Researchers use AI to optimize sensor design for ultrasensitive vitamin C detection
Biomedical EngineeringChemistryBiology
THE AI ANGLE
Predicting optimal sensor fabrication conditions from limited experimental dataResearchers trained a machine learning model on 25 initial tests to optimize an electrochemical polymer sensor for vitamin C out of 625 possible fabrication combinations. The fabricated sensor achieved a detection limit of 778 picomolar in human serum and artificial urine. The method allows sensor designers to skip hundreds of physical trial-and-error experiments when setting fabrication parameters.
THE TEACHING ANGLE
Classes can discuss whether a model trained on only 25 experimental data points can reliably find the best result among hundreds of untested parameter combinations.Read the original at phys.org Generate teaching or study materials