, and publication bias. Further, we recalculated all the meta-analyses from observational studies and classified evidence into five levels (i.e., convincing, highly suggestive, suggestive, weak, and non-significant) by considering stringent thresholds for P-values, sample size, robustness, heterogeneity, and testing for biases. FINDINGS: In total, 154 meta-analysed associations (interventional = 44, observational = 110) between green spaces and health outcomes were graded. Among meta-analyses from interventional studies, zero, four (wellbeing, systolic blood pressure, negative affect, and positive affect), 20, and 20 associations between green spaces and health outcomes were graded as high, moderate, low, and very low credibility evidence, respectively. Among meta-analyses from observational studies, one (cardiovascular disease mortality), four (prevalence/incidence of diabetes mellitus, preterm birth, and small for gestational age infant, and all-cause mortality), 12, 22, and 71 associations were categorized as convincing, highly suggestive, suggestive, weak, and non-significant evidence, respectively. INTERPRETATION: The current evidence largely confirms beneficial associations between green spaces and human health. However, only a small subset of these associations can be deemed to have a high or convincing credibility. Hence, future better designed primary studies and meta-analyses are still needed to provide higher quality evidence for informing health promotion strategies. FUNDING: The National Natural Science Foundation of China of China; the Guangzhou Science and Technology Program; the Guangdong Medical Science and Technology Research Fund; the Research Grant Council of the Hong Kong SAR; and Sino-German mobility program.

Multiparametric MRI-Based Interpretable Radiomics Machine Learning Model Differentiates Medulloblastoma and Ependymoma in Children: A Two-Center Study.

RATIONALE AND OBJECTIVES: Medulloblastoma (MB) and Ependymoma (EM) in children, share similarities in age group, tumor location, and clinical presentation. Distinguishing between them through clinical diagnosis is challenging. This study aims to explore the effectiveness of using radiomics and machine learning on multiparametric magnetic resonance imaging (MRI) to differentiate between MB and EM and validate its diagnostic ability with an external set. MATERIALS AND METHODS: Axial T2 weighted image (T2WI) and contrast-enhanced T1weighted image (CE-T1WI) MRI sequences of 135 patients from two centers were collected as train/test sets. Volume of interest (VOI) was manually delineated by an experienced neuroradiologist, supervised by a senior. Feature selection analysis and the least absolute shrinkage and selection operator (LASSO) algorithm identified valuable features, and Shapley additive explanations (SHAP) evaluated their significance. Five machine-learning classifiers-extreme gradient boosting (XGBoost), Bernoulli naive Bayes (Bernoulli NB), Logistic Regression (LR), support vector machine (SVM), linear support vector machine (Linear SVC) classifiers were built based on T2WI (T2 model), CE-T1WI (T1 model), and T1 + T2WI (T1 + T2 model). A human expert diagnosis was developed and corrected by senior radiologists. External validation was performed at Sun Yat-Sen University Cancer Center. RESULTS: 31 valuable features were extracted from T2WI and CE-T1WI. XGBoost demonstrated the highest performance with an area under the curve (AUC) of 0.92 on the test set and maintained an AUC of 0.80 during external validation. For the T1 model, XGBoost achieved the highest AUC of 0.85 on the test set and the highest accuracy of 0.71 on the external validation set. In the T2 model, XGBoost achieved the highest AUC of 0.86 on the test set and the highest accuracy of 0.82 on the external validation set. The human expert diagnosis had an AUC of 0.66 on the test set and 0.69 on the external validation set. The integrated T1 + T2 model achieved an AUC of 0.92 on the test set, 0.80 on the external validation set, achieved the best performance. Overall, XGBoost consistently outperformed in different classification models. CONCLUSION: The combination of radiomics and machine learning on multiparametric MRI effectively distinguishes between MB and EM in childhood, surpassing human expert diagnosis in training and testing sets.

Deep learning on longitudinal CT scans: automated prediction of treatment outcomes in hospitalized tuberculosis patients.

Three deep learning (DL)-based prediction models (PMs) using longitudinal CT images were developed to predict tuberculosis (TB) treatment outcomes. The internal dataset consists of 493 bacteriologically confirmed TB patients who completed the anti-tuberculosis treatment with three-time CT scans, including a pretreatment CT scan and two follow-up CT scans. PM1 was trained using only pretreatment CT scans, and PM2 and PM3 were developed by adding follow-up scans. An independent testing was performed on external dataset comprising 86 TB patients. The area under the curve for classifying success and drug-resistant (DR)-TB was improved on both internal (0.609 vs. 0.625 vs. 0.815) and external (0.627 vs. 0.705 vs. 0.735) dataset by adding follow-up scans. The accuracy and F1-score also showed an increasing tendency in the external test. Regular follow-up CT scans can aid in the treatment prediction, and special attention should be given to early intensive phase of treatment to identify high-risk DR-TB patients.

Deep learning and radiomics of longitudinal CT scans for early prediction of tuberculosis treatment outcomes.

BACKGROUND: To predict tuberculosis (TB) treatment outcomes at an early stage, prevent poor outcomes ofdrug-resistant tuberculosis(DR-TB) and interrupt transmission. METHODS: An internal cohort for model development consists of 204 bacteriologically-confirmed TB patients who completed anti-tuberculosis treatment, with one pretreatment and two follow-up CT images (612 scans). Three radiomics feature-based models (RM) with multiple classifiers of Bagging, Random forest and Gradient boosting and two deep-learning-based models (i.e., supervised deep-learning model, SDLM; weakly supervised deep-learning model, WSDLM) are developed independently. Prediction scores of RM and deep-learning models with respectively highest performance are fused to create new fusion models under different fusion strategies. An additional independent validation was conducted on the external cohort comprising 80 patients (160 scans). RESULTS: For RM scheme, 16 optimal radiomics features are finally selected using longitudinal scans. The AUCs of RM for Bagging, Random forest and Gradient boosting were 0.789, 0.773 and 0.764 in the internal cohort and 0.840, 0.834 and 0.816 in the external cohort, respectively. For deep learning-based scheme, AUCs of SDLM and WSDLM were 0.767 and 0.661 in the internal cohort, and 0.823 and 0.651 in the external. The fusion model yields AUCs from 0.767 to 0.802 in the internal cohort, and from 0.831 to 0.857 in the external cohort. CONCLUSIONS: Fusion of radiomics features and deep-learning model may have the potential to predict early failure outcome of DR-TB, which may be combined to help prevent poor TB treatment outcomes.

Outdoor environmental exposome and the burden of tuberculosis: Findings from nearly two million adults in northwestern China.

We simultaneously assessed the associations for a range of outdoor environmental exposures with prevalent tuberculosis (TB) cases in a population-based health program with 1940,622 participants ≥ 15 years of age. TB status was confirmed through bacteriological and clinical assessment. We measured 14 outdoor environmental exposures at residential addresses. An exposome-wide association study (ExWAS) approach was used to estimate cross-sectional associations between environmental exposures and prevalent TB, an adaptive elastic net model (AENET) was implemented to select important exposure(s), and the Extreme Gradient Boosting algorithm was subsequently applied to assess their relative importance. In ExWAS analysis, 12 exposures were significantly associated with prevalent TB. Eight of the exposures were selected as predictors by the AENET model: particulate matter ≤ 2.5 µm (odds ratio [OR]=1.01, p = 0.3295), nitrogen dioxide (OR=1.09, p < 0.0001), carbon monoxide (OR=1.19, p < 0.0001), and wind speed (OR=1.08, p < 0.0001) were positively associated with the odds of prevalent TB while sulfur dioxide (OR=0.95, p = 0.0017), altitude (OR=0.97, p < 0.0001), artificial light at night (OR=0.98, p = 0.0001), and proportion of forests, shrublands, and grasslands (OR=0.95, p < 0.0001) were negatively associated with the odds of prevalent TB. Air pollutants had higher relative importance than meteorological and geographical factors, and the outdoor environment collectively explained 11% of TB prevalence.

Short-Term Clinical Response and Changes in the Fecal Microbiota and Metabolite Levels in Patients with Crohn's Disease After Stem Cell Infusions.

Recent studies have shown a close relationship between the gut microbiota and Crohn's disease (CD). This study aimed to determine whether mesenchymal stem cell (MSC) treatment alters the gut microbiota and fecal metabolite pathways and to establish the relationship between the gut microbiota and fecal metabolites. Patients with refractory CD were enrolled and received 8 intravenous infusions of MSCs at a dose of 1.0 × 106 cells/kg. The MSC efficacy and safety were evaluated. Fecal samples were collected, and their microbiomes were analyzed by 16S rDNA sequencing. The fecal metabolites at baseline and after 4 and 8 MSC infusions were identified by liquid chromatography-mass spectrometry (LC--MS). A bioinformatics analysis was conducted using the sequencing data. No serious adverse effects were observed. The clinical symptoms and signs of patients with CD were substantially relieved after 8 MSC infusions, as revealed by changes in weight, the CD activity index (CDAI) score, C-reactive protein (CRP) level, and erythrocyte sedimentation rate (ESR). Endoscopic improvement was observed in 2 patients. A comparison of the gut microbiome after 8 MSC treatments with that at baseline showed that the genus Cetobacterium was significantly enriched. Linoleic acid was depleted after 8 MSC treatments. A possible link between the altered Cetobacterium abundance and linoleic acid metabolite levels was observed in patients with CD who received MSCs. This study enabled an understanding of both the gut microbiota response and bacterial metabolites to obtain more information about host-gut microbiota metabolic interactions in the short-term response to MSC treatment.

Clinical symptoms and immune injury reflected by low CD4/CD8 ratio should increase the suspicion of HIV coinfection with tuberculosis.

Background: Patients who are coinfected with human immunodeficiency virus 1 (HIV) and Mycobacterium tuberculosis (TB) benefit from timely diagnosis and treatment. In the present study frequencies of CD3+, CD4+, and CD8+ T cells among peripheral blood mononuclear cells (PBMCs) of patients in the Kashi region of China infected with HIV, TB, and both HIV and TB (HIV-TB) were investigated to provide a basis for rapid identification of coinfected patients. Methods: A total of 62 patients with HIV, TB, or HIV-TB who were first hospitalized at our institution were included in the study, as were 30 controls. PBMCs were isolated, and the frequencies of CD3+, CD4+, and CD8+ T cells were determined via flow cytometry. Results: The frequency of CD4+ T cells and the CD4/CD8 ratio were significantly lower in the HIV-TB group than in the other three groups. In fever patients the frequency of CD4+ T cells and the CD4/CD8 ratio were significantly lower in the HIV-TB group than in the HIV group and the TB group. In patients who exhibited rapid weight loss there were no significant differences in the frequency of CD4+ T cells or the CD4/CD8 ratio between the groups. The results of treatment were compared in the HIV, TB, and HIV-TB groups after 7 days, and there were obvious improvements in the frequency of CD4+ T cells and the CD4/CD8 ratio. Conclusion: Clinical symptoms and the degree of immune injury can heighten suspicion for HIV-TB coinfection.

Key genes associated with non-alcoholic fatty liver disease and hepatocellular carcinoma with metabolic risk factors.

Background: Hepatocellular carcinoma (HCC) has become the world's primary cause of cancer death. Obesity, hyperglycemia, and dyslipidemia are all illnesses that are part of the metabolic syndrome. In recent years, this risk factor has become increasingly recognized as a contributing factor to HCC. Around the world, non-alcoholic fatty liver disease (NAFLD) is on the rise, especially in western countries. In the past, the exact pathogenesis of NAFLD that progressed to metabolic risk factors (MFRs)-associated HCC has not been fully understood. Methods: Two groups of the GEO dataset (including normal/NAFLD and HCC with MFRs) were used to analyze differential expression. Differentially expressed genes of HCC were verified by overlapping in TCGA. In addition, functional enrichment analysis, modular analysis, Receiver Operating Characteristic (ROC) analysis, LASSO analysis, and Genes with key survival characteristics were analyzed. Results: We identified six hub genes (FABP5, SCD, CCL20, AGPAT9(GPAT3), PLIN1, and IL1RN) that may be closely related to NAFLD and HCC with MFRs. We constructed survival and prognosis gene markers based on FABP5, CCL20, AGPAT9(GPAT3), PLIN1, and IL1RN.This gene signature has shown good diagnostic accuracy in both NAFLD and HCC and in predicting HCC overall survival rates. Conclusion: As a result of the findings of this study, there is some guiding significance for the diagnosis and treatment of liver disease associated with NAFLD progression.

Protective effect of astaxanthin on tuberculosis-associated inflammatory lung injury.

Mycobacterium tuberculosis (MTB) invades the lungs and is the key cause of tuberculosis (TB). MTB induces immune overreaction and inflammatory damage to lung tissue. There is a lack of protective drugs against pulmonary inflammatory damage. Herein, the protective roles and mechanisms of Astaxanthin (ASTA), a natural compound, in inflammatory injured lung epithelial cells were investigated. Lipopolysaccharide (LPS) was used to establish inflammatory injury model in the murine lung epithelial (MLE)-12 cells. Cell counting kit-8 was used for screening of compound concentrations. Cell proliferation was observed real-time with a high content analysis system. Flow cytometry assessed apoptosis. The changes of apoptotic proteins and key proteins in nuclear factor kappa-B (NF-κB) pathway were measured with the western blot. LPS was used to establish an animal model of pulmonary injury. The pathological changes and degree of inflammatory injury in lung tissue were observed with hematoxylin and eosin (HE) staining. The levels of inflammatory mediators were detected with enzyme-linked immunosorbent assay. The results showed that ASTA reduced lung inflammation and attenuated inflammatory damage in lung tissues. ASTA reduced apoptosis stimulated by LPS through suppressing the NF-κB pathway in MLE-12 cells. We believe that ASTA may have great potential for protection against inflammatory damage to lung tissue.

A symmetric prior knowledge based deep learning model for intracerebral hemorrhage lesion segmentation.

Background: Accurate localization and classification of intracerebral hemorrhage (ICH) lesions are of great significance for the treatment and prognosis of patients with ICH. The purpose of this study is to develop a symmetric prior knowledge based deep learning model to segment ICH lesions in computed tomography (CT). Methods: A novel symmetric Transformer network (Sym-TransNet) is designed to segment ICH lesions in CT images. A cohort of 1,157 patients diagnosed with ICH is established to train (n = 857), validate (n = 100), and test (n = 200) the Sym-TransNet. A healthy cohort of 200 subjects is added, establishing a test set with balanced positive and negative cases (n = 400), to further evaluate the accuracy, sensitivity, and specificity of the diagnosis of ICH. The segmentation results are obtained after data pre-processing and Sym-TransNet. The DICE coefficient is used to evaluate the similarity between the segmentation results and the segmentation gold standard. Furthermore, some recent deep learning methods are reproduced to compare with Sym-TransNet, and statistical analysis is performed to prove the statistical significance of the proposed method. Ablation experiments are conducted to prove that each component in Sym-TransNet could effectively improve the DICE coefficient of ICH lesions. Results: For the segmentation of ICH lesions, the DICE coefficient of Sym-TransNet is 0.716 ± 0.031 in the test set which contains 200 CT images of ICH. The DICE coefficients of five subtypes of ICH, including intraparenchymal hemorrhage (IPH), intraventricular hemorrhage (IVH), extradural hemorrhage (EDH), subdural hemorrhage (SDH), and subarachnoid hemorrhage (SAH), are 0.784 ± 0.039, 0.680 ± 0.049, 0.359 ± 0.186, 0.534 ± 0.455, and 0.337 ± 0.044, respectively. Statistical results show that the proposed Sym-TransNet can significantly improve the DICE coefficient of ICH lesions in most cases. In addition, the accuracy, sensitivity, and specificity of Sym-TransNet in the diagnosis of ICH in 400 CT images are 91.25%, 98.50%, and 84.00%, respectively. Conclusion: Compared with recent mainstream deep learning methods, the proposed Sym-TransNet can segment and identify different types of lesions from CT images of ICH patients more effectively. Moreover, the Sym-TransNet can diagnose ICH more stably and efficiently, which has clinical application prospects.

Deep learning based CT images automatic analysis model for active/non-active pulmonary tuberculosis differential diagnosis.

Active pulmonary tuberculosis (ATB), which is more infectious and has a higher mortality rate compared with non-active pulmonary tuberculosis (non-ATB), needs to be diagnosed accurately and timely to prevent the tuberculosis from spreading and causing deaths. However, traditional differential diagnosis methods of active pulmonary tuberculosis involve bacteriological testing, sputum culturing and radiological images reading, which is time consuming and labour intensive. Therefore, an artificial intelligence model for ATB differential diagnosis would offer great assistance in clinical practice. In this study, computer tomography (CT) scans images and corresponding clinical information of 1160 ATB patients and 1131 patients with non-ATB were collected and divided into training, validation, and testing sets. A 3-dimension (3D) Nested UNet model was utilized to delineate lung field regions in the CT images, and three different pre-trained deep learning models including 3D VGG-16, 3D EfficientNet and 3D ResNet-50 were used for classification and differential diagnosis task. We also collected an external testing set with 100 ATB cases and 100 Non-ATB cases for further validation of the model. In the internal and external testing set, the 3D ResNet-50 model outperformed other models, reaching an AUC of 0.961 and 0.946, respectively. The 3D ResNet-50 model reached even higher levels of diagnostic accuracy than experienced radiologists, while the CT images reading and diagnosing speed was 10 times faster than human experts. The model was also capable of visualizing clinician interpretable lung lesion regions important for differential diagnosis, making it a powerful tool assisting ATB diagnosis. In conclusion, we developed an auxiliary tool to differentiate active and non-active pulmonary tuberculosis, which would have broad prospects in the bedside.

Astaxanthin Prevents Tuberculosis-Associated Inflammatory Injury by Inhibiting the Caspase 4/11-Gasdermin-Pyroptosis Pathway.

Pyroptosis is a programmed cell death caused by inflammation. Multiple studies have suggested that Mycobacterium tuberculosis infection causes tissue pyroptosis. However, there are currently no protective drugs against the inflammatory damage caused by pyroptosis. In this study, anti-pyroptotic effects of the natural compound astaxanthin (ASTA) were explored in a simulated pulmonary tuberculosis-associated inflammatory environment. The results showed that ASTA maintained the stability of MLE-12 lung epithelial cell numbers in the inflammatory environment established by lipopolysaccharide. The reason is not to promote cell proliferation but to inhibit lipopolysaccharide-induced pyroptosis. The results showed that ASTA significantly inhibited the expression of key proteins in the caspase 4/11-gasdermin D pathway and the release of pyroptosis-related inflammatory mediators. Therefore, ASTA inhibits inflammation-induced pyroptosis by inhibiting the caspase 4/11-gasdermin D pathway and has the potential to protect lung tissue from tuberculosis-related inflammatory injury. ASTA, a functional food component, is a promising candidate for protection against tuberculosis-associated inflammatory lung injury.

Machine learning identifies prominent factors associated with cardiovascular disease: findings from two million adults in the Kashgar Prospective Cohort Study (KPCS).

BACKGROUND: Identifying factors associated with cardiovascular disease (CVD) is critical for its prevention, but this topic is scarcely investigated in Kashgar prefecture, Xinjiang, northwestern China. We thus explored the CVD epidemiology and identified prominent factors associated with CVD in this region. METHODS: A total of 1,887,710 adults at baseline (in 2017) of the Kashgar Prospective Cohort Study were included in the analysis. Sixteen candidate factors, including seven demographic factors, 4 lifestyle factors, and 5 clinical factors, were collected from a questionnaire and health examination records. CVD was defined according to International Clinical Diagnosis (ICD-10) codes. We first used logistic regression models to investigate the association between each of the candidate factors and CVD. Then, we employed 3 machine learning methods-Random Forest, Random Ferns, and Extreme Gradient Boosting-to rank and identify prominent factors associated with CVD. Stratification analyses by sex, ethnicity, education level, economic status, and residential setting were also performed to test the consistency of the ranking. RESULTS: The prevalence of CVD in Kashgar prefecture was 8.1%. All the 16 candidate factors were confirmed to be significantly associated with CVD (odds ratios ranged from 1.03 to 2.99, all p values < 0.05) in logistic regression models. Further machine learning-based analysis suggested that age, occupation, hypertension, exercise frequency, and dietary pattern were the five most prominent factors associated with CVD. The ranking of relative importance for prominent factors in stratification analyses showed that the factor importance generally followed the same pattern as that in the overall sample. CONCLUSIONS: CVD is a major public health concern in Kashgar prefecture. Age, occupation, hypertension, exercise frequency, and dietary pattern might be the prominent factors associated with CVD in this region.In the future, these factors should be given priority in preventing CVD in future.

Human umbilical cord-derived mesenchymal stem cells ameliorate experimental colitis by normalizing the gut microbiota.

BACKGROUND: Crohn's disease (CD) is a chronic non-specific inflammatory bowel disease. Current CD therapeutics cannot fundamentally change the natural course of CD. Therefore, it is of great significance to find new treatment strategies for CD. Preclinical and clinical studies have shown that mesenchymal stromal cells (MSCs) are a promising therapeutic approach. However, the mechanism by which MSCs alleviate CD and how MSCs affect gut microbes are still unclear and need further elucidation. METHODS: We used 2,4,6-trinitrobenzenesulfonic acid (TNBS) to induce experimental colitis in mice and analysed the microbiota in faecal samples from the control group, the TNBS group and the TNBS + MSC group with faecal 16S rDNA sequencing. Subsequent analyses of alpha and beta diversity were all performed based on the rarified data. PICRUStII analysis was performed on the 16S rRNA gene sequences to infer the gut microbiome functions. RESULTS: MSC Treatment improved TNBS-induced colitis by increasing survival rates and relieving symptoms. A distinct bacterial signature was found in the TNBS group that differed from the TNBS + MSC group and controls. MSCs prevented gut microbiota dysbiosis, including increasing α-diversity and the amount of Bacteroidetes Firmicutes and Tenericutes at the phylum level and decreasing the amount of Proteobacteria at the phylum level. MSCs alleviated the increased activities of sulphur and riboflavin metabolism. Meanwhile some metabolic pathways such as biosynthesis of amino acids lysine biosynthesis sphingolipid metabolism and secondary bile acid biosynthesis were decreased in the TNBS group compared with the control group and the TNBS + MSC group CONCLUSIONS: Overall, our findings preliminarily confirmed that colitis in mice is closely related to microbial and metabolic dysbiosis. MSC treatment could modulate the dysregulated metabolism pathways in mice with colitis, restoring the abnormal microbiota function to that of the normal control group. This study provides insight into specific intestinal microbiota and metabolism pathways linked with MSC treatment, suggesting a new approach to the treatment of CD.

Segmentation only uses sparse annotations: Unified weakly and semi-supervised learning in medical images.

Since segmentation labeling is usually time-consuming and annotating medical images requires professional expertise, it is laborious to obtain a large-scale, high-quality annotated segmentation dataset. We propose a novel weakly- and semi-supervised framework named SOUSA (Segmentation Only Uses Sparse Annotations), aiming at learning from a small set of sparse annotated data and a large amount of unlabeled data. The proposed framework contains a teacher model and a student model. The student model is weakly supervised by scribbles and a Geodesic distance map derived from scribbles. Meanwhile, a large amount of unlabeled data with various perturbations are fed to student and teacher models. The consistency of their output predictions is imposed by Mean Square Error (MSE) loss and a carefully designed Multi-angle Projection Reconstruction (MPR) loss. Extensive experiments are conducted to demonstrate the robustness and generalization ability of our proposed method. Results show that our method outperforms weakly- and semi-supervised state-of-the-art methods on multiple datasets. Furthermore, our method achieves a competitive performance with some fully supervised methods with dense annotation when the size of the dataset is limited.