Researchers are looking at robotic skeletons to see if they can grow strong enough to transplant.

If the proof-of-concept can be improved with further research, tissue grown on humanoid robots could one day be used to repair torn muscles in a real person.

Today, surgeons can try to repair tendons, but the results are not always good. Scientists have been looking into alternative strategies for more than two decades.

It is difficult to grow new tendons from cells outside of the body. It is currently done in small bioreactor chambers, which mimic the conditions of a joint.

There is a soft transparent chamber. Fisher Studios.

Even the most cutting edge bioreactors fall short in mimicking the range and magnitude of movements expected of a tendon, despite evidence suggesting that dynamic movement like stretching and flexing is key in tendon development. The result may not live up to the task required.

The problem could be solved by the use of robots.

The concept has only been proved using a robotic shoulder joint. Human cells grown in a bioreactor chamber that could bend and extend with a robot grew faster than those grown in a static environment.

Different genes were expressed by the cells.

It makes sense that similar movements could help engineered versions grow.

Tendons connect human muscles to our bones, so they must be strong and flexible.

If they are deprived of stress, they will fail. The very composition of the tissue begins to fall apart.

The human shoulder's rotator cuff is busy. This joint has the greatest range of movement in the entire human body, which means that the tendons are hard at work keeping the ball and sockets in place.

The shoulder blade is connected to the arm's humerus by the supraspinatus muscle and tendon. It helps the arm flap from a person's side upwards.

Researchers were able to manipulate the cells that were grown in a flexible bioreactor and attached to a robotic shoulder by using a human body model.

The repeated motions of flapping, known as arm abduction and adduction, seemed to give the developing tissue some flexibility and strength.

The growth of the human cells was influenced by the force of the robotic movements.

The proof-of-concept suggests that flexible bioreactors, when attached to humanoid robots, are more realistic platforms on which to engineer tendons.

There is still a lot of testing that needs to be done. What bioreactor materials are best to use, what cell types respond to pushes and pulls, and which robotic movements are most useful to grow human tissue are some of the questions researchers want to investigate.

Possible long-term benefits from a humanoid bioreactor-based strategy include the production of functional tissue grafts for patients, the creation of an improved in vitro culture model for preclinical work and the opportunity to support the development of advanced robotic systems.

The study was published in a journal.