A humanoid robot has accelerated vision restoration research by finding the best conditions to grow replacement retinal layers from human stem cells.
The AI system known as Maholo took just 185 days to complete experiments that would have taken humans two and a half years.
In just a quarter of the time, Maholo processed trial-and-error research with 200 million possible conditions.
The robot was created by a joint research group at the RIKEN Center for Biosystems Dynamic Research (BDR) in Kobe, Japan, to grow functional retinal pigment epithelial cells (RPE) from stem cells.
The degeneration of the RPE, a supporting cell layer that lies beneath our photoreceptors, is often seen in progressive disease that is the leading cause of blindness in the elderly.
RPE transplants have had some clinical success in the past.
The robot continuously repeated a series of exact movements and was able to evaluate the results to formulate the next experiment.
For every 100 stem cells, there were 50 RPE cells. These cells showed many of the biological markers that would make them suitable for transplantation.
The automation of life science research experiments that depend on a number of variables means that labor-intensive experiments that take months are avoided.
In the case of cell differentiation, the process by which stem cells are made from specific tissue, variables include finding the optimal type, dose, and timing of reagents.
Physical variables such as temperature, pipette strength and cell transfer time are extremely important “because minute differences in physical conditions have a significant impact on quality,” says team leader Genki Kanda.
The success of the new system extends beyond the researchers’ findings, as Kanda explained: “We chose to differentiate RPE cells from stem cells as a model, but in principle, combining a precision robot with the optimization algorithms will provide autonomous enabling trial and error experiments in many areas of the life sciences.”
Despite this, the research is not aimed at replacing humans with robots.
Kanda said, “Using robots and AI to conduct experiments will be of great interest to the public.”
“However, it is a mistake to see them as substitutes.”
“Our vision is that people do what they are good at, which is to be creative.”
“We can use robots and AI for the trial-and-error parts of experiments that require repeatable precision and time-consuming, but don’t require any thinking.”
The study was published on June 28 in the scientific journal eLife†
This story was provided to Newsweek by Zenger News†