Smart sensor patch which can detect safety problems in real-time

A new multimodal sensor patch, developed by Japanese researchers, detects arrhythmia, coughing, and falls in real time using edge computing.

A team led by Professor Kuniharu Takei of Hokkaido University and Associate Professor Kohei Nakajima of The University of Tokyo has developed a flexible sensor patch capable of real-time health monitoring. The patch is outline in their paper “Real-time personal healthcare data analysis using edge computing for multimodal wearable sensors” published this week in the journal Device.

The patch tracks vital signs such as cardiac activity, respiration, skin temperature, and humidity through small, skin-adhering sensors. By connecting to a smartphone via Bluetooth, the patch uses edge computing to process this data locally.

Unlike cloud-based systems, edge computing performs data analysis on the smartphone itself, reducing latency and avoiding the potential security risks associated with transmitting sensitive data. As Professor Takei explains, “Our goal was to design a multimodal sensor patch that could process and interpret data using edge computing, and detect early stages of disease during daily life.” The team tested the patch on volunteers and found it effective in detecting conditions such as heart arrhythmia, coughing, and even falls.

Lab safety enhanced by wearable technology

Although initially developed for healthcare, these sensors hold substantial potential for improving safety in lab environments, where workers are often exposed to hazardous conditions such as chemicals, high temperatures, or dangerous machinery. For instance, laboratories operating in high-temperature conditions or using equipment that generates significant heat could benefit from real-time monitoring of workers’ skin temperature and humidity.

The research team tested the sensor patch in extreme heat conditions, monitoring volunteers under wet-bulb globe temperatures—a measure used to assess heat stress. “Although our test group was small,” explains Takei, “we could observe vital signs change during time-series monitoring at high temperature. This observation may eventually lead to identifying symptoms of early-stage heat stress.” This early detection is particularly important in labs, where rapid intervention can prevent heat-related illnesses or accidents.

By providing continuous monitoring of workers’ physiological responses, this technology could serve as a safeguard in labs, alerting H&S teams to potential risks before they escalate into serious incidents.

Addressing fall and accident risks

Falls are a common concern in laboratory environments, especially where there is heavy equipment, delicate instruments, or hazardous substances. The wearable sensors can detect when a fall has occurred and trigger alerts in real time. This could be especially useful for workers who are alone in specific lab areas, ensuring a swift response from safety teams.

Using machine learning algorithms, the wearable patch can accurately detect falls, arrhythmia, and coughing. According to Associate Professor Kohei Nakajima, “We achieved prediction accuracy of over 80%,” showing the system’s reliability for early warning and rapid response. This makes it particularly promising for safety-critical environments such as laboratories, where early detection of health incidents is key to preventing accidents and ensuring worker wellbeing.

Challenges and future applications

Despite its benefits, the current system has limitations. One notable challenge is that while the data analysis can be performed on the smartphone, initial machine learning training must still be done on a computer. This extra step adds complexity, but researchers believe future developments will streamline the process, allowing fully independent operation on smartphones.

Looking ahead, the integration of this wearable sensor system with edge computing could extend beyond the lab environment. Professor Takei envisions its broader application in telemedicine and telediagnosis. For laboratories, however, the immediate benefits are clear. Incorporating these sensors into H&S protocols could significantly reduce the likelihood of accidents or health emergencies.

By continuously monitoring vital signs and analysing data in real time, these sensors offer a practical solution for improving worker safety in labs. As technology advances, wearable sensor systems may soon become a standard feature of lab safety, enhancing both monitoring and response capabilities.