Prevention starts with better data

Detecting changes in a person’s health before symptoms develop or diseases progress is a key goal of preventive medicine. The medicine of the future should therefore not only treat diseases, but also help prevent it from developing in the first place. This raises a key question: How can health and disease be assessed objectively at the earliest possible stage? High-quality medical data are essential. What matters is not only what is measured, but also how.

Many medical decisions are still based on measurements taken at individual points in time, such as during a doctor’s appointment, medical examination or hospital stay. Wearables such as smartwatches or smart rings allow users to collect data continuously over extended periods. However, measuring individual health parameters is often not enough for medical research and healthcare. To understand complex physiological interactions, different types of health data, such as heart activity, respiration and movement, need to be considered together. Commercial systems are often not ideally suited to this purpose, as they are generally designed as closed, standalone solutions and provide only limited access to the underlying raw data.

From individual measurements to the bigger picture

The first part of the series on Science Year 2026 showed how continuous ECG measurements using innovative medical sensor technology can support the early detection of cardiac arrhythmias. However, many questions in preventive medicine go beyond this and require the synchronous recording of multiple biosignals.

The real challenge is not building sensors, but capturing different physiological parameters simultaneously, with precise synchronization and under real-world conditions: in everyday life, outside controlled laboratory environments and without restricting users’ freedom of movement. With maphera^®, Fraunhofer IIS has developed a decentralized sensor network that connects different sensors to form a mobile system. The Bluetooth-based platform enables the multimodal acquisition of physiological data and synchronizes different measurement sources with an accuracy of 30 microseconds. This allows different biosignals to be precisely aligned in time and analyzed together.

Prevention needs context

Such datasets offer possibilities that go far beyond measuring individual vital parameters in isolation. By considering heart activity, respiration, movement and other biosignals together, physical and mental strain as well as changes in a person’s health can be assessed much more accurately. This provides a more comprehensive picture of how physical or mental strain affects people in everyday life, which factors may contribute to long-term health risks, and how individual tolerance limits can be determined objectively.

Multimodal measurements also open up new possibilities for occupational health and safety. For example, understanding the strain associated with physically or cognitively demanding tasks requires more than simply knowing how much a person moves. Only by combining different biosignals does a more complete picture emerge. While ECG and optical measurement methods such as photoplethysmography (PPG) provide information about strain on the cardiovascular system, inertial measurement units (IMUs) capture posture, movement patterns and acceleration. Together, these data allow physical strain, fatigue and adverse movement patterns to be analyzed more precisely.
 

When data becomes knowledge

Beyond prevention, new research approaches in other areas of healthcare can also benefit from this approach. One example is the flagship project ROBO.KIWI. To preserve the valuable experiential knowledge of nursing professionals and make it available for training, maphera® captures their movement and interaction patterns with patients in real-world care settings. The resulting data are combined into multimodal datasets from which characteristic movement and interaction patterns can be derived. These curated reference movements are then transferred to humanoid robots, which are intended to serve as physical learning partners in nursing education.

This example shows that modern medical sensor technology can do far more than measure individual vital parameters. By recording different biosignals synchronously and analyzing them together, it is possible to create datasets that reveal complex physiological interactions. This is precisely where the potential of this key technology lies, both for preventive medicine and for many other areas of healthcare.