Stanford Medicine researchers are transforming the way type 2 diabetes is managed with a groundbreaking voice-based artificial intelligence application.
This new tool, designed for the ubiquitous “smart speaker,” simplifies insulin dosage without the need for constant medical supervision, a boon for patients who struggle with technology or lack easy access to healthcare professionals.
Type 2 diabetes, a condition affecting blood sugar levels, requires patients to regularly monitor their blood glucose and adjust their insulin intake accordingly.
Traditionally, this involves using a glucometer at home and then communicating the results to a healthcare team through an app, a web portal, or by phone.
However, this process can be daunting, especially for those not comfortable with technology or the mathematics involved in adjusting insulin doses.
Recognizing these challenges, Ashwin Nayak, MD, a clinical assistant professor of medicine at Stanford and the first author of the study published in JAMA Network Open, spearheaded the development of this AI application.
The system is uniquely user-friendly, leveraging the familiarity and simplicity of interacting with a smart speaker.
In a randomized trial involving 32 participants with type 2 diabetes, half were provided with smart speakers equipped with this AI software. This group showed remarkably better control of their blood glucose levels compared to the control group, who received standard care.
Not only did they reach optimal blood sugar levels more quickly, but they also adhered better to their insulin schedules and experienced less emotional distress related to diabetes management.
Unlike type 1 diabetics who often use continuous glucose monitors or insulin pumps, type 2 diabetes patients usually have less consistent insulin needs due to insulin resistance rather than insulin deficiency. This inconsistency makes managing their condition more challenging.
The AI-enabled smart speaker offers a simple yet effective solution. Participants using this system checked in daily with a prompt like “Check in with clinical trial,” followed by reporting their insulin usage and blood glucose readings.
The AI then analyzed this data and provided immediate feedback and dosage recommendations, such as maintaining or adjusting the insulin dose.
This approach not only facilitates remote monitoring but also enables remote patient intervention, a significant advancement in diabetes care.
Participants in the standard care group used their smart speakers for reminders but did not have access to the AI app. They followed their usual routine of reporting blood glucose levels to their healthcare providers, waiting for adjustments in their insulin dosages.
The results of the study were striking. Over an eight-week period, 81% of the AI group achieved control of their diabetes, compared to only 25% in the traditional care group.
Additionally, those managed by the AI required fewer doctor’s appointments and had more frequent insulin dose adjustments, leading to better diabetes control.
This voice-activated AI tool holds immense potential, particularly for older patients with type 2 diabetes. It simplifies diabetes management, making it more accessible and convenient.
Kevin Schulman, MD, professor of medicine at Stanford and leader of the study, emphasized the need for technology to meet patients where they are, especially given the complexity of diabetes treatments.
Currently programmed in English, the system can be adapted for other languages and is capable of monitoring responses to various diabetes medications, not just insulin.
The researchers are optimistic about its potential to manage other chronic diseases as well, signaling a new era in patient-centered, technology-assisted healthcare.
If you care about diabetes, please read studies about new way to achieve type 2 diabetes remission, and one avocado a day keeps diabetes at bay.
For more information about diabetes, please see recent studies about 5 dangerous signs you have diabetes-related eye disease, and results showing why pomegranate is super fruit for people with diabetes.
The research findings can be found in JAMA Network Open.
Copyright © 2024 Knowridge Science Report. All rights reserved.
