Chatgpt for diabetes: the new advisory diabetes

We present the new "Advisory Diabetes", a GPT built with Openai and Chatgpt technology.

It is an initial version that will surprise you, we continue working to improve it, so we hope it will continue to surprise you with the new improvements that we will incorporate !!!

We appreciate that you leave us here your opinions about its operation, remember that we just launched it and that we are still working to improve it, so you are not very hard !!!

to try access to the following link: Link

For those who do not know, I share here a conversation with the "Advisory Diabetes", set you that includes the pages of the forum where the information has obtained, and that can be visited with a click to check the origin of it.

Link

The question: Explain me about glucose monitoring.

The answer:

Glucose monitoring is a crucial component in diabetes management, and there are several available methods and devices that can facilitate this process.For example, Sugarbeat is a glucose meter that adheres to the skin as a sticker and allows glucose to be measured without punctures, which represents great comfort for the user [source] .

Another important concept in glucose monitoring is "time in rank" (IRR), which refers to the percentage of time that glucose measurements are maintained within the limits considered normal (70-180 mg/dl) according to theContinuous glucose monitoring.It is recommended that the objective be to maintain a tir greater than 70% 【source】 .

In addition, there are developments such as the B-My Watch, a development phase bracelet that measures glucose levels through the skin without the need to extract blood, which could represent a significant advance in the comfort and ease of monitoring for the monitoring forpeople with diabetes https://www.diabetesforo.com/discusion/13890/b-my-watch-brazalete-para-monitorizar-glucosa-en-desarrollo#:~:text=%23%20%E3%80%903%E2%80%A0B, CICESE%20Trabajan%20en%20Se%20Project

B-MY Watch: Bracelet for monitoring glucose, in development.
An appropriate monitoring of glucose levels would benefit the quality of life of millions of people suffering from diabetes and students of the National Autonomous University of Mexico (UNAM) and Cici...

As for continuous monitoring devices, such as Medtronic IPRO2, these are useful for detecting fluctuations in glucose levels that are often not captured in standard glycosylated hemoglobin tests and can help identify and solve problems beforebecome complications https://www.diabetesforo.com/discusion/7356/monitor-continuo-de-glucosa-ipro2-de-medtronic#:~:text=,Muy%20%C3%Batil%20Para%20cual any

Continuous IPRO2 glucose monitor of Medtronic
Ibérica Medtronic has presented in Spain its latest advance in technology applied to diabetes management: a new system of continuous glucose monitoring that will allow diabetes patients to carry o...

In addition, A1C hemoglobin is an established method for monitoring patients diagnosed with diabetes, since it does not require fasting for the sample and provides a vision of long -term glucose control 【source】 .

To obtain detailed information about these and other glucose monitoring methods, I recommend visiting the links to the discussions and articles in forum diabetes where these devices and techniques are discussed.This forum is an excellent source of knowledge shared by an active community of people with diabetes and their families.

IMPORTANT: To be able to use the "Assistant Diabetes" it is necessary to have a user with the Plus Plan of Chatgpt, hope that our OpenAi friends have planned to open the new GPT's for all users soon!

I use it to tell the rations of carbohydrates from my meals

I am an ambulatory doctor and I have tried to adjust slow insulin doses in type 2 diabetes and makes it perfect.

EYE.Just handle in case of being a health professional and having notions of AI (minimum knowing how to ask the question properly, that is the key).I do not recommend using at the user level to change insulin doses, but I do drop that I am very amazed.

I put the prompt (significa "la pregunta exacta" que hice) :

"I have a diabetic patient treated with insulin Lantus 8 units at night. He gives me today glycemias of the last week taken early in the day in fasting. They are 189, 220, 210, 250, 245 and 228 mg/dl.

Do I have to make any change in the medication?"

Now the chat gpt response:

"With the morning glychemics that you mean (consistently high above 180 mg/dl), it seems that the dose of basal insulin (lantus) is insufficient to maintain adequate control. It is likely that you need to adjust the dose of Lantus.

Here are some considerations:

1. Progressive increase in Lantus: you could increase the basal insulin dose gradually, perhaps in increases of 2 units every 3-4 days, until fasting glycemia are in the target range (usually between 80-130 mg/dl, depending on the patient and its characteristics).

2. Discard the effect of the Alba or Somogyi phenomenon: Make sure these high glycemia are not due to night bounce hypoglycemia.You can ask you to measure glycemia at 3 am in one day to rule out this.

3. Review the nightfall and exercise: it also evaluates whether you are following an adequate diet before sleeping and if there is any other factor, such as exercise or schedules, which may be influencing.

If after the adjustment there is no improvement, it may be necessary to assess other aspects such as the inclusion of a rapid insulin analog or a more complete review of its treatment.

Do you think adequate gradually increase the dose? "

==========

I want to say today I am trying to elaborate an algorithm to see if Chat GPT "gets wet" in changing insulin guidelines in type 1 diabetes, but it is much more complex, as you will imagine.

Rogerix said:
@ROGERIX said:

I use it to tell the rations of carbohydrates from my meals