Revolutionizing Diabetes Care: Key Insights Every Doctor Should Know

Diabetes is quickly becoming one of the important public health challenges of the 21st century. With its rising prevalence, the significant health complications that come with it, and the substantial economic burden it places on healthcare systems.

As we are aware managing and preventing diabetes is not a straight forward task, especially in traditional face-to-face healthcare settings. One of the major hurdles is early diagnosis most of the people live with diabetes for years without even knowing it. Preventing and identifying the disease early could make a huge difference, but that remains a major challenge.

Another key problem is the complexity of managing diabetes. It requires regular monitoring of blood glucose levels, constant follow-ups to check for complications, and a holistic approach that involves healthcare professionals across multiple specialties, like endocrinologists, podiatrists, nutritionists, nephrologists, and ophthalmologists. This makes it difficult to distribute medical resources effectively, especially in places where access to high-quality care and trained professionals is limited.

Perhaps most challenging is the fact that diabetes management demands active participation from patients. It’s not just about taking medication; it requires ongoing attention to diet, exercise, and regular self-monitoring. With so many aspects of the disease to manage, it can feel overwhelming for patients and healthcare providers alike.

Digital era is evolving creating hope on the horizon. Digital health technologies particularly those powered by artificial intelligence (AI), are showing great promise in tackling some of these challenges. AI-based tools can help in early detection, improve prevention strategies, and manage diabetes more effectively even for patients who can’t make it to a doctor’s clinic regularly. These technologies can offer real-time health updates, provide insights on metabolic health, encourage better self-management opening doors for making diabetes manageable.

Digital health solutions could be the key to alleviating the burden of diabetes, making it easier for people to live healthier lives and ensuring more equitable access to care.

AI to Diagnose Diabetic Retinopathy

AI algorithms can analyse retinal images with remarkable precision, identifying tiny changes in the blood vessels of the retina even before symptoms show up. When these tiny blood vessels start to leak yellow fluid or blood, it’s often an early warning sign of diabetic retinopathy. Sometimes these changes in the retina can occur even before someone has been diagnosed with diabetes.

Early detection of diabetic retinopathy is key to preventing vision loss, and that’s where AI can make a significant impact. By spotting these subtle signs early on, AI can prompt further testing for diabetes, ensuring that people receive the right diagnosis and care before the condition progresses.

In 2024, US FDA granted clearance to AEYE Health Diagnostic Screening technology (AEYE-DS), a fully autonomous AI system that diagnoses referable diabetic retinopathy from retinal images captured by a handheld camera.

AI in Continuous Glucose Monitoring

Continuous glucose monitors are transforming the way we manage diabetes by providing real-time data on glucose levels throughout the day. The true potential is unlocked when this data is paired with AI

AI send alerts if glucose levels go outside the normal range. These real-time notifications helps users to take immediate action, whether it’s adjusting their diet, increasing physical activity, or seeking medical advice. By detecting issues early and providing actionable insights, AI is helping individuals better manage their health and prevent the progression of diabetes.

With AI’s ability to continuously analyse glucose data, patients are empowered with real-time insights about their glucose levels, enabling them to make informed decisions that can help prevent long-term complications. Whether managing diabetes or working to prevent it, AI serves as a powerful tool for gaining better control over glucose levels and improving overall health outcomes.

In 2024, U.S. Food and Drug Administration has approved the first OTC continuous glucose monitor, the Dexcom Stelo Glucose Biosensor System. This integrated CGM (iCGM) is designed for individuals aged 18 and older who do not use insulin, including those managing diabetes with oral medications.

AI to Diagnose Diabetic Foot Ulcers

Diabetic foot ulcers (DFU) are a significant concern for individuals with diabetes, often leading to complications like infections and amputations. Managing DFUs requires early detection, precise monitoring, and effective treatment but traditional methods can be time-consuming and subjective.

Machine learning (ML) is making a profound impact, offering more accurate, faster, and automated solutions. AI can quickly analyse medical images to determine whether a diabetic foot ulcer is present. This automatic detection helps healthcare providers catch potential issues earlier, reducing the risk of complications. Once a wound is detected, ML algorithms can pinpoint the exact location within the image. This makes it easier for healthcare professionals to assess the wound’s size and severity, which is critical for planning treatment.

Beyond simply detecting and localizing the wound, advanced ML models can accurately outline the borders of the ulcer. This precise segmentation is invaluable for tracking healing progress over time and adjusting care strategies as needed.

The FDA granted clearance to an AI-based imaging system developed by DiaMonTech for assessing diabetic foot ulcers.

The Use of AI to Diagnose Diabetic Nephropathy

Managing chronic kidney disease (CKD) is a critical aspect of diabetes care, as diabetes is one of the leading causes of kidney damage. If left unmanaged, it can lead to kidney failure, and eventually, require dialysis, a renal transplant. For years, healthcare providers have used clinical data like serum creatinine levels and albumin-to-creatinine ratios to assess kidney function.

KidneyIntelX received FDA clearance for assessing the risk of CKD in patients with diabetes, including diagnosing and monitoring diabetic nephropathy. This approval marks a significant leap in the use of artificial intelligence in healthcare, offering a more personalized and accurate approach to managing kidney health.

AI combines clinical data with ML algorithms to predict how kidney disease will progress in patients with diabetes. It analyses a variety of important biomarkers, such asserum creatinine levels, albumin-to-creatinine ratio and other key clinical indicators

By examining these data points AI generates a risk score that helps healthcare providers identify patients who are at high risk for worsening kidney function, including those who may develop diabetic nephropathy. It can track disease progression, offering valuable insights that allow for timely interventions and more effective management strategies.

Artificial intelligence is making significant impact in the world of diabetes care, especially when it comes to predicting and diagnosing complications like retinopathy, foot ulcers, nephropathy, and even continuous glucose monitoring. By implementing AI-powered algorithms, healthcare providers can detect these complications earlier, allowing for more proactive and personalized treatment. As these technologies continue to evolve, AI is expected to play an increasingly important role in diagnosing and managing diabetic complications, offering a future where early intervention and better outcomes are the norm for patients.

Reference

1. Guan Z, Li H, Liu R, Cai C, Liu Y, Li J, Wang X, Huang S, Wu L, Liu D, Yu S, Wang Z, Shu J, Hou X, Yang X, Jia W, Sheng B. Artificial intelligence in diabetes management: Advancements, opportunities, and challenges. Cell Rep Med. 2023 Oct 17;4(10):101213.
2. Khalifa M, Albadawy M. Artificial intelligence for diabetes: Enhancing prevention, diagnosis, and effective management. Computer Methods and Programs in Biomedicine Update. 2024 Feb 12:100141.
3. Huang J, Yeung AM, Armstrong DG, Battarbee AN, Cuadros J, Espinoza JC, Kleinberg S, Mathioudakis N, Swerdlow MA, Klonoff DC. Artificial Intelligence for Predicting and Diagnosing Complications of Diabetes. J Diabetes Sci Technol. 2023 Jan;17(1):224-238.
4. PR Newswire. FDA clears first fully autonomous AI for portable diabetic retinopathy screening. PR Newswire; 2024 Apr 30. Available from: https://www.prnewswire.com/news-releases/fda-clears-first-fully-autonomous-ai-for-portable-diabetic-retinopathy-screening-302131559.html
5. U.S. Food and Drug Administration. FDA clears first over-the-counter continuous glucose monitor. FDA; 2024 Apr 18. Available from: https://www.fda.gov/news-events/press-announcements/fda-clears-first-over-counter-continuous-glucose-monitor