Showing posts with label Ambient Intelligence. Show all posts
Showing posts with label Ambient Intelligence. Show all posts

14 July 2025

Robot Removes Pig Gallbladders

Automated surgery could be trialled on humans within a decade, after an AI-trained robot armed with tools to cut, clip and grab soft tissue successfully removed pig gall bladders without human help. The robot surgeons were schooled on video footage of human medics conducting operations using organs taken from dead pigs. Eight operations were conducted on pig organs with a 100% success rate by a team led by experts at Johns Hopkins University in Baltimore in the US. The technology allowing robots to handle complex soft tissues such as gallbladders, which release bile to aid digestion, is rooted in the same type of computerised neural networks that underpin widely used AI tools such as Chat GPT or Google Gemini.

A machine in a room

AI-generated content may be incorrect.

The surgical robots were slightly slower than human doctors, but they were less jerky and plotted shorter trajectories between tasks. The robots were also able to repeatedly correct mistakes as they went along, asked for different tools and adapted to anatomical variation, according to a peer-reviewed paper published in the journal Science Robotics. In the Johns Hopkins trial, the robots took just over five minutes to carry out the operation, which required 17 steps including cutting the gallbladder away from its connection to the liver, applying six clips in a specific order and removing the organ. The robots on average corrected course without any human help six times in each operation.

More information:

https://www.theguardian.com/science/2025/jul/09/robot-surgery-on-humans-could-be-trialled-within-decade-after-success-on-pig-organs

25 April 2025

Robots Learn by Watching Videos

Cornell researchers have developed a new robotic framework powered by AI, called RHyME (Retrieval for Hybrid Imitation under Mismatched Execution). It allows robots to learn tasks by watching a single how-to video. RHyME could fast-track the development and deployment of robotic systems by significantly reducing the time, energy and money.

RHyME paves the way for robots to learn multiple-step sequences while significantly lowering the amount of robot data needed for training, the researchers said. RHyME requires just 30 minutes of robot data; in a lab setting, robots trained using the system achieved a more than 50% increase in task success compared to previous methods.

More information:

https://news.cornell.edu/stories/2025/04/robot-see-robot-do-system-learns-after-watching-how-tos