When your fitness tracker or doctor's pulse oximeter shines a light on your skin to measure your heart rate, that technology was probably tested on fake skin first. For years, those test models have mostly looked like light-colored skin. But a team in Finland just changed that.
Researchers at the VTT Technical Research Centre of Finland have built realistic skin-like test models, called optical phantoms, that look like lighter, medium, and darker skin tones. They even have tiny artificial blood vessels inside with a blood-like liquid flowing through them, connected to a miniature pump.
The problem these researchers are solving is real. Many medical devices that use light — like pulse oximeters, which clip onto your fingertip to measure oxygen levels, and fitness trackers that monitor your heart rate — can give less accurate results for people with darker skin. This raises serious concerns about healthcare fairness and patient safety.
The VTT team built their phantoms in layers, just like real skin. There is an outer layer that mimics the epidermis, a deeper layer for the tissue underneath, and a bottom layer for fat. They embedded a network of tiny fake blood vessels and hooked them up to a pump that circulates a liquid that behaves like blood.
The researchers tested three skin tones: lighter, medium, and darker. These matched the range of skin colors found in European, South Asian, and African populations. When they shined light through the models and measured what came back, the results closely matched what happens with real human skin.
The tests revealed something important. In lighter skin models, the flowing blood-like liquid created clear, detectable signals. As the skin models became darker, those signals grew harder to pick up. In the darkest phantom, the pigmented outer layer largely blocked the signal from the blood vessels below.
This finding matches what happens in the real world, according to the researchers. More pigment in the skin makes it harder for light to reach the blood vessels underneath, which can affect how accurately devices measure vital signs.
The phantoms also turned out to be remarkably durable. After nine months of use, their optical properties had changed only slightly. That means labs and companies could use the same models over and over for testing and quality checks, rather than constantly replacing them.
The team, led by researcher Anni Ranta-Lassila, published their work in the Journal of Biomedical Optics. They now plan to expand the range of skin tones even further and compare their phantom results directly with measurements from real human skin.
Ultimately, the goal is straightforward: help device makers catch and fix biases during the design phase, so that the smartwatches, medical monitors, and health sensors of tomorrow work just as well for everyone who wears them.
