AI is Helping Me Manage My Pre-Diabetes

Health | Blood Sugar | Diabetes

How I’m Using Technology to Control Blood Sugar, Lose Weight, and Stay Healthy

Disclaimer:I am not a doctor, and no part of this article should be considered medical advice. I’m sharing my own explorations into how I’m managing weight loss and avoiding diabetes.

All your health care questions and challenges should be discussed with your personal health care professional. This article should only be considered entertainment and not be used for education or healthcare.

The Present

If you’ve been following me, you’ll know I’m pre-diabetic and obese. In recent days, I’ve been trying to figure out how I got here and what I should do next.

I’ve decided to believe I have a genetic predisposition to diabetes arising out of being the descendant of generations of South Asians who suffered famines in the past. Deeper dive here.

This predisposition makes South Asians (and other populations that survived famines) more susceptible to developing diabetes. It seems our ancestors evolved to survive the chronic shortages of food paradoxically by impairing insulin production.

Impaired insulin production allows for blood sugars to go well outside of normal ranges and while this can increase food storage as fat, it also predisposes us to diabetes .

It worked well to help my ancestors survive but when presented with periods of food abundance, the survival mechanisms go awry. We develop diabetes at much higher rates than almost every other demographic, whether obese or not.

And, like most patients who are pre-diabetic or have type 2 diabetes, we would also have been exposed to an excess of sugars and likely proteins and fats as well in our diets.

My hypothesis then — for me to reverse prediabetes, I’d need to re-create artificial conditions of famine in my own life.

I’ll have to cut down, cut out and give up excesses in sugars/carbs, proteins and fats in my diet and include moderate exercise, enough to promote weight loss.

Enter Artificial Intelligence

I’ve been tracking blood sugars diligently every morning after fasting between 16–18 hours. For the past week, I’ve observed a downward trend moving from 123 mg/dL to below 100 mg/dL. That has been my objective.

All I’ve done is input the dates and readings into ChatGPT and asked it to generate a table. Here’s what that looks like:

I kept close track of my daily input of carbs using ChatGPT. Again, I input exactly what I ate and asked it to calculate the number of carbs I ingested and to give me the result in a table.

Note, all I kept track of was what I ate and how much. The Ai did the rest.

Here’s what that looked like yesterday when, for the first time, I got a fasting blood sugar below my target of 100 mg/dL:

I wanted to know how many calories I ate, so I simply asked it to calculate the number of calories I ate based on the data I’d already given it. It came up with the table below:

I have to say it — this was wonderful. I no longer needed to track every detail by hand and I could enter data just once to manipulate it in different ways.

What I plan to do going forward is to input my meals and blood sugars every day in ChatGPT and generate a daily report to get a sense of how my diet affects my blood sugars, with specific focus on carb intake and subsequent fasting blood sugars.

Based on the data, I can figure out what sends it over my goal reading of 100 mg/dL., and what keeps it within range. It’s targeted, personalized help for a condition I need to manage. For me, it works beautifully.

Tables aren’t everything

Then I asked my Ai companion this question: Do you think that my meals today were appropriate for a prediabetic trying to lose weight and increase my insulin sensitivity?

The answer it returned was based on the food data I input. That blew my mind. It analyzed my specific food choices and returned multiple positive aspects e.g.

Low-Carb Choices:You kept your carb intake relatively low, which is important for blood sugar control in pre-diabetes. Your total carb intake was around 28.3–29.3 grams, which is well within a moderate low-carb range.

It then suggested several areas for improvement including:

More Focus on Balanced Meals:While your meals were healthy overall, you might want to include more variety and balance. For example, incorporating moderate portions of low-GI fruits (like berries, which you did), whole grains (if tolerated), or legumes could improve insulin sensitivity over time.

And finally it gave me suggestions on what I could do going forward:

Replace whole cream milk with lower-carb alternatives.

Add more non-starchy veggies to increase fiber and volume.

Continue focusing on healthy fats and lean proteins but keep an eye on overall calorie intake to align with your weight loss goals.

Your meal choices were largely appropriate for managing blood sugar and promoting insulin sensitivity, but slight adjustments could further support your weight loss efforts.

Don’t trust the Machine

Okay, I know it’s a machine. I know I cannot rely on it’s advice and neither should you gentle reader. But damn, it gave me enough to work with for when I speak to my doctor next.

I will cut out the whole cream milk though. That’s sensible advice.

I plan to generate a table of blood sugars vs. dates and take it to my doctor, and if she asks about my diet, I can share the entire GPT chat with her via a link. That way she can see exactly what I’ve been doing — whether right or wrong and then make adjustments as needed.

Now isn’t that too cool for school? Smiling.

Conclusions:

I’m enthusiastic about using Ai sensibly. This chat with the Ai opened my eyes to the possibility of using artificial intelligence to manage large amounts of critical data in my personal life. It’s because I’m using Ai to monitor my diet and blood sugars that I can take better control of my condition.

Could I do it with a pencil and a notepad? Yes.

But it would take more time, more effort and would generate much more friction. At the same time, I wouldn’t be able to share the details of my diet and pre-diabetes management with you as easily as I have.

One more thing — I didn’t mention my weight loss. When I started taking weight readings two weeks ago, I weighed 225 pounds. Today I weighed 219. Six pounds gone.

All I can say, this is working for me. Consider figuring out how you can incorporate Artificial intelligence into your life. It can be scary for some. I’m already retired, yet I figured it out. You can too. It’s easy. It’s mostly free and so useful.

Do you think you can use artificial intelligence to help manage your chronic health conditions? I’d love to know. If you’re willing to try it, what challenges do you face? Maybe we can get over them together.

In the meantime,Walk Good.Mitch.