In the segment 'A Look Around Innovation,' we introduce research and development (R&D) sites that have led to social implementation. In the 29th installment of this series, JST News interviewed Lecturer Ayumi Amemiya at Chiba University's Graduate School of Nursing and her team. They are developing a system to detect in advance the unintentional removal of IV lines by hospitalized patients with cognitive decline and to prevent this "self-removal" from occurring. Through a "cyclical development" partnership between the university, a corporation, and a medical facility, they aim to achieve dignified care for patients with dementia.
Reducing physical restraints for patients: Partnering with a corporation for full-scale development
Chiba University's Inohana Campus is located about a 15-minute bus ride from Chiba Station. Here, Lecturer Ayumi Amemiya, a specialist in nursing engineering, launched the "REGARD" project in collaboration with medical device manufacturer I-Medex Holdings (Chiba City) and Izumi Chuo Hospital (Chiba City). The team is developing a system that uses contact-detection sensors to reduce physical restraints for patients with cognitive decline.
The motivation behind this research is a strong desire to "achieve dignified dementia care where no one is left behind." Patients with reduced cognitive function sometimes unconsciously touch and pull out their IV tubes. When this happens, treatment is interrupted and the IV must be reinserted. To prevent this, many hospitals reluctantly resort to physically restraining dementia patients (Figure 1). Before becoming a researcher, Amemiya worked as a hospital nurse and experienced several situations where she had no choice but to use physical restraints.
"I had been thinking for a long time about whether there was a safe method that could reduce physical restraints on patients without increasing the burden on medical staff," Amemiya says. She actually began research and development about 10 years ago. Her idea was to use a sensor that responds when a patient touches an IV tube, thereby detecting movements likely to lead to self-removal. Initially, Amemiya built prototypes by herself. However, realizing that full-scale development required corporate collaboration, she reached out to President Makoto Ichida of I-Medex Holdings, in 2017.
A wave-shaped contact-sensor sheet: Usefulness and high accuracy proven in hospital setting
With the medical device manufacturer on board, research and development progressed significantly. Based on the rough mechanism Amemiya had already conceptualized, Ichida's company and Amemiya's laboratory worked together to refine the sensor component. Ultimately, they adopted a method that detects when a hand or finger touches a contact-detection sensor sheet placed over the area where the IV needle is inserted and secured. The sensor sheet measures 6.0 centimeters wide by 13.7 centimeters long. "To ensure the sheet fits even the thin arms of elderly patients, we cut it into a wave shape rather than a rectangle," Ichida explains. "We also colored the circuits on the sheet so they wouldn't bother the patient." (Figure 2).
The system is designed so that when the sensor reacts to a patient's touch, the tablet at the nurse's station displays "Detecting" and sounds an alert. This software was also developed from scratch. Furthermore, to ensure the system only detects movements where a patient is intentionally fumbling with the IV out of curiosity or discomfort, it is configured to trigger an alert only when touched two or more times within 1.5 seconds (Figure 3). This detailed setting is a direct result of Amemiya's deep familiarity with clinical settings.
Following the system's completion, Amemiya approached numerous hospitals to conduct a clinical validation study. However, introducing new equipment into clinical settings faces substantial barriers, and she struggled to find cooperative facilities. It was then that Ichida introduced her to his long-time friend and surgeon, Dr. Kazuya Yamahatsu. Dr. Yamahatsu had just left his previous hospital job to become the Director of Surgery at Izumi Chuo Hospital, which was managed by his family.
"I wanted to cooperate by all means if the research benefited patients and reduced the burden on our staff," Dr. Yamahatsu recalls. It was also fortunate that the hospital was a small-to-medium-sized facility with 98 beds, and Dr. Yamahatsu was the person responsible for ward management. The study began with a few patients and eventually included 14 participants. When inviting them to join, Amemiya carefully explained the study to both the families and the patients themselves. Many families reportedly said, "If it means avoiding physical restraints, we definitely want to participate."
The hospital study showed that when this system was used, physical restraints could be removed for 440 minutes out of a 480-minute observation period. On the other hand, without the system, restraints could be removed for only 8 minutes out of 480 minutes, proving the system's utility. Amemiya adds, "Without the system, when physical restraints were removed, patients often touched the IV out of curiosity, or nurses felt anxious and quickly reapplied the restraints." Furthermore, over a total of 70 hours of use, there were only two false alarms, demonstrating the system's high accuracy.
Advantages of cyclical development: Speed and deep understanding — Post-sale support forms the next challenge
The REGARD project, which aims to reduce physical restraints through this contact-detection sensor system, won the 2025 Minister of Education, Culture, Sports, Science and Technology Award at the "STI for SDGs" Awards, a JST initiative recognizing efforts to solve social issues through science, technology, and innovation. The project also drew attention because it is driven by cyclical development, where a university, a company, and a medical site collaborate to repeat on-site testing and university-led improvements.
Amemiya highlights speed and a deep mutual understanding as the core strengths of cyclical development. "In many cases, company engineers develop devices based solely on information provided by researchers. This time, however, Ichida came to the hospital himself to see the actual situation, which allowed us to make necessary improvements quickly." Having Dr. Yamahatsu on-site as the manager also enabled smooth coordination with the hospital staff. Ichida notes, "Cyclical development is taken for granted overseas. I think it will gradually catch on in Japan as well."
Currently, the team is discussing the commercial incorporation of REGARD. Among the challenges toward commercialization, Amemiya believes post-sale support is the most critical. "We want to think beyond just selling the product and ensure we provide ongoing support for the nurses on the ground," she says.
Since 2023, Amemiya has also been working on research to detect cognitive decline early in home care under JST's Fusion Oriented Research for disruptive Science and Technology (FOREST) program. Whether in hospitals or at home, she keeps her sights set on realizing a society where no one is left behind and "dignity is protected until the very end of life."
(Article: Noriko Higo, Photography: Erika Shimamoto)