Aviation’s safety relies on pilots’ well-being, making health monitoring vital. Traditional health monitoring methods, including self-reported assessments, often fail to accurately capture emerging health problems, necessitating a more proactive and reliable solution. This study explores the use of deep learning (DL) models to predict pilots’ health alerts based on physiological data—heart rate (HR), heart rate variability (HRV), and respiratory rate (RR)—from both experienced and non-experienced pilots in flight simulations. A deep neural network (DNN) was developed, achieving 94.87% accuracy, 95.45% precision, 94.07% recall, and 99.28% AUC in classifying pilot health statuses. These results highlight the model’s potential to identify at-risk pilots, enabling timely interventions and advancing proactive health monitoring, ultimately enhancing aviation safety and operational efficiency.