Early diagnosis and prevention play a crucial role in the treatment of patients with ARDS. The definition of ARDS requires an arterial blood gas to define the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen (PaO 2 /FiO 2 ratio). However, many patients with ARDS do not have a blood gas measured, which may result in under-diagnosis of the condition. Using data from MIMIC-III Database, we propose an algorithm based on patient non-invasive physiological parameters to estimate P/F levels to aid in the diagnosis of ARDS disease. The machine learning algorithm was combined with the filter feature selection method to study the correlation of various noninvasive parameters from patients to identify the ARDS disease. Cross-validation techniques are used to verify the performance of algorithms for different feature subsets. XGBoost using the optimal feature subset had the best performance of ARDS identification with the sensitivity of 84.03%, the specificity of 87.75% and the AUC of 0.9128. For the four machine learning algorithms, reducing a certain number of features, AUC can still above 0.8. Compared to Rice Linear Model, this method has the advantages of high reliability and continually monitoring the development of patients with ARDS.

No abstract

ObjectiveTo screen lung adenocarcinoma (LUAC)-specific cell-cycle-related genes (CCRGs) and develop a prognostic signature for patients with LUAC.MethodsThe GSE68465, GSE42127, and GSE30219 data sets were downloaded from the GEO database. Single-sample gene set enrichment analysis was used to calculate the cell cycle enrichment of each sample in GSE68465 to identify CCRGs in LUAC. The differential CCRGs compared with LUAC data from The Cancer Genome Atlas were determined. The genetic data from GSE68465 were divided into an internal training group and a test group at a ratio of 1:1, and GSE42127 and GSE30219 were defined as external test groups. In addition, we combined LASSO (least absolute shrinkage and selection operator) and Cox regression analysis with the clinical information of the internal training group to construct a CCRG risk scoring model. Samples were divided into high- and low-risk groups according to the resulting risk values, and internal and external test sets were used to prove the validity of the signature. A nomogram evaluation model was used to predict prognosis. The CPTAC and HPA databases were chosen to verify the protein expression of CCRGs.ResultsWe identified 10 LUAC-specific CCRGs (PKMYT1, ETF1, ECT2, BUB1B, RECQL4, TFRC, COCH, TUBB2B, PITX1, and CDC6) and constructed a model using the internal training group. Based on this model, LUAC patients were divided into high- and low-risk groups for further validation. Time-dependent receiver operating characteristic and Cox regression analyses suggested that the signature could precisely predict the prognosis of LUAC patients. Results obtained with CPTAC, HPA, and IHC supported significant dysregulation of these CCRGs in LUAC tissues.ConclusionThis prognostic prediction signature based on CCRGs could help to evaluate the prognosis of LUAC patients. The 10 LUAC-specific CCRGs could be used as prognostic markers of LUAC.

No abstract

No abstract