OPTIMIZATION OF PREDICTION OF LUNG DISORDERS USING LSTM COMPARISON OF RMSPROP AND ADAM

Accurate prediction of pulmonary disorders is essential to support early diagnosis and clinical decision-making. Medical time-series data are inherently nonlinear and temporally dependent, making conventional statistical approaches insufficient. This study formulates pulmonary disorder prediction as a regression problem and proposes an optimized Long Short-Term Memory (LSTM) model by comparing two widely used optimization algorithms, RMSProp and Adam. The dataset consists of 30,000 clinical records obtained from an open-source Kaggle repository, including demographic, behavioral, and health-related variables relevant to respiratory conditions. Data preprocessing involved categorical encoding and Min–Max normalization, followed by an 80:20 train–test split. Model performance was evaluated using Mean Squared Error (MSE), Mean Absolute Error (MAE), and the coefficient of determination (R²). Experimental results demonstrate that the Adam optimizer achieves superior performance with lower prediction errors and more stable convergence compared to RMSProp and the baseline SGD optimizer. These findings highlight the critical role of optimizer selection in LSTM-based medical time-series modeling.