Pan-mediastinal neoplasm diagnosis via nationwide federated learning: a multicentre cohort study.

BACKGROUND: Mediastinal neoplasms are typical thoracic diseases with increasing incidence in the general global population and can lead to poor prognosis. In clinical practice, the mediastinum's complex anatomic structures and intertype confusion among different mediastinal neoplasm pathologies severely hinder accurate diagnosis. To solve these difficulties, we organised a multicentre national collaboration on the basis of privacy-secured federated learning and developed CAIMEN, an efficient chest CT-based artificial intelligence (AI) mediastinal neoplasm diagnosis system. METHODS: In this multicentre cohort study, 7825 mediastinal neoplasm cases and 796 normal controls were collected from 24 centres in China to develop CAIMEN. We further enhanced CAIMEN with several novel algorithms in a multiview, knowledge-transferred, multilevel decision-making pattern. CAIMEN was tested by internal (929 cases at 15 centres), external (1216 cases at five centres and a real-world cohort of 11 162 cases), and human-AI (60 positive cases from four centres and radiologists from 15 institutions) test sets to evaluate its detection, segmentation, and classification performance. FINDINGS: In the external test experiments, the area under the receiver operating characteristic curve for detecting mediastinal neoplasms of CAIMEN was 0·973 (95% CI 0·969-0·977). In the real-world cohort, CAIMEN detected 13 false-negative cases confirmed by radiologists. The dice score for segmenting mediastinal neoplasms of CAIMEN was 0·765 (0·738-0·792). The mediastinal neoplasm classification top-1 and top-3 accuracy of CAIMEN were 0·523 (0·497-0·554) and 0·799 (0·778-0·822), respectively. In the human-AI test experiments, CAIMEN outperformed clinicians with top-1 and top-3 accuracy of 0·500 (0·383-0·633) and 0·800 (0·700-0·900), respectively. Meanwhile, with assistance from the computer aided diagnosis software based on CAIMEN, the 46 clinicians improved their average top-1 accuracy by 19·1% (0·345-0·411) and top-3 accuracy by 13·0% (0·545-0·616). INTERPRETATION: For mediastinal neoplasms, CAIMEN can produce high diagnostic accuracy and assist the diagnosis of human experts, showing its potential for clinical practice. FUNDING: National Key R&D Program of China, National Natural Science Foundation of China, and Beijing Natural Science Foundation.

Immunomodulatory Factor TIM3 of Cytolytic Active Genes Affected the Survival and Prognosis of Lung Adenocarcinoma Patients by Multi-Omics Analysis.

[Objective] Using multi-omics research methods to explore cytolytic activity-related genes through the immunoregulatory factors HAVCR2 (TIM3) affecting the survival and prognosis of lung adenocarcinoma. [Methods] We combined Cox single factor regression and lasso regression feature selection algorithm to screen out the key genes of cytolytic activity in lung adenocarcinoma, and applied multi-omics research to explore the clinical predictive value of the model, including onset risk, independent prognosis, clinical relevance, signal transduction pathways, drug sensitivity, and the correlation of immune regulatory factors, etc. TCGA data are used as the experimental group, and GEO data is used as the external data control group to verify the stability of the model. The survival curve was generated by the Kaplan-Meier method and compared by log-rank, and the Cox proportional hazard model was used for multivariate analysis. In this study, 10 fresh tissue samples of lung adenocarcinoma were collected for cellular immunohistochemical experiments to analyze the expression of immunoregulatory factors in cancer tissues, and the key immunoregulatory factors were verified and screened out. [Results] A total of 450 genes related to cytolytic activity were differentially expressed, of which 273 genes were up-regulated and 177 genes were down-regulated. A total of 91 key genes related to cytolytic activity related to the prognosis of lung adenocarcinoma were screened through Cox single factor regression. The ROC curve results showed that the AUC values of 1, 3, and 5 years in the training set and test set were all greater than 0.7, indicating that the model has a valid verification. The level of risk score is significantly related to the sensitivity of patients to AKT inhibitor VIII, Lenalidomide, and Tipifarnib. In addition, our study also found that receptor and MHC genes related to immunomodulatory, and chemokines, including HAVCR2, are more highly expressed in the low-risk group. [Conclusions] HAVCR2 (TIM3) immunoregulatory factors affect the expression of key genes that affect cytolytic activity in lung adenocarcinoma cells, and to some extent indirectly affect the survival and prognosis of patients with lung adenocarcinoma.

Selective poly adenylation predicts the efficacy of immunotherapy in patients with lung adenocarcinoma by multiple omics research.

The aim of this study was to find the application value of selective polyadenylation in immune cell infiltration, biological transcription function and risk assessment of survival and prognosis in lung adenocarcinoma (LUAD). The processed original mRNA expression data of LUAD were downloaded, and the expression profiles of 594 patient samples were collected. The (APA) events in TCGA-NA-SEQ data were evaluated by polyadenylation site use Index (PDUI) values, and the invasion of stromal cells and immune cells and tumor purity were calculated to group and select the differential genes. Lasso regression and stratified analysis were used to examine the role of risk scores in predicting patient outcomes. The study also used the GDSC database to predict the chemotherapeutic sensitivity of each tumor sample and used a regression method to obtain an IC50 estimate for each specific chemotherapeutic drug treatment. Then CIBERSORT algorithm was used to conduct Spearman correlation analysis, immune regulatory factor analysis and TIDE immune system function analysis for gene expression level and immune cell content. Finally, the Kaplan-Meier curve was used to analyze the correlation between stromal score and the immune score of LUAD. In this study, APA's LUAD risk score prognostic model was constructed. KM survival analysis showed that immune score affected the prognosis of LUAD patients ( P = 0.027) but the matrix score was not statistically significant ( P = 0.1). We extracted 108 genes with APA events from 827 different genes and based on PUDI clustering and heat map, the survival rate of patients in the four groups was significantly different ( P = 0.05). Multiple omics studies showed that risk score was significantly positively correlated with Macrophages M0, T cells Follicular helper, B cells naive and NK cells resting. It is significantly negatively correlated with dendritic cells resting, mast cells resting, monocyte, T cells CD4 memory resting and B cells memory. We further explored the relationship between the expression of immunosuppressor genes and risk score and found that ADORA2A, BTLA, CD160, CD244, CD274, CD96, CSF1R and CTLA4 genes were highly correlated with the risk score. Selective poly adenylation plays an important role in the development and progression of LUAD, immune invasion, tumor cell invasion and metastasis and biological transcription, and affects the survival and prognosis of LUAD patients.

Long-chain noncoding ribonucleic acids affect the survival and prognosis of patients with esophageal adenocarcinoma through the autophagy pathway: construction of a prognostic model.

Autophagy-related long-chain noncoding ribonucleic acids play a vital role in the development of esophageal adenocarcinoma. This study aimed to construct a prognostic model of autophagy-related long-chain noncoding ribonucleic acids and identify potential therapeutical targets for esophageal adenocarcinoma. We downloaded 261 long-chain noncoding RNA transcript samples and clinical data of 87 esophageal adenocarcinoma patients from the Cancer Genome Atlas and 307 autophagy-related genes from www.autophagy.com. We performed Kyoto Encyclopedia of Genes and Genomes and Gene Ontology enrichment analyses and Gene Set Enrichment Analysis to determine risk characteristics and bioinformatics functions of signal transduction pathways. Univariate and multivariate Cox regression analyses were used to determine the correlation between autophagy-related long-chain noncoding ribonucleic acids and independent risk factors. The receiver operating characteristic analysis was used to evaluate the feasibility of the prognostic model. Finally, we performed survival analysis, risk analysis and independent prognostic analysis to verify the prognostic model of esophageal adenocarcinoma. We identified 22 autophagic long-chain noncoding ribonucleic acids that were highly correlated with the overall survival of esophageal adenocarcinoma patients. The areas under the receiver operating characteristic curve (0.941) and the calibration curve were significantly similar. Moreover, univariate and multivariate Cox regression analyses indicated that autophagy-related long-chain noncoding ribonucleic acids were independent predictors of esophageal adenocarcinoma. We found that autophagy-related long-chain noncoding ribonucleic acids might affect tumor development and prognosis in esophageal adenocarcinoma patients. The findings indicate that the prognostic model of esophageal adenocarcinoma has potential therapeutic applications in patients with esophageal adenocarcinoma.

Comprehensive evaluation of the efficacy and safety of LPV/r drugs in the treatment of SARS and MERS to provide potential treatment options for COVID-19.

Coronavirus disease 2019 (COVID-19) experienced an outbreak that expanded worldwide. Lopinavir/ritonavir (LPV/r), which is used effectively for severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) coronavirus infections, was applied for COVID-19 treatment given similarities in the molecular structures of these viruses. We performed a systematic review and meta-analysis to evaluate the efficacy and safety of lopinavir/ritonavir antiviral treatment in patients with SARS, MERS, and COVID-19. After registration with INPLASY, a search was conducted in PubMed, Embase, China National Knowledge Infrastructure (CNKI), Cochrane Library, WanFang Data, China Biomedical Literature Database (CBM) and other databases for all relevant literature on lopinavir/ritonavir treatment of SARS, MERS and COVID-19. The Cochrane Collaboration's bias risk assessment tool and the Newcastle-Ottawa Scale (NOS) were used to evaluate the quality of the literature, and RevMan 5.3 software was used to evaluate the relevant outcome indicators of the efficacy and safety of lopinavir/ritonavir in the treatment of COVID-19. A total of 18 eligible studies (including randomized controlled studies, cohort studies, and case-control studies) were retrieved and included with a total of 2273 patients. The lopinavir/ritonavir group exhibited an increased nucleic acid conversion rate (P=0.004), higher virus clearance rate (P<0.0001), lower mortality rate (P=0.002), and reduced incidence of acute respiratory distress syndrome (ARDS) (P=0.02) compared with the control group. No significant benefit in the improvement rate of chest CT (P=0.08) or incidence of adverse events (P=0.45) was noted. The lopinavir/ritonavir group had a lower incidence of acute respiratory distress syndrome (P=0.02). According to the clinical prognostic results, the incidence of adverse events between the two groups was not statistically significant (P<0.0001). The efficacy of lopinavir/ritonavir in the treatment of patients with SARS, MERS and COVID-19 was significantly better than that of the control. Furthermore, the incidence of adverse events did not significantly increase. Lopinavir/ritonavir is effective in the treatment of COVID-19, and this combination should be further assessed in RCT studies. In addition, when we analyzed the differences in age and sex, we found that the differences were statistically significant in the safety and effectiveness of lopinavir/ritonavir in patients with COVID-19, and both of these factors played a significant role in the trial.

Reliability of information about the use of antiepileptic drugs during pregnancy from three major web search engines in China.

OBJECTIVES: The objective of this study was to assess the reliability of online information, as provided by three major search engines in China, about the usage of antiepileptic drugs (AEDs) during pregnancy. METHOD: Over eight weeks, six physicians conducted a literature search on six computers and six smartphones at a frequency of once per week. During each web search on each computer and smartphone, three major search engines in China were used, namely, Baidu, Sogou and 360. The search terms used were a combination of words, including one AED name (valproate/oxcarbazepine/levetiracetam/lamotrigine) and one Chinese word ("huaiyun" or "renshen", which means pregnancy in Chinese). The top ten websites retrieved from each search were recorded. After the content of each website was evaluated, the sites were categorized into 9 types. Meanwhile, commercial advertisements on each web page were also registered. RESULTS: A total of 16,411 search results were assessed. After excluding the redundant web pages, 4840 search results were included in the data analysis. Only 12.05% of the search results were reliable, 47.75% were partly reliable, and 40.21% were unreliable. A total of 4139 (85.52%) webpages contained commercial advertisements. The results from a multivariate analysis suggested that websites with no advertisements and professional websites have an independent positive impact on reliability. CONCLUSION: Overall, little information on AED usage during pregnancy provided by major search engines in China was reliable. PRACTICE IMPLICATIONS: Accurate and professional online information for female patients with epilepsy should be provided through major efforts by the government, search engine companies, professional websites and epilepsy physicians.