Nowadays, Machine Learning (ML) techniques play a very important role in health applications, where model transparency and real-time decision-making are paramount. In contrast to complex ML models, lightweight approaches offer a more efficient and resource-friendly solution, making them suitable for deployment on resource-constrained wearable devices for health applications. Moreover, the interpretability of models ensures that healthcare professionals can trust the ML models and comprehend them. Hence, it leads to foster collaborations in health applications (to diagnosis, treatment, and patient care). By considering both simplicity and interpretability, ML solutions hold great promise for improving the accessibility and efficiency of healthcare.