Researchers propose framework to make tear-based health sensors more reliable

A team of researchers at the Terasaki Institute for Biomedical Innovation, working with colleagues at UC San Diego and USC, has published a framework in Nature Sensors aimed at solving a fundamental problem with tear-based health sensors: they treat tears as a static source of biomarkers when, in reality, tears are a constantly shifting physiological system.

The appeal of monitoring tears is straightforward. For people with dry eye disease, glaucoma, or other eye conditions, current care relies on occasional visits to the clinic and single measurements taken at one moment in time—snapshots that can easily miss how symptoms fluctuate from day to day. Tears, by contrast, are continuously produced and replaced, and they carry chemical signals tied to eye health and sometimes to broader systemic health. This makes them an attractive target for continuous, minimally invasive monitoring. Yet the sensors developed so far have focused primarily on two goals: detecting more biomarkers and making the hardware smaller. What they have largely ignored is the fact that tears themselves are in constant motion, shaped by blinking, evaporation, and the turnover of fluid.

Dr. Yangzhi Zhu and his collaborators argue that this oversight is precisely what has prevented tear-based sensors from becoming reliable long-term tools. The framework they propose treats the tear film itself—not just the biomarkers within it—as central to how a sensor should be designed. Different biological signals require different approaches. A measurement of pH or electrolytes, for instance, calls for a different sensing frequency and device format than tracking larger proteins or inflammatory markers. The team also emphasizes the need for systems that combine chemical sensing with contextual data: information about blink patterns, tear flow, and other physical parameters that can help distinguish genuine biological change from the normal fluctuations that occur simply because tears are constantly moving.

Dr. Chia-Wei Liu, a postdoctoral researcher at the institute, put it plainly: if a sensor does not account for blinking, evaporation, and fluid turnover, it becomes very easy to mistake the tear film's ordinary behavior for a meaningful biological signal. The framework spans four areas: the biological interface between sensor and tear film, how the device performs under real-world conditions, the path to clinical translation, and the trust required for patients and clinicians to adopt the technology. It also acknowledges that no single device design will work for every application. Smart contact lenses, punctal devices (placed at the tear drainage point), and periocular patches each have different strengths and suit different monitoring needs.

Zhu emphasized that the field needs to move beyond simply counting how many biomarkers a device can detect. The real question is whether the readings can actually be trusted over time. That requires designing sensors and the methods to interpret their data together, not as separate problems solved independently. Dry eye disease emerges as the natural first testing ground for this approach, given how closely it is tied to tear-film instability and symptom variation. Glaucoma monitoring and metabolic tracking are identified as additional potential applications down the line.

Bringing this vision to reality will demand a shift in how research is conducted. Short-term wear tests—the current standard—are not sufficient. The field needs longitudinal studies that follow patients over weeks or months and compare tear-based sensor readings against established clinical measures to validate whether the technology actually works. The Terasaki Institute is continuing to invest in translational research aimed at moving tear-based ocular monitoring from the laboratory into clinical practice, but the path forward depends on whether researchers can build sensors that respect the dynamic nature of the tear film itself.