Researchers have newly developed a reinforce–ment learning (RL)-based artificial intelligence (AI) algorithm that calculates the amount of in–sulin required by a patient with diabetes and au–tomatically injects it, according to a study pub–lished in IEEE Journal of Biomedical and Health Informatics. The upgraded artificial pancreas sys–tem avoids the inconvenience of having to input meal intake information as required with cur–rently available systems. To do so, the researchers added a pharmacologic concept to RL, known widely as the algorithm of AlphaGo. The AI algorithm, when studied, was able to achieve a mean glucose of 124.72 mg/dL and a time in the normal range of 89.6%, performance that is comparable with conventional artificial pancre–as systems, but without needing to input meal intake. “The fully automated artificial pancreas is like autonomous driving for the medical in–dustry,” said the lead study author. “The newly developed AI algorithm enables fully automated blood sugar control without the hassle of input–ting meal or exercise information. We expect this algorithm to be extended to other drug-based treatments.”
