Researchers at the University of Chicago Pritzker School of Molecular Engineering, in collaboration with Argonne National Laboratory, have developed a novel skin-like computing patch capable of running artificial intelligence algorithms directly on the human body. Unlike current wearable devices that transmit data to external servers for analysis, this new device performs computations locally within milliseconds. This capability is critical for time-sensitive medical emergencies where even a few seconds of latency can be fatal. The technology relies on organic electrochemical transistors printed onto flexible surfaces. These transistors differ from standard computer chips by processing information through both electrical currents and ion movement within a gel-like electrolyte layer. This mechanism provides the components with built-in memory, allowing them to store numerical values stably over time, mimicking the synaptic function of the human brain. The primary engineering challenge was manufacturing these sensitive components using standard microelectronics techniques, as the flexible materials react poorly to heat and solvents, and the liquid electrolytes tend to merge and cause short circuits. The team overcame these obstacles by engineering a new polymer gel that can be hardened into precise patterns using ultraviolet light. This innovation enables the production of 10,000 transistors per square centimeter on stretchable substrates. Sihong Wang, an associate professor at UChicago and co-senior author of the study published in Nature Electronics, emphasized that the goal is to create wearable devices that function as personal, instantaneous doctors integrated directly into the body. To validate the device's potential, researchers tested the stretchable array using real cardiac mapping data from a donor human heart. The system successfully located electrical wavefronts associated with ventricular fibrillation, a dangerous condition that can lead to sudden cardiac arrest. The patch achieved 99.6% accuracy in identifying wave positions while being stretched to 1.5 times its original length. This performance is vital because wavefronts move through the heart so rapidly that remote computing is not feasible. By processing data on-site, the device could enable treatments that deliver precise electrical pulses to stop these storms before they become fatal. In a separate demonstration, the neural network analyzed a combination of vital signs, including cholesterol levels, blood sugar, maximum heart rate, and ECG readings. It assessed a patient's risk of a heart attack with 83.5% accuracy. The researchers envision this computing array as a core component of a future health platform that integrates sensing, analysis, and response capabilities. Current efforts are focused on pairing the computing array with stretchable wireless communication components and advanced sensors to create a fully integrated system that processes data where life is happening. This advancement eliminates the need to send health data to remote servers, offering a new paradigm for real-time, AI-driven medical monitoring.