Background: Surgical resection is the standard treatment for early-stage lung cancer. Since postoperative lung function is related to mortality, predicted postoperative lung function is used to determine the treatment modality. The aim of this study was to evaluate the predictive performance of linear regression and machine learning models.Methods: We extracted data from the Clinical Data Warehouse and developed three sets: set I, the linear regression model; set II, machine learning models omitting the missing data: and set III, machine learning models imputing the missing data. Six machine learning models, the least absolute shrinkage and selection operator (LASSO), Ridge regression, ElasticNet, Random Forest, eXtreme gradient boosting (XGBoost), and the light gradient boosting machine (LightGBM) were implemented. The forced expiratory volume in 1 second measured 6 months after surgery was defined as the outcome. Five-fold cross-validation was performed for hyperparameter tuning of the machine learning models. The dataset was split into training and test datasets at a 70:30 ratio. Implementation was done after dataset splitting in set III. Predictive performance was evaluated by R2 and mean squared error (MSE) in the three sets.Results: A total of 1,487 patients were included in sets I and III and 896 patients were included in set II. In set I, the R2 value was 0.27 and in set II, LightGBM was the best model with the highest R2 value of 0.5 and the lowest MSE of 154.95. In set III, LightGBM was the best model with the highest R2 value of 0.56 and the lowest MSE of 174.07.Conclusion: The LightGBM model showed the best performance in predicting postoperative lung function.

Kwon OB, et al. J Korean Med Sci. 2025 Jan;40:e73. https://doi.org/10.3346/jkms.2025.40.e73

Chronic obstructive pulmonary disease (COPD) is one of the most frequently occurring concomitant diseases in patients with non-small cell lung cancer (NSCLC). It is characterized by small airways and the hyperinflation of the lung. Patients with hyperinflated lung tend to have more reserved lung function than conventionally predicted after lung cancer surgery. The aim of this study was to identify other indicators in predicting postoperative lung function after lung resection for lung cancer. Patients with NSCLC who underwent curative lobectomy with mediastinal lymph node dissection from 2017 to 2019 were included. Predicted postoperative FEV₁ (ppoFEV₁) was calculated using the formula: preoperative FEV₁ × (19 segments-the number of segments to be removed) ÷ 19. The difference between the measured postoperative FEV₁ and ppoFEV₁ was defined as an outcome. Patients were categorized into two groups: preserved FEV₁ if the difference was positive and non-preserved FEV₁, if otherwise. In total, 238 patients were included: 74 (31.1%) in the FEV₁ non-preserved group and 164 (68.9%) in the FEV₁ preserved group. The proportion of preoperative residual volume (RV)/total lung capacity (TLC) ≥ 40% in the FEV₁ non-preserved group (21.4%) was lower than in the preserved group (36.1%) (p = 0.03). In logistic regression analysis, preoperative RV/TLC ≥ 40% was related to postoperative FEV1 preservation. (adjusted OR, 2.02, p = 0.041). Linear regression analysis suggested that preoperative RV/TLC was positively correlated with a significant difference. (p = 0.004) Preoperative RV/TLC ≥ 40% was an independent predictor of preserved lung function in patients undergoing curative lobectomy with mediastinal lymph node dissection. Preoperative RV/TLC is positively correlated with postoperative lung function.

There have been previous studies conducted to predict postoperative lung function with pulmonary function tests (PFTs). Computing tomography (CT) can quantitatively measure small airway walls’ thickness, lung volume, pulmonary vessel volume, and emphysema area, which reflect the severity of respiratory diseases. These measurements are considered imaging biomarkers. This study aimed to predict postoperative lung function with imaging biomarkers. A retrospective analysis of 79 patients with lung cancer who had undergone lung surgery was completed. Postoperative lung function measured by forced expiratory volume in one second (FEV₁) was defined as an outcome. Preoperative clinico-pathological parameters and imaging biomarkers representing airway walls’ thickness, severity of emphysema, total lung volume, and pulmonary vessel volume were measured quantitatively in chest CT by an automated segmentation software, AVIEW COPD. Pi1 was defined as the first percentile along the histogram of lung attenuation that represents the degree of emphysema. Wafw was defined as the airway thickness, which was calculated by the full-width at half-maximum method. Logistic and linear regressions were used to assess these variables. If the actual postoperative FEV₁ was higher than the postoperative FEV₁ projected by a formula, the group was considered to be preserved. Among the 79 patients, 16 of the patients were grouped as a non-preserved group, and 63 of them were grouped as a preserved group. The patients in the preserved FEV₁ group had a higher vessel volume than the non-preserved group. Pi1 and Wafw were independent predictors of postoperative lung function. Imaging biomarkers can be considered significant variables in predicting postoperative lung function in patients with lung cancer.

Background: Accurate confirmation of nasogastric (NG) tubes is essential for patient safety, but delays and variability in interpretation remain common in clinical practice. Deep learning (DL) models have shown potential for assisting in this task, but real-world performance, particularly in detecting malpositioned tubes, remains insufficiently characterized. Methods: We conducted a pilot evaluation of a previously developed DL model using 135 chest radiographs from Kangwon National University Hospital. Expert physicians established the reference standard. Model performance was assessed and receiver operating characteristic (ROC) curve and precision recall curve (PRC) analyses were performed. Differences between correctly classified and misclassified cases were examined using Wilcoxon rank-sum and Fisher’s exact tests to explore potential clinical or radiographic contributors to model failure. Results: The model correctly classified 129 of 135 cases. The sensitivity was 96.1% (95% confidence interval (CI): 92.2–98.9%), specificity was 85.7% (95% CI: 42.2–97.7%), positive predictive value (PPV) was 99.2% (95% CI: 96.1–99.9%), negative predictive value (NPV) was 54.5% (95% CI: 25.4–80.8%), balanced accuracy was 90.8%, and F1-score was 0.976. The area under the ROC curve was 0.970 (95% CI: 0.929–1.000) and that under the PRC was 0.727 (95% CI: 0.289–1.000), reflecting substantial uncertainty related to the very small number of incomplete cases (n = 6). No statistically significant differences in clinical or radiographic characteristics were observed between correctly classified and misclassified cases. Conclusions: The DL model performed well in identifying correctly positioned NG tubes but demonstrated limited and unstable performance for detecting incomplete placements. Given the safety implications of misclassification, the model should be used only as an assistive tool with mandatory physician oversight. Larger, multi-center studies with greater representation of incomplete cases are required to obtain more reliable estimates and support safe clinical implementation.

Smoking remains a primary cause of cancers, cardiovascular and respiratory diseases and death. Globally, efforts have been made to reduce smoking rates, but the addictive nature of nicotine, a key component of tobacco, makes cessation challenging for smokers. Medical interventions including medical advice and pharmacotherapies are effective methods for smoking cessation. The frequency of medical interventions correlates with success in smoking cessation. This study aims to compare the characteristics of the patients who visited the smoking cessation clinic once with those who visited more than once, in order to identify factors that are associated with repeat clinic visits. A total of 81 patients who have visited the smoking cessation clinic in Kangwon National University Hospital were included. Patients answered the questionnaire at their first visit. If the patient visited only once, the outcome was defined as negative and if the patient visited more than once, the outcome was defined as positive. The proportion of patients who answered “within 5 min” to the Fagerstrom Test for Nicotine Dependence’s (FTND) 1st question and answered “yes” to the FTND’s 6th question was higher in the negative outcome group. In the logistic regression, patients who had withdrawal symptoms previously were associated with positive outcomes (adjusted OR 3.466, 95% CI 1.088–11.034 and p value = 0.0354). Withdrawal symptoms during previous attempts were positively related to visiting the clinic more than once.