{"id":14029,"date":"2023-10-05T17:48:38","date_gmt":"2023-10-05T15:48:38","guid":{"rendered":"https:\/\/sano.science\/?post_type=research&p=14029"},"modified":"2024-01-05T13:36:40","modified_gmt":"2024-01-05T12:36:40","slug":"robust-simulation-of-internal-body-organs","status":"publish","type":"research","link":"https:\/\/sano.science\/research\/robust-simulation-of-internal-body-organs\/","title":{"rendered":"Robust simulation of internal body organs"},"content":{"rendered":"\n

Micha\u0142 Naskr\u0119t<\/h2>\n\n\n\n
<\/div>\n\n\n\n

A virtual representation of body organs can be useful in a number of ways. One of the most innovative is to recreate a real-life scenario environment which can be a playground for Artificial Intelligence algorithms. This is especially convenient in the context of surgical procedures, where it is impossible to create a real environment in which an algorithm could try to perform millions of actions using for example a collaborative robot arm. Here, our goal is to leverage the possibilities of parallel computations using Graphics Processing Units (GPUs) and Differentiable Physics to boost the performance of Reinforcement Learning (RL) algorithms.<\/p>\n\n\n\n

This will hopefully allow us to try to transfer the knowledge obtained by AI in virtual environments into true environments with more difficult scenarios.<\/p>\n\n\n\n

The project is conducted in collaboration with teams from University of Verona and Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\n