Fig. 1

AI-based mapping platform for optimizing the neurostimulation parameters. The subject is interacting with the user interface, simultaneously perceiving neurostimulation selected by the RL algorithm, that is eliciting the electro-touch. The system consists of three parts: the AI brain (A), the neurostimulation unit (B) and a user interface (C). A The AI model is an iterative RL machine which initializes and updates the neurostimulation parameters sent to the stimulator. B The stimulator receives the parameters and stimulates each of 3 channels accordingly, through a pair of superficial electrodes placed on the skin of the subject in correspondence of the specific nerve. C When the stimulation ends, the subject can describe the perceived sensation through comprehensive questionnaires, which include the perceived intensity, type and location of sensation and the intensity of the sensation perceived under the electrodes. The subject’s answers are sent to the AI which: can either finish the characterization, if the desired sensation has been reached or update the neurostimulation parameters and repeat described steps to optimize the sensation