Would you trust artificial intelligence to manage your diabetes?

When we talk about advances in artificial intelligence (AI) applied to diabetes, we cannot fail to mention the work of Peter G. Jacobs, a visionary who is transforming the way we manage this condition.During the recent IFAC Workshop on Engineering Diabetes Technology, Jacobs shared his vision of the present and future of AI in diabetes, and made clear an essential point: the patient's confidence in technology is key to his success.

Jacobs, since his Aims Laboratory (artificial Intelligence for Medical Systems), has led research that has demonstrated significant improvements in time in the range of patients who use advanced algorithms for insulin management.An example is its KNN-DS project, a support system to the clinical decision that has allowed patients to increase the time in range by up to 6.3%, something that for us, that we deal with constant rises and declines, is a real achievement.

The most interesting is his human and ethical approach to technology.For Jacobs, the challenge is not only technical, but also human.He insists that people should trust algorithms to use them correctly.And this makes me reflect, because when I started with the insulin bomb, at first I doubted how it would work.However, over time, trust and good results made me believe in the system.I imagine that the same can happen with these new advances in AI.

Another aspect that caught my attention is the use of digital twins.According to Jacobs, these virtual models allow to precisely replicate the behavior of a person with diabetes, which facilitates adjusting the treatments in a totally personalized way.This sounds incredible, since each of us is a world, and having a technology that understands how we react at all times would be a before and after.

Jacobs also spoke of a delicate topic: the privacy of the data.In his speech, he warned that many health professionals are using AI tools without considering privacy risks.This is something we should demand: transparency and clear protocols to protect our medical information.

And I could not miss the issue of biases in AI.According to Jacobs, an algorithm is as fair as the data with which it is trained.To combat this, in their laboratory they have created a training program for new generations to learn to identify and mitigate these biases.

💡 The future looks promising, but it will only be possible if we trust and adapt to these new technologies.

I would love to know what you think:

Would you trust an algorithm to help control your diabetes?

Would you like to try a predictive and personalized system?

I read you in the comments!💬💙

"Up to 6.3% more time in rank" seems very little.

Although it does not seem much, if improvements of that magnitude are accumulated continuously, it can end up being a brutal increase, it is a matter of time, in my opinion it is a real way of potential improvement.

Sometimes we read very promising, but unimarable studies to reach patients, in this case, I would say that it is very likely that we can benefit, you just have to see the degree of adoption that the AI ​​is having in all sectors.

I hope you get the percentage much more, and that we can use it soon!

It is true that the potential of AI in treatment is very high.That is why I miss me so little improvement.

Because, if it were 6.3% yet, but 6.3 maximum, it will mean that the majority barely noticed improvement.

Not flatly!

I trust a thousand times more of my human brain than of an AI.I consider it unable to manage all the factors that influence diabetes day by day (I think 42 factors were mentioned in another thread).

If one day I lose my head, before self-rising making mistakes with the insulin dose or the type of insulin, it may be interesting to delegate the control of diabetes in an AI.

It seems that we have a good bud debate @ricki21!😋

It is understandable that doubts and some distrust arise, especially when we talk about a control as delicate and personal as diabetes.However, I would like to share a reflection on this subject.

Artificial intelligence does not replace our decision -making capacity, but to complement it.Technology, when integrated correctly, acts as a constant support that minimizes human errors and improves precision in decisions, especially in critical situations.If we think about it, we already trust advanced technology in many aspects of our life (GPS navigation, medical assistants, financial management, etc.), and diabetes should not be an exception.

I understand skepticism, but 6.3% more time in rank is not a simple number.For many of us, that means less hypoglycemia, less sudden increases and, above all, more stability.And this, accumulated throughout months and years, becomes a significant improvement in the quality of life.In addition, as technology evolves, these percentages will increase.

It may not be "blindly" to trust the AI, but to understand that it can be another tool to help us, such as glucose sensors or insulin pumps.It is a more step towards more automated and safe management.

What do you think?Do you think that in the near future we will see a qualitative leap in these percentages?I read you!😊

Good afternoon everyone

It seems to me a very good topic to discuss.In advance that I am not very much put on the new technologies, but I would already bet on AI.

My reasoning is as simple as primary.After many years dealing with this, and we already know the amount of daily decisions that the treatment entails and therefore the fatigue that it implies, especially psychological, continually playing all those variables that have already been listed recently in an open thread, I think they were 42, flatly yes.

No one can escape that this ability to decide, to control diabetes, in short, is diminishing with age.The AI ​​could be that support cane in Middle/Advanced Ages, and I refer after 65 years.It is worth the cane simile.I am not exactly of the optimists with the future that awaits us, in fact I refrain from giving my opinion on issues or threads such as diabetes and the world of work, because of how negative it can be, according to my personal experience.

In his day I saw a great advance those basal insulins, Lantus type, toujeo or threeiba, which are helping us so much.Needless to say, they have the MCG, I mean the sensors;correct?They have drastically improved our lives.

I imagine that these percentages of improvement in time in range will be initial estimates and that they will improve even more and with it the control of our diabetes.

Greetings.