A team of researchers has created a soft, stretchable electronic patch that can run artificial intelligence (AI) directly on the human body without needing to send data to a computer or cloud server.
The new device was developed by scientists at the University of Chicago Pritzker School of Molecular Engineering in collaboration with researchers at Argonne National Laboratory. Their findings were published in the journal Nature Electronics.
Many wearable devices today, such as smartwatches and fitness trackers, can collect health information like heart rate and movement. However, they usually cannot analyze that information themselves.
Instead, the data must be sent to a smartphone or remote server for processing.
That delay may only take a few seconds, but in some medical emergencies, those seconds can matter.
The new skin-like patch performs AI calculations directly on the body in just milliseconds. Researchers hope this could one day help wearable and implantable medical devices make faster decisions and provide immediate health monitoring.
“The future that we’re trying to realize is to make wearable and implantable devices smarter,” said Associate Professor Sihong Wang, one of the study’s senior authors.
The breakthrough became possible because the team developed a new way to manufacture tiny flexible electronic components called organic electrochemical transistors.
Unlike the transistors found in ordinary computer chips, these devices process information using both electrical signals and charged particles moving through a gel-like material.
One important advantage is that these transistors can store information, similar to how connections between brain cells strengthen and weaken during learning. This makes them useful for AI systems that mimic some functions of the human brain.
Building large numbers of these devices was not easy. The flexible materials are sensitive to heat and chemicals used in traditional chip manufacturing. In addition, the gel inside the transistors tends to spread and mix with nearby components, causing electrical problems.
To solve this challenge, the researchers created a new polymer gel that hardens when exposed to ultraviolet light. This allowed them to manufacture highly detailed transistor arrays containing about 10,000 transistors per square centimeter.
The team then tested whether the technology could handle real medical tasks.
In one experiment, the stretchable device analyzed electrical activity in the heart to identify dangerous wave patterns linked to ventricular fibrillation, a life-threatening heart rhythm disorder. Using data collected from a donor human heart, the device located these wave patterns with 99.6% accuracy, even when stretched to more than one-and-a-half times its original size.
In another test, a neural network stored within the device analyzed health information such as cholesterol levels, blood sugar, heart rate, and ECG readings. It predicted heart attack risk with an accuracy of 83.5%.
Researchers believe this technology could eventually become part of advanced health-monitoring systems that can sense, analyze, and respond to medical problems in real time.
Instead of sending health data to distant servers, future devices may be able to understand what is happening inside the body instantly—right where it matters most.
