When the body stops moving due to stroke or ALS (Amyotrophic Lateral Sclerosis), "AI" reads the images envisioned in the mind to move machines, regardless of disability or age. This science-fiction-like technology, Brain-Machine Interface (BMI), is becoming a reality. Professor Junichi Ushiba of the Department of Biosciences and Informatics at the Keio University Faculty of Science and Technology has elevated this technology from a mere tool to rehabilitation that draws out the "plasticity" and flexible power of the brain to promote the recovery of lost functions. Based on the philosophy of "gentle technology that stays close to feelings," the IoB (Internet of Brains) interface research connects the brain and machines to create hope for social participation and overcoming physical barriers for people with severe disabilities. The first part of this series, which considers the future of "AI" and the "innovation ecosystem," approaches the front lines of this field.
Profile
Junichi Ushiba
Researcher/Professor, Department of Biosciences and Informatics, Faculty of Science and Technology, Keio UniversityGraduated from the Department of Applied Physics and Physico-Informatics, Faculty of Science and Technology, Keio University in March 2001. Completed the Master's Program in School of Fundamental Science and Technology at the Keio University Graduate School of Science and Technology in September 2002. Completed the Doctoral Programs in School of Fundamental Science and Technology at the Keio University Graduate School of Science and Technology in March 2004, earning a Ph.D. (Engineering). In the same year, he was appointed as an assistant at the Department of Biosciences and Informatics, Faculty of Science and Technology, Keio University. After serving as a senior lecturer in 2007 and an associate professor in 2012, he has held his current position since 2022. From April 2014 to March 2019, he was a principal investigator at the Keio Institute of Pure and Applied Sciences (KiPAS). Since 2019, he has concurrently served as the President and CEO of LIFESCAPES Inc. (formerly Connect Inc.), a company utilizing research results. Co-author of "Biocybernetics: From Physiology to Control Engineering" (Corona Publishing). Recipient of numerous awards, including The BCI Research Award 2019, 2017, 2013, 2012, 2010 Top 10-12 Nominees, and the Ministry of Education, Culture, Sports, Science and Technology's "FY2015 Young Scientists' Prize (Neuro-medical research using brain-machine interfaces)."
Profile
Junichi Ushiba
Researcher/Professor, Department of Biosciences and Informatics, Faculty of Science and Technology, Keio UniversityGraduated from the Department of Applied Physics and Physico-Informatics, Faculty of Science and Technology, Keio University in March 2001. Completed the Master's Program in School of Fundamental Science and Technology at the Keio University Graduate School of Science and Technology in September 2002. Completed the Doctoral Programs in School of Fundamental Science and Technology at the Keio University Graduate School of Science and Technology in March 2004, earning a Ph.D. (Engineering). In the same year, he was appointed as an assistant at the Department of Biosciences and Informatics, Faculty of Science and Technology, Keio University. After serving as a senior lecturer in 2007 and an associate professor in 2012, he has held his current position since 2022. From April 2014 to March 2019, he was a principal investigator at the Keio Institute of Pure and Applied Sciences (KiPAS). Since 2019, he has concurrently served as the President and CEO of LIFESCAPES Inc. (formerly Connect Inc.), a company utilizing research results. Co-author of "Biocybernetics: From Physiology to Control Engineering" (Corona Publishing). Recipient of numerous awards, including The BCI Research Award 2019, 2017, 2013, 2012, 2010 Top 10-12 Nominees, and the Ministry of Education, Culture, Sports, Science and Technology's "FY2015 Young Scientists' Prize (Neuro-medical research using brain-machine interfaces)."
■ The Brain Can Rebuild New Networks
What if one day you collapsed from a stroke and were left with severe physical paralysis? In the medical field, it was long common knowledge that "once brain cells are destroyed, it's over. Lost functions will not return."
"Is that really true?" Starting from that question, Professor Ushiba re-examined the mechanisms of the brain from the perspective of information engineering, eventually leading to the development of rehabilitation devices using BMI.
Simply put, BMI is a technology that allows you to move your body or a machine just by imagining "I want to do this" in your head. It reads brain waves with a sensor worn on the head and links them to the movement of external devices or physical exercise. For example, for a person with paralysis in their fingers due to a stroke, the fingers move in response to the brain waves that appear when they intend to move.
Up to this point, BMI is merely an alternative technology to compensate for lost physical functions. However, Professor Ushiba viewed BMI as an environment that supports the brain's relearning. He thought that even if the damaged brain cells themselves do not return, the brain might be able to rebuild a different neural network for moving the body.
Brain waves are unique to each individual, like fingerprints. Professor Ushiba created a system where a computer instantly analyzes even these subtle differences, and the device reacts only when an appropriate signal is output as an intention to move.
The fingers move only when the brain is used correctly. Otherwise, they do not move. By repeating this strict feedback, the brain learns that "using this neural network will move the fingers" and establishes it as a new motor circuit. Professor Ushiba verified this process in clinical settings and showed that improvement and recovery of function could be seen.
In the 2010s, he published a series of papers showing that the brain has "plasticity" to rebuild new networks and that this is effective for rehabilitation treatment, which attracted international attention. This was because neither the idea of using BMI for rehabilitation nor the clinical results had been envisioned until then.
■ Believing in Intuition and Implementing Rehabilitation Devices in Society
Looking only at the results, BMI feels like a magic wand, but what is required of those who make the wand is the accumulation of sincere and steady research.
First, while a student in the Faculty of Science and Technology, he asked to study in a laboratory in the School of Medicine. He achieved results in an anatomy exam assigned by a faculty member of the School of Medicine and gained a place in the clinical setting. Believing that he could not talk about technology without understanding the brain, diseases, and the lives and dignity of patients, he took as many lectures in the School of Medicine as possible during graduate school.
Eventually, he began conducting neurophysiology experiments in collaboration with the School of Medicine, steadily accumulating correlation data between the movements and brain waves of healthy individuals. He developed an AI program to analyze brain reactions in real time.
The turning point came during clinical research using BMI. In his late 20s, after finishing his doctoral course and teaching students in the laboratory as an assistant in the Faculty of Science and Technology, a student he was researching with spoke to him with a troubled expression.
"When a patient in rehabilitation concentrates on 'moving their hand,' signals appear not only in the brain waves but also from the muscles that were supposed to have stopped. Since there is a lot of noise (interfering signals), why don't we change the program so the machine can move the fingers using the muscle signals instead?"
At that moment, a shock ran through Professor Ushiba.
"That's not noise! Doesn't it mean that the brain of a person who couldn't be cured has been 'rewired' to the point where they can move their own muscles through the machine!"
His own intuition regarding the brain's plasticity, which he had always believed in. "By being able to share that process, I felt a flow of increasing allies."
He proceeded with preparations for social implementation through industry-academia collaboration, but the project was scrapped due to the impact of the 2008 Lehman Shock. Feeling that "it would be a disservice to the patients who cooperated in the research if the results were left to gather dust," he founded "Connect (now LIFESCAPES)" as a university-launched startup in 2018.
After designing a BMI system that could be handled by general medical practitioners and accumulating clinical evidence, the usefulness of training using BMI was highlighted in the 2021 stroke treatment guidelines, and in 2023, it received a high evaluation for its effectiveness in recovering finger function.
He says, "It was a big challenge to see if the product could cure patients with the same reproducibility as us even when it left my hands and was used by others," but by releasing it to the world, new possibilities for moving research to the next stage also became visible.
"There were happy reports, such as it being effective for spinal cord injuries or curing polio. If we conduct research sincerely and tenaciously, we can definitely move toward the future. I felt that science is hope and power."
■ What Was Seen Through Research on Artificial Tails
Apart from the BMI rehabilitation device, Professor Ushiba worked on research into artificial tails from 2014 to 2019. He spent five years immersed in research with free ideas, utilizing "KiPAS," a unique system of the Faculty of Science and Technology that provides an environment where principal investigators selected from full-time faculty members can immerse themselves in research in basic academic fields.
"Can the brain recognize even a body part it doesn't originally possess as its own body and adapt to it? It was a challenge for science."
Multiple participants wearing tails moved by BMI lived in the same space and mastered the tails through trial and error. In the process, he observed how the control methods acquired by individuals were transmitted to others and shared and matured within the group.
Professor Ushiba likens this experiment to the transmission of advanced physical skills acquired later in life, such as piano playing or ballet. The research on artificial tails was also an attempt to explore the mechanisms of human learning and transmission through the tool of BMI.
In the experiment, there were individual differences in the speed and quality of adaptation. "People who accepted trial and error positively learned faster, while cautious people took more time." Intentions and the attitude toward the experiment also influenced the brain's adaptation.
Since 2020, he has participated in the Cabinet Office's Moonshot Research and Development Program. In the Internet of Brains (IoB) project that Professor Ushiba is involved in, they combine BMI and AI to explore the possibilities of technology that transcends physical and cognitive constraints. To make "non-invasive" BMI, which does not require head surgery, usable in daily life, they developed a headphone-type wireless electroencephalograph and applied it to avatar operation in metaverse spaces.
In 2022, the e-sports event "BMI Brainpic 2022" was held in Shibuya, Tokyo, providing an experience of operating avatars and other objects using brain waves across disabilities and ages. A similar exhibition was held at the 2025 Osaka/Kansai Expo.
While the global competition for BMI development is accelerating, in 2024, invasive BMI, which involves embedding a chip in the brain, was administered to a human for the first time as a clinical trial, attracting significant attention. Meanwhile, Professor Ushiba has consistently worked on the development of non-invasive BMI that can be put on and taken off without surgery.
"The goal is to use neurofeedback (*) with BMI rehabilitation devices to draw out the brain's inherent abilities, and ultimately to make the natural brain work more healthily without BMI." This is the idea of using BMI as support for people to regain their own power.
(* "Neurofeedback": A treatment where brain activity is measured in real time using an electroencephalograph or similar device, and the brain learns its own activity while checking those signals.)
■ How to Draw Out the Vitality Inherent in the Brain
"The potential of the brain is quite high. But we just haven't noticed yet how to fully draw out that power. If we use AI, we should be able to further draw out the brain's ability to 'rewrite its functions through experience.'" This is the starting point of Professor Ushiba's research and has remained unchanged.
How to harmonize artificial creations such as AI and BMI with human life and the cycles of nature. Humans can also be said to be part of a large cycle within nature.
"All the principles of the natural world are science, and things created by human handiwork are art," says Professor Ushiba. Researchers create new technologies from science, and artists create crafts and other works with the help of the power of design.
"Art enriches the hearts of those who see it, and by vicariously experiencing the worldview captured by the artist through the work, the way one's heart moves when seeing something next changes. There is much to learn from art."
On the other hand, in his daily morning running routine, he senses the transition of the seasons from the scent of flowers and the color of the leaves on the trees. "I sometimes notice new scenery even in places I always pass, and even if flowers wither, they bloom again the following year. I can feel the things that are 'Reborn' (regenerated) in nature."
The workings of nature are connected to the human brain. Humans have the power to adapt and recover again even if they are hurt.
"What I want to create is technology for people to use their own minds and bodies to the maximum, in their own way, to regain their modest daily lives and connections with others." Professor Ushiba says that passion wells up when he thinks about how to draw out the vitality inherent in the brain.
Driven by passion, people learn from each other regardless of whether they are in the humanities or sciences, and clarify the essence of things. That way of thinking is common to both science and art.
"People can become passionate through learning, and the temperature of the university, which is a place for learning, should rise further. That is something I learned from my artist friends and felt while being involved in management; I believe it is the mission and possibility of a comprehensive university."
Composition: Toru Tamakawa, Editor-in-Chief, Asahi Shimbun GLOBE+
Interview/Text: Maiko Watanabe
Photography: Hiromi Shinada