News & Resources

How is AI helping those with diabetes

In Consulting Services, Thought Leadership, Indirect Resellers, Direct Customers

Over 400 million people have diabetes but as many as 46% remain completely undiagnosed and untreated. In fact, diabetes accounts for 12% of the world’s health expenditure through the vast amount of research, analysis and treatments, conducted and required today. Artificial intelligence (AI) and its applications such as Machine Learning (ML) are promising significant breakthroughs in diabetes care. AI advances in the field relevant to those with diabetes, those who remain undiagnosed, their families and caregivers.

 

The problem

Years of research has gone into finding effective therapies that can delay the onset and slow the progression of diabetes-related complications. Studies have found that despite the associated risks such as kidney failure and the loss of limbs, a staggering 31% of patients discontinue their diabetes medication after the first three months (Dankwa-Mullan et all 2019).

Caring for diabetics has traditionally been hampered by a lack of real-time data and knowledge to make informed decisions. We know there is a huge amount of literature in the field, and that people generate over 1 million gigabytes of health-related data in a lifetime. However much of this health data is unstructured, including the essential notes from doctors and nurses, images and more recently IoT data.

Advances in AI have made unstructured information more accessible to experts in the field, allowing them to collate meaningful data. ML algorithms alongside methods like computer vision, can convert this data into useful and insightful information that can help make the decision process more efficient or even immediate. 

How can AI be used in diabetes care?

There are three primary objectives for using AI in diabetes care.

1.      AI can see patterns in behaviors and build its own logic

2.      AI can help with early diagnosis of diabetes

3.      AI can provide personalized healthcare recommendations

We will look at each of these separately to show why breakthroughs in AI could be life-changing for so many people.

AI is used to spot patterns in behavior that lead to either high or low blood sugar levels in diabetes patients. Continuous glucose monitors used by those with diabetes collect a huge amount of data that has previously not been used efficiently.

Those with Type 1 diabetes need to test blood sugar levels multiple times a day to determine how much insulin they should inject. Until recently, the only way of doing this was using a finger stick needle to obtain a blood sample. The continuous glucose monitor doesn’t require finger sticks. It is a small device under the skin that can simply be scanned using a smartphone app.

The device can change patiences’ lives and enables healthcare providers with the opportunity to analyze real-time data. The monitor automatically tells patients which direction their levels are trending to ensure they act appropriately. It can even predict the likely outcome of meals on their sugar levels.

The technology is still in relatively early stages but as more data is acquired over the years, it could change the ways diabetes patients handle their medication, to better manage their condition and ultimately giving them a higher quality of life.

Early detection

Many of us use wearable technology such as a FitBit or Samsung Health to track our steps, sleep patterns and calories. These devices have the potential to revolutionize how we diagnose diabetes. With the amount of data collected on our health it is clear to see how they might be able to help prevent ailments such as Type 2 Diabetes. For example, they could send alerts if your data showed signs that matched the typical attributes of Type 2 diabetes, even if you don’t yet have it. They can track your stats to inform you of any prevention methods, such as measuring sugar content in your daily meals and advising how to eat healthy.

Fitbit is one of the early adopters that have committed to various health ventures and investments specifically in the last two or three years. The projects are particularly aimed at providing better health management for those with diabetes. Fitbit is also working with third parties to help you see blood glucose readings on your device, where data is synced from the glucose monitor straight to your watch.

AI is being used to check for the early detection of diabetes related conditions such as diabetic retinopathy which is a condition that leads to a loss of vision. The AI software can determine if the patient needs to be referred to an ophthalmologist right away. These technologies, with their early detection capabilities can prevent an otherwise un-treatable condition.

AI in diabetes care

Some people with Type 2 diabetes find it tough to maintain a balanced lifestyle and are most at risk from straying from their diabetes management plans. A common misconception of diabetes is that it is all about ‘managing sugar intake’ but those with diabetes need to have a very well-balanced diet to properly mitigate any of the risks.

Food apps such as Ascensia Diabetes Care have been designed to create personalized diet plans specifically for diabetics. In the app, users can create personalized food choices to set up a plan that suits them rather than being told what they need to eat. The aim is to help them make smarter diet choices without being overly restrictive.

If people can take it upon themselves to manage their diabetes with the help of AI, we will see a reduction in appointments at the doctor’s and a reduction in medical interventions.

AI enabling you to make informed decisions

One of the biggest problems facing healthcare today, is gaining access to essential data to help make informed decisions efficiently, and also accessing data that is backed by vast real-life data samples. The solutions proposed in this article are designed to start bringing data to the patients directly and helping to create awareness of how they can proactively help in prevention and cure of Diabetes related conditions. The likes of continuous glucose monitors, wearable technology and smartphone apps empower people with the crucial information to take control of their diabetes.

Although today AI cannot completely prevent diabetes, in years to come with enough data collated, we could find a cure.

Crayon continues to invest to be in the forefront of technological development, with our dedicated in-house research teams. Our collaborations within the healthcare sector, allow us to work shoulder to shoulder with our customers across the globe. We have helped in identifying data collection points, data analysis and technology suggestions based from the data outcomes. With our expertise in machine learning as well as IoT we are working on projects to help customers embrace their end-to-end AI journey. If you’re looking for guidance in how emerging technologies can provide true business value, contact one of our experts today.

Erling Hesselberg - VP at Crayon