Press "Enter" to skip to content

Artificial Intelligence: The cure for diabetes?

An estimated 34.2 million people in the United States have diabetes, excluding an estimated 7.3 million not yet diagnosed, according to the Diabetes Research Institute Foundation. Diabetes is a metabolic disorder that is due to the lack or inefficient use of insulin, which is a hormone produced by the pancreas that regulates the amount of glucose in the blood (blood sugar). 

 

Although diabetics are no longer able to naturally regulate their glucose levels, there are new and unique medical solutions that are in development to help track and regulate blood sugar levels.

 

Artificial intelligence (AI) is being developed in the medical field to help with the treatment and management of diabetes. This new technology can help those affected by the metabolic disorder painlessly monitor and regulate blood sugar, and maybe one day replace the pancreas.

 

“Almost every medical area is using some form of robotics, artificial intelligence, and augmented and virtual reality to educate patients and students,” said ORU’s Vice President of Technology and Innovation Mike Mathews. “Even surgeons are using these technologies for research purposes.” 

 

If AI is fed enough data, it can be used to make many decisions and recommendations that can supplement human judgment.

 

“Artificial intelligence is ultimately capable of learning millions of behavioral patterns and making predictions,” Mathews explained. “If you put it into the daily lives of diabetic people and allow artificial intelligence to ask behavior questions, it starts learning personal behavior and recommends lifestyle changes, including what you should or shouldn’t eat.”

 

We use this type of behavior-learning AI in our daily lives. Siri, Netflix, Pandora, Spotify — they all use pattern-recognition algorithms to pick up on users’ favorite things to watch or listen to in order to successfully suggest relative entertainment. 

 

It’s through this learning that AI is able to regulate a person’s blood sugar relative to that individual’s behavior and body.

 

As an example of this behavior-learning AI, apps like Glucose Buddy are now being developed by medical engineers and used by diabetics to help monitor their blood glucose levels, thus eliminating the need for finger pricks. Glucose Buddy is free, has an online forum to connect with other users, can graph both glucose levels and medication on the same chart and has an A1C estimator based on past blood-glucose levels. 

 

“If someone is on an artificial-enabled phone, they would dial-in with voice recognition which accesses the worldwide database that has all of your medical records in it. The automated medical record prompts the patient on recommended medicines, dosage frequencies and amounts, while digitally recording your up-to-date fitness and blood sugar levels,” Mathews said.

 

Not only are more and more apps becoming available for diabetics, but smartwatch-like technology is being researched and developed. As of 2018, the University of Waterloo is working on adapting Google’s Project Soli to become a wearable monitor. This technology, instead of measuring gestures like Project Soli, would measure different blood properties via radar and relay that information back to the chips in the hardware.

 

“Even though we’re not perfect human beings, diabetes is controllable for the most part,” Matthews said. “Artificial intelligence has the capability of prompting and reminding me of my personal well-being and behavior patterns.”

 

Lastly, an artificial pancreas is being developed by a research team at McGill University, with Dr. Ahmad Haidar, an assistant professor in the department of biomedical engineering at McGill and a scientist at the Research Institute of McGill University Health Centre. This artificial pancreas would automate “insulin pump delivery in type 1 diabetes based on glucose sensor readings and a dosing algorithm,” according to Dr. Haidar.

 

Once this technology is perfected, it will be offered to patients—but will it all be equitable?

 

“I think one day artificial intelligence will be affordable and seamless with human lifestyles and make a worldwide impact,” Mathews said.