Counting steps, tracking the heart rate, logging how long we sleep: many people already use smartphone apps and fitness trackers to keep an eye on their health. These numbers give a rough idea of whether we reach our goals, but they are too fragmented to tell whether someone is living healthily over time. This is where SensorTech4Health (ST4H) comes in. The international project develops assistance systems for preventive healthcare that help people live independently in their own homes for as long as possible.

According to Dr. Stefan Finkbeiner, CEO of project coordinator Bosch Sensortec, the partners want sensors to become more than technical components: companions for a healthier life that fit seamlessly into everyday routines. Putting prevention at the center should enable people to stay fit and self-determined for longer.

The new systems are designed around their users, with smart technology that makes them easy and intuitive to operate. Industry partners work together with the sleep research team at the Medical Center – University of Freiburg and with gas sensor experts at Saarland University, with the aim of developing market-ready products.

Smells, sounds and AI for a healthier society

Recognizing health risks early matters for each individual, and in an ageing society with expensive care and a shortage of healthcare workers, it matters for everyone. People who keep track of their health, stress level and sleep can maintain their quality of life for longer and help avoid high costs for the healthcare system.

The ST4H systems collect health information from their surroundings, for example in a home or a hospital room, or directly on the body. Using machine learning, they learn to interpret these measurements, link them to a specific situation and suggest suitable steps, even if that is just opening a window when the air is stale. Dr. Christian Bur, project PI at Saarland University, explains that the solutions are meant for everyone: “The aim is to develop technical solutions for everyone – from young people, who want to lead healthy lifestyles, to the elderly, those in need of care and people with chronic illnesses, all within their own homes.”

The devices will monitor sleep quality, stress level and health parameters automatically and discreetly. They are designed to detect substances emitted by the body and in exhaled breath as well as volatile organic compounds in the air. Artificial intelligence then relates these measurements to data from other sensors.

Privacy as a design principle

The partners develop assistance systems that can be integrated into a smart home. Such technology could help people stay in their own homes for longer. Based on typical sounds and smells, an assistant could, for example, infer whether someone is active and cooking, and that everything is fine.

This works through the combination of gas sensors and microphones. The microphones do not record speech; they only capture sound as pressure changes on a tiny membrane. A single measurement says little on its own. Only when data from several sensor types are combined and analyzed with AI does a new kind of assistance technology emerge.

Privacy is a priority for the whole consortium. The systems use no cameras and no speech recognition, and they do not upload data to a cloud: all signal processing happens locally on the device. The technology is meant to work quietly and automatically in the background, without anyone having to press a button.

From bedside devices to hearing implants

Beyond the connected home, the partners are developing smart devices for the bedside and even integrating gas sensors into textiles such as bed sheets. These could continuously monitor health parameters of people in need of care without disturbing them.

Since hearing loss increases the risk of dementia, ST4H also works on hearing support: cochlear implants that perform better thanks to improved microphones, and other devices that reduce background noise and make conversations easier to follow.

Finding single molecules among billions

A central goal of the project is to monitor health and wellbeing with new gas sensor technology. The body constantly releases metabolic products through the skin and breath. These mixtures of molecules are highly characteristic, even though the human nose usually cannot detect them.

Christian Bur, a postdoctoral researcher who leads the medical gas sensor research team in Andreas Schütze’s measurement technology research group at Saarland University, develops gas sensor systems that detect such volatile organic compounds with increasing precision. The challenge is to pick out tiny concentrations of individual substances among billions of other particles in the air. Together with partner companies, the researchers develop highly sensitive metal-oxide semiconductor gas sensors with new sensing layers that continuously measure the concentration of specific substances, and use machine learning to identify them. Saarland University receives 1.2 million euros within the project.

About SensorTech4Health

SensorTech4Health brings together partners from academia and industry in the field of digital health. The international project receives 15 million euros from national funding organizations within the EUREKA Cluster Xecs. The German partners are funded by the Federal Ministry of Research, Technology and Space (BMFTR) through VDI/VDE Innovation + Technik GmbH. The project partners come from industry and research in Belgium, Finland, Germany and the Netherlands.