top of page

INTERVIEW

Case Study Interview

visioncare_article.png

Searching for cell recipes for regenerative medicine using AI. Achieving efficiency equal to or greater than that of an expert.

Background

  • Because iPS cell differentiation induction culture is a complicated process that takes more than a month, the know-how and techniques tend to be concentrated in the hands of a few experienced workers. In addition, because it takes time, it is difficult to gain experience, and this makes it difficult to pass on the skills.

  • In order to conduct large-scale clinical studies and ultimately deliver treatment to many patients, cell culture techniques needed to be scaled up so that more people could learn them.

Effects after implementation

  • We discovered a method for inducing iPS cell differentiation for robots that achieves differentiation efficiency equal to or greater than that of an experienced technician (accepted by an international journal).

  • The results have also been applied to cell manufacturing processes.

Masayo Takahashi was the first in the world to successfully perform a retinal transplant using iPS cells in September 2014. She is currently the CEO of Vision Care Co., Ltd. and continues to work to bring cutting-edge treatments to many patients. We spoke to her about the background to her collaborative research using Epistra Accelerate and her future prospects.

20240510-005 (1).jpg

iPS cell culture is difficult and relies on the skills of a select few experienced personnel

- First of all, please tell us about the background and challenges you faced when starting this joint research project.

We are working on developing treatments for intractable eye diseases using iPS cells. To create cells suitable for treatment, it is important to culture iPS cells in an undifferentiated state while multiplying them and inducing their differentiation into the desired cells (in this case, retinal pigment epithelial cells). However, this is an extremely difficult task, and until now only a small number of skilled workers have been able to consistently produce cells of the highest quality.

On the other hand, in order to conduct larger-scale clinical research and ultimately deliver treatment to many patients, it is not enough for only a few skilled individuals to be able to culture cells; cell culture techniques need to be scaled up so that many people can learn them.

Of course, there are proper manuals, and by gaining experience under the guidance of experts, the technique gradually improves. However, cell culture takes time (induction of differentiation of retinal pigment epithelium takes about 40 days), so it also takes time to gain experience, and people often end up doing things their own way without realizing it. Another issue is that it is difficult to convey the tips that experts use unconsciously.

20240510-042 (1).jpg

The best team of AI x robots x biology will tackle the bottlenecks of regenerative medicine

- Because it is a complicated process, the know-how and skills tend to be concentrated in a small number of skilled workers, and because it takes time, it is difficult to gain experience and pass on the skills. What made you decide to embark on joint research with Epistola?

The direct opportunity came at the beginning of 2018, when Mr. Ozawa from Epistula, Mr. Koichi Takahashi from RIKEN, and Mr. Toru Natsume from AIST proposed a project to combine robots and AI to culture iPS cells.

I had the opportunity to see the robot (Mahoro) originally developed by Natsume and his team when I visited AIST in 2015, and from that time I had the intuition that the robot's ability to accurately execute the same procedures over and over again made it suitable for manufacturing.

However, to actually have a robot manufacture the cells, it is necessary to have a system that systematically extracts the key points of differentiation induction from the mind of an expert. When I first heard the proposal, I thought there was a lack of understanding of how to incorporate these points.

So we all got together and decided to combine AI and robots to extract the key points of differentiation induction methods and create a differentiation induction method that surpasses that of experts. Retinal pigment epithelial cells, which contain pigment, are easy to judge whether a differentiation induction method is good or bad, so it was a very good practice problem, and we thought it was definitely worth the challenge.

Above all, it would be cool if we could make it with a robot (laughs).

20240510-063 (1).jpg

Right: Masayo Takahashi, President and CEO of Vision Care Co., Ltd.

Left: Yosuke Ozawa, Ph.D., co-founder and CEO of Epistra Inc.

-In 2018, there were still very few examples of autonomous exploration combining robots and AI (A mobile robot chemist was introduced in 2020), and Epistra had only just been founded and had no track record, so did you have any concerns?

Of course, I was a little worried, but my expectations were far greater than my worries.

In basic research, you first need the depth and expertise to delve into a specific theme. However, in order to use that as a base to launch a new industry, it is not enough to have a single, deep expertise; you need to have a broad perspective and be able to creatively solve the problems that arise one after another.

At this time, too, I had a hunch that if we could use the best recipe found by AI to accurately culture cells using robots, it might be possible to eliminate the various bottlenecks that stand in the way of widespread use of regenerative medicine, such as issues with personal skills and production management, and make it possible to deliver cutting-edge treatments to more people.

In addition, one of the conditions for taking on a new challenge that I learned during my time at the Salk Institute is to assemble the best people in each field. With regard to the members of this team, I was able to believe that we were bringing together the highest level of talent, which was a big boost for me.

Once the team's gears meshed, we were able to achieve results that easily surpassed previous limits.

-How did you feel when you actually started the collaborative research?

I remember it was very exciting to see AI and robotics experts gathering and discussing in my lab, which specializes in regenerative medicine. They all gathered around the expert culturing expert in my lab, asking him everything in detail about what he was looking at and what he was thinking. Then, seeing how he put together various new hypotheses and methods based on that information, I felt a definite sense of accomplishment in transforming biology into biotechnology.

-Was there anything that surprised you?

I thought that the differentiation induction method we had been using was quite sophisticated, and therefore thought that it would be very difficult to improve it any further, so I was surprised when the AI (and the robot) actually found conditions that easily surpassed it.

Moreover, excluding the lead-up period for making contracts and understanding each other, this was achieved in just six months, which made me realize that the combination of AI and robots creates a wonderful synergy. AI can come up with better conditions based on data, and robots can accurately repeat specified actions and conditions as many times as necessary. The strengths of these two teams have been combined beautifully.

20240510-015 (1).jpg

This technology has great potential to transform biology, and its application in regenerative medicine is just one example.

- Were there any results other than what you originally expected?

Our initial goal was to improve differentiation induction methods, but as we continued our research, we realized that this system should be used for cell production itself. Cell production is very demanding work even for experienced workers. However, using this system, we could achieve high efficiency and mass production by using robots, which also improved the stability.

Also, having experienced collaborative research using AI once gave me many new perspectives. I began to think that maybe I could solve that problem in my project, or maybe AI could be used here. I feel extremely lucky to have had such an initial experience with top-level people.

-Please tell us about your future plans.

As I mentioned earlier, this result is currently being put to use in cell manufacturing for regenerative medicine. However, I believe that revolution brought about by "AI" or "the combination of AI and robots" is not limited to regenerative medicine.

I believe that this technology has the potential to advance biology research and development at an unprecedented speed, and I personally would like to utilize this technology to witness the evolution of biology.

20240510-042 (1).jpg
VC.png

Vision Care Co., Ltd.

President and CEO

Ms. Masayo Takahashi

image 10.png

Recommend

Other recommended articles

spiber_article.png

Research and development that would have taken two years shortened to just a few weeks: AI discovers conditions for high production of structural proteins in a short period of time

Spiber Inc. is a bio venture that develops and produces structural protein "Brewed Protein™ material." Brewed Protein™ fiber is environmentally degradable and is expected to be a next-generation material that can provide an alternative solution to conventional animal-derived, plant-derived, and synthetic materials. 71 products are sold by 16 brands, including The North Face and Sacai. We spoke with Toru Takahashi, head of the biotechnology division, about the background to the introduction of Epistra Accelerate and the changes that have occurred since then.

image 10.png

Read the article

visioncare_article.png

Searching for cell recipes for regenerative medicine using AI. Achieving efficiency equal to or greater than that of an expert.

Masayo Takahashi was the first in the world to successfully perform a retinal transplant using iPS cells in September 2014. She is currently the CEO of Vision Care Co., Ltd. and continues to work to bring cutting-edge treatments to many patients. We spoke to her about the background to her collaborative research using Epistra Accelerate and her future prospects.

image 10.png

Read the article

bottom of page