We've build a prototype of method potentially feasible to become the basis for assistive apps for deaf-blind individuals. For instance in communication or spatial navigation with just a sensor screen of a smartphone and haptic feedback.

Wondering how to better approach real patients to test. And also curious how steep the learning curve would be.

Any thoughts and advices are highly appreciated.

https://www.kaggle.com/competitions/google-gemma-3n-hackathon/writeups/feelu-is-a-mobile-assistant-for-deaf-blind-individ

https://github.com/nullhtp/feelu