Paired with a smartphone app called DailyDose, the recommendations from the algorithm were shown to be in agreement with physicians 67.9 per cent of the time.
"Our system design is unique. We designed the AI algorithm entirely using a mathematical simulator," study researcher Nichole Tyler from Oregon Health & Science University (OHSU) in the US, said in a paper published in the journal Nature Metabolism.
"When the algorithm was validated on real-world data from people with type 1 diabetes at OHSU, it generated recommendations that were highly similar to recommendations from endocrinologists," Tyler added.
That's significant because the people with diabetes typically go three to six months between appointments with their endocrinologist.
According to the researchers, people with type 1 diabetes do not produce their own insulin, so they must take it continuously through the day using an insulin pump or through multiple daily injections.
The algorithm developed by the research team used data collected from a continuous glucose monitor and wireless insulin pens to provide guidance on adjustments.
The new study involved monitoring 16 people with type 1 diabetes over the course of four weeks, showing that the model can help reduce hypoglycemia, or low glucose.
If left untreated, hypoglycemia can cause coma or death.
"Very few algorithms have shown a statistically relevant outcome - and most do not compare algorithm recommendations with those of a physician," said study researcher Peter Jacob.
"In addition to showing improvement in glucose control, our algorithm-generated recommendations that had a very high correlation with physician recommendations with over 99 per cent of the algorithm's recommendations delivered across 100 weeks of patient testing considered safe by physicians," Jacob added.