Background:This present work focused on predicting prognostic outcomes of inpatients developing acute exacerbation of chronic obstructive pulmonary disease (AECOPD), and enhancing patient monitoring and treatment by using objective clinical indicators. Methods:The present retrospective study enrolled 322 AECOPD patients. Registry data downloaded based on the chronic obstructive pulmonary disease (COPD) Pay-for-Performance Program database from January 2012 to December 2018 were used to check whether the enrolled patients were eligible. Our primary and secondary outcomes were intensive care unit (ICU) admission and in-hospital mortality, respectively. The best feature subset was chosen by recursive feature elimination. Moreover, 7 machine learning (ML) models were trained for forecasting ICU admission among AECOPD patients, and the model with the most excellent performance was used. Results:According to our findings, a random forest (RF) model showed superb discrimination performance, and the values of area under the receiver operating characteristic curve were 0.973 and 0.828 in training and test cohorts, separately. Additionally, according to decision curve analysis, the net benefit of the RF model was higher when differentiating patients with a high risk of ICU admission at a <0.55 threshold probability. Moreover, the ML-based prediction model was also constructed to predict in-hospital mortality, and it showed excellent calibration and discrimination capacities. Conclusion:The ML model was highly accurate in assessing the ICU admission and in-hospital mortality risk for AECOPD cases. Maintenance of model interpretability helped effectively provide accurate and lucid risk prediction of different individuals.

In this study, we investigated whether COL4A3 mRNA expression levels were associated with clinical outcomes after treatment with a combination of gemcitabine (Gem)/cis-diamminedichloroplatinum(II) (CDDP) regimen for patients with advanced stage non-small cell lung cancer (NSCLC). Response and survival were correlated with the level of COL4A3 expression in 58 patients with advanced (stage IIIb or IV) NSCLC treated as part of a multicenter randomized trial with Gem 1,250 mg/m2 on days 1 and 8 plus CDDP 100 mg/m2 on day 1 every 3 weeks. mRNA was isolated from paraffin-embedded pretreatment primary tumor specimens, and relative expression levels of COL4A3/β-actin were measured using a quantitative reverse transcription-PCR (Taqman) system. COL4A3 expression was detectable in all tumors. There were no significant differences in COL4A3 levels by gender, age, performance status, weight loss, or tumor stage. The overall response rate was 45.8 %. There were no significant associations between COL4A3 expression and response. Median overall survival was significantly longer in patients with low COL4A3 expression tumors compared to patients with high expression tumors. COL4A3 expression, Eastern Cooperative Oncology Group performance status, and presence of weight loss were significant prognostic factors for survival in a Cox proportional hazards multivariable analysis. These data suggest that COL4A3 expression is a predictive factor for survival after CDDP/Gem therapy in advanced NSCLC. Although there was a trend toward decreased response with high COL4A3 mRNA levels, this difference failed to reach statistical significance. This result may reflect the impact of Gem and the requirement for COL4A3 expression for CDDP/Gem synergism or may be attributable to the relatively small patient sample size in this study. Prospective studies of COL4A3 as a predictive marker for activity of CDDP-based regimens in NSCLC are warranted.