Scientists have created tiny biological "robots" from lung cells that they hope could one day travel around our bodies, regenerating damaged tissues and treating disease.

The robots, called anthrobots, range in size from the width of a human hair to the point of a sharpened pencil and can self-assemble, TuftsNow reported.

They were created by a team of scientists from Tufts and Harvard universities who altered the chemical composition of anonymously-donated lung cells.

The scientists discovered that when they made cilia – tiny hair-like structures on the surface of cells – face outwards, they began moving. And when a cluster of anthrobots, known as a "superbot," was placed with damaged neuron cells it was able to encourage new growth.

They don't yet know how the cell regeneration occurs, but believe that the anthrobots could once day be used to clear blocked arteries, repair spinal cords, identify cancer cells, or deliver drugs to specific parts of the body, per TuftsNow.

Their research, published Thursday in the journal Advanced Science, raises new questions about how human cells assemble and work together.

The same research team had previously created similar tiny robots, or xenobots, from cells sourced from embryos of the African clawed frog. But the anthrobots prove that cells don't have to be amphibian or embryonic to regenerate.

Coauthor Michael Levin told CNN: "I don't think this has anything to do with being an embryo. This has nothing to do with being a frog. I think this is a much more general property of living things."

Alhough they're created from human cells, Levin said the anthrobots didn't have a full life cycle and so aren't considered as fully-fledged organisms.

"It reminds us that these harsh binary categories that we've operated with – is that a robot, is that an animal, is that a machine? – don't serve us very well. We need to get beyond that," Levin told CNN.

While doctors may not be using anthrobots in the near future, they're another example of advancements in medical technology.

The US Food and Drug Administration has already approved more than 500 medical AI algorithms that are helping to diagnose and treat patients.