Whether it’s sniffing your breath, watching the way you type or sampling your poop, there’s no shortage of ways machines can make observations about our health. IBM has a new method to add to the list, however — and, unlike some of the approaches we’ve already described, this one could monitor you continuously. And it could do so in a way that, quite frankly, is easier to explain to house guests than a poop-testing lavatory.

Unveiled today, December 21, IBM has created what it describes as a “first-of-a-kind” fingernail sensor prototype, designed to measure how your fingernails bend and move throughout the day. On the face of it, that might sound pointless. In fact, it serves as an impressively accurate measure of grip strength. This, in turn, can be an indicator of various medical conditions. By gathering this data and using algorithms to parse it, changes can be monitored over time and potentially used to alert clinicians of health concerns.

“It’s designed to capture everyday movement by people in their daily lives — whether they are at home, outside, or in a car,” Ajay Royyuru, vice president of healthcare and life sciences research at IBM, told Digital Trends. “It’s lightweight, wireless, and unobtrusive. We have a team working to get the device even smaller, so that it is just a speck on your fingertip. [Research] looking at grip strength has revealed insights into chronic conditions such as schizophrenia and Parkinson’s Disease. In one cardiovascular disease study, reduced grip strength was a better predictor of death than blood pressure.”

ibm fingernail grip strength sensor shot  pill bottle
IBM Research

Once attached to a wearer, the sensor streams the data it collects to a smart watch and then on to the cloud, so that it can be analyzed by machine-learning algorithms.

“For example, the data can show us if someone is likely to be performing an activity in their home, activities such as holding and operating everyday objects in the kitchen,” Royyuru continued. “Over time, the A.I. system can not only identify what the actions are, but also detect patterns in the data which could reveal insights about the user. [For instance], we could see abnormalities in the data that show someone’s medication is wearing off and tremors are increasing or that grip strength is weakening during certain periods of the day. This can help provide insights about disease progression or state of well-being, such as spikes in stress.”

Don’t expect it to ship any time soon, though. This is a research project, rather than a commercial technology IBM will be launching in the near future. However, while there’s no guarantee that this research will ever be embedded into IBM’s commercial offerings, the fact that this kind of assessment is even possible hints at how the future of medicine may look.