MIT Technology Review is reporting that strokes leave 5 million people with long-term impairments each year, and a growing shortage of physical therapists makes it hard for them to get the care they need. It said a new approach tries to solve that problem by combining artificial intelligence and human therapy.

The outlet said Johannes Lachner, a former MIT postdoc who is now an assistant professor at Purdue University, and Noah Geiger, a former visiting student at MIT, developed a robotic system that uses generative AI and real-time force feedback to learn from real therapists. Drawing on what it has learned, the system gives patients just the right amount of assistance during physical rehabilitation exercises such as reaching, lifting their arms, and performing tasks of daily living.

The goal is "not to replace therapists, but to extend their reach," Lachner said, according to MIT Technology Review.

Lachner said initial robotic experiments have already demonstrated the feasibility of the approach. He said the next phase of the project aims to develop therapist-specific models and evaluate them in a long-term clinical study with the same patients who previously received manual therapy. Beyond that, he envisions possible applications in contact-rich industrial settings or collaborative workspaces.

MIT Technology Review also highlighted other recent coverage, including a piece arguing that large language models do not reason, a critique of this summer's AI hype, a report on an AI "mind-reading" tool that can reconstruct what a person is looking at from a brain scan, and a report on why OpenAI agents hacked Hugging Face.

Full Article: Robotic therapy for stroke patients