Predicting pulmonary function using deep-learning-generated synthetic data

• Using MLR and GLM, chest radiography score, albumin level, and Brinkman index were identified as significant predictors, underlining the value of incorporating clinical data beyond age, sex, and height. • Neural network models were developed to predict pulmonary function (FEV₁ and FVC) using an expanded training dataset generated by CTGAN. • Neural network models demonstrated better performance than traditional linear models (MLR and GLM) for predicting FVC in all patients and FEV₁ in males. • This approach may provide more accurate pulmonary function estimates than the traditional reference equations in patients with complex or severe respiratory diseases. Pulmonary function test (PFT) is used to assess the risk of postoperative complications. However, in patients requiring emergency surgery, PFT is often not feasible. This study aimed to develop useful predictive models using synthetic data by Conditional Tabular Generative Adversarial Network (CTGAN). Data were collected from patients who visited the respiratory medicine departments. Model performance was evaluated using R 2 , mean absolute error (MAE), and root mean square error (RMSE). A total of 930 patient datasets were collected from two medium-sized hospitals. Prediction models for forced vital capacity (FVC) and forced expiratory volume in forced expiratory volume in 1 second (FEV₁) were constructed using multiple linear regression (MLR) and generalized linear models (GLM). After generating 100,000 synthetic datasets, neural network models (NNM) were developed. The prediction accuracy for FVC was higher than that of MLR in both males and females (R² = 0.63 for NNM, 0.53 for MLR in males, and 0.62 for MLR in females), whereas for FEV₁, the prediction accuracy was higher in males but lower in females (R² = 0.58 for NNM, 0.50 for MLR in males, and 0.65 for MLR in females). The prediction accuracy for FEV₁ was lower in females but higher in males (R² = 0.58 for NNM, 0.50 for MLR in males, and 0.65 for MLR in females). NNM may be especially more effective than linear regression for patients with more severe and complex respiratory diseases and may therefore be useful for perioperative surgical risk assessment in such patients.