This paper explores an innovative approach to emotional design, that biological signal data is quantified and converted into music by integrating Artificial Intelligence (AI). Based on introduction of the theoretical foundations of emotion quantification and the techniques for biological signal collection, the design and implementation of fuzzy logic models is elaborated, capable of transforming physiological data into musical elements. Then the music is produced by using musical generative AI models. At the end, the application of the proposed route is validated through real-world cases and user evaluations. The results indicate that the design significantly alleviates stress, and improves emotional states. Finally, the potential applications and future research directions is discussed. This research provides a novel technological pathway for the field of emotional design and offers an innovative non-pharmacological intervention for mental health.