Introduction:
Due to long sea voyages, harsh environmental conditions, and limited medical access, seafarers face unique health challenges¹. Effective treatment and improved health outcomes can only be achieved through early diagnosis, despite the remote medical advice provided by Telemedical Maritime Assistance Services (TMAS)^2,3. Proactive health predictions and enhanced healthcare for maritime workers can be achieved using Machine Learning (ML) ⁴. The objective of this study is to investigate the application of machine learning techniques in the early diagnosis of diseases among seafarers by utilizing data from wearable devices, environmental sensors, and health records. The goal is to provide timely medical interventions and improve the health outcomes of seafarers.

Methods:
The study utilized physiological data from wearable sensors (e.g., heart rate, sleep patterns) along with anonymized health records. Three supervised ML models were adopted namely logistic regression, decision trees, and neural networks. To improve model accuracy and reduce overfitting, feature selection and data preprocessing techniques were employed. Cross-validation was applied to evaluate model performance.

Results:
The ML models accurately predicted the onset of common diseases among seafarers, including cardiovascular and respiratory diseases, as well as mental health conditions. Traditional statistical methods were used as a benchmark for comparison. Neural networks outperformed the other models with 89.6% accuracy, showing the highest accuracy in diagnosing diseases.

Conclusions:
ML has significant potential to transform healthcare in the maritime industry. By enabling early diagnosis and proactive health management, ML models can ensure timely medical intervention and improve the safety and well-being of seafarers. The integration of advanced data analytics into maritime health practices is recommended for sustainable health protection in the maritime sector.