ASSESSING RISK FACTORS AFFECTING LUNGS CAPACITY USING STATISTICAL LEARNING MODELS

Abstract

Lung capacity is a key indicator of respiratory health and reflects the functional ability of the lungs to facilitate effective gas exchange. It is influenced by a variety of biological, lifestyle, and early-life factors. This study aims to assess the impact of age, height, gender, smoking status, and mode of delivery (caesarean versus vaginal birth) on lung capacity using both statistical and machine learning approaches. A publicly available lung capacity dataset obtained from Kaggle was analyzed using the R programming language. Multiple Linear Regression, Seventh-Degree Polynomial Regression, Artificial Neural Networks (ANN), and K-Nearest Neighbors (KNN) were applied to model linear and nonlinear relationships between lung capacity and the selected predictors. Model performance was evaluated using Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²).

The results indicate that lung capacity is significantly associated with age, height, gender, smoking status, and mode of delivery. Higher lung capacity was observed among taller, younger individuals, males, non-smokers, and those born via vaginal delivery. Among all models, ANN demonstrated the highest predictive accuracy, followed closely by polynomial and multiple linear regression models. These findings highlight the effectiveness of integrating traditional statistical methods with machine learning techniques for understanding and predicting lung capacity. The study provides valuable insights for clinical assessment, preventive strategies, and future respiratory health research.