Personalized Deformable Masks: A Workflow for AI-Based Landmark Filtering and Morphing Material Integration in Health Monitoring

This thesis presents an ai-driven workflow for designing personalized deformable masks that dynamically adapt to individual facial geometries. Addressing key limitations in existing health-monitoring wearables—specifically, poor facial conformity, limited portability, and lack of user-specific customization—the system leverages AI-based facial reconstruction and morphing material integration to create masks that are both functional and responsive. By selectively filtering facial landmarks and focusing only on mask-relevant regions, the method significantly reduces data complexity and speeds up the design-to-fabrication cycle. The proposed workflow prioritizes accessibility, using consumer-grade tools to enable rapid customization without requiring specialized scanning hardware. While formal technical evaluations remain future work, the workflow’s comparative efficiency is demonstrated through reduced modeling steps, material usage, and hands-on fabrication time. This research contributes a scalable, low-cost pipeline for rapid custom mask production, with potential applications in healthcare, personal protective equipment, and responsive wearable systems.