With summer right around the corner in the northern hemisphere and climate changing causing it to be hotter every year, proper hydration is becoming more and more important. One of the issues that not many people seem to be familiar with is that sometimes our bodies start feeling the effects of dehydration before we even notice we are thirsty, and even though our bodies are usually very good at knowing when we need water, a little help is never a bad thing. Considering we are attached to our phones all day, some researchers have put their focus on technology that could connect to our smartphones and help us with this problem.
The proposal comes from a team from King Abdullah University of Science and Technology (KAUST), they have developed a noninvasive way to measure hydration using the same capacitive sensors already built into most smartphone screens and their research, published in the IEEE Sensors Journal in May 2025, shows that by analyzing how your skin reacts electrically, it is possible to determine how hydrated, or dehydrated, you are.
While there are many ways to determine how dehydrated a person is like electronic patches or wearable biosensors, the team leading this project wanted to use something simpler, like tech already sitting in your hand, as Tareq Al-Naffouri, lead researcher, explained.
Can your cell phone see if you need to drink more water?
To test whether or not the sensors on your phone would in fact work to see if you are dehydrated they used a board equipped with two sensors mimicking a phone screen and had people press their fingers on it. They then fed the data into machine learning models to analyze. For testing, they focused on two very different but dehydration-prone groups: athletes after intense workouts, and people observing Ramadan, who go hours without food or water.
In total, the team studied 35 fasting individuals and 10 athletes. For the fasting group, they took five readings daily from morning to evening. Athletes were checked before and after exercise. To make profiles even more complete, they also factored in personal details like gender and weight, so the models could be tailored to individual differences, which helped the algorithms become more accurate, especially since dehydration does not show up the same way in everyone.
As Al-Naffouri pointed out, “It’s also reasonable to expect that the approach could one day benefit other vulnerable groups, including the very elderly, the very young, and those with kidney disease.”
They used different AI models, from simple two-level classifications (hydrated or not) to more nuanced five-level categories to ensure that they interpreted the data correctly and the logistic regression model was the one that had the greatest rate of success, achieving up to 92% accuracy for athletes and 87% for those fasting.
What makes this system exciting is how effortless it could be to use as it would just require users to place their finger on a screen and get real-time hydration information. As an added plus, since the tech is already in most smartphones, there would be no need to invent new hardware, which lowers the barrier for adoption dramatically and could eventually just become a new app that you could download.
Applications could be numerous, but something to consider is that it could be adopted in places where traditional healthcare is hard to access or for people who need daily monitoring. It could also be useful for those who have chronic conditions where dehydration becomes a crucial factor, like kidney problems or heart strain.
As Soumia Siyoucef, one of the authors, said: “We envision easy-to-use, daily, real-time hydration monitoring, where users simply place their finger on their smartphone screen to assess their hydration status.”