“Information technology and business are becoming inextricably interwoven. I don’t think anybody can talk meaningfully about one without the talking about the other” – Bill Gates wrote this in 1999 and, nowadays, the same can certainly be said about information technology and healthcare.
The ability to collect, analyze, and use data is becoming ubiquitous with many of healthcare’s most exciting advancements. Whether it’s Google’s DeepMind using artificial intelligence (AI) to help doctors predict the onset of life-threatening diseases or companies like Amelieff using bioinformatics for genetic testing, it looks like data could be the key to answering some of healthcare’s most enduring questions.
I believe that machine learning, in particular, is an important aspect of future diabetes management and healthcare more generally. Machine learning is a type of AI that is programmed to automatically learn and improve through its own experience. To some this may sound like a distant idea from a science fiction movie but, today, machine learning is already sweeping through our world and it has great potential to improve our lives.
In a recent report, Deloitte and MedTech Europe estimated that applications of AI in healthcare could result in annual savings of €200 billion ($224 billion) and 1.8 billion staff hours in Europe alone. More tangibly, the report suggests that 400,000 European lives could be saved every year as a result – a staggering figure that emphasizes AI’s potential.
Machine learning has multiple purposes in healthcare and can be used wherever data is generated and utilized, such as in clinical trials, diagnosis and even in managing chronic conditions such as diabetes. In theory, machine learning brings automation, personalization and simplicity to aspects of healthcare which are crying out for just that.
Diabetes is almost always a life-long condition where continuous support and monitoring are required, making it an ideal candidate for a data-driven approach to management. With 463 million people estimated to have diabetes worldwide, it is no surprise that the condition sits high on the global health agenda. Given its prevalence, any improvements for people with diabetes (PWDs) afforded through AI and machine learning will have a huge impact on the lives of millions.
We are seeing this in action across the entire PWD journey – from prediction and prevention to monitoring and management – so how could machine learning revolutionize diabetes?
Prediction, prevention and early diagnosis
One of the keys to combatting the impacts of chronic conditions, and disease more generally, is early diagnosis. AI and its subfield of machine learning is starting to facilitate this, along with the prediction of risk, earlier than ever before, whether it be through its application to radiology, genomics or electronic health data.
Earlier diagnosis could vastly improve outcomes in type 1 diabetes (T1D) as it would allow people to understand their body and learn how they can manage their condition from an earlier age and, ultimately, better avoid associated health complications throughout life. For type 2 diabetes (T2D) the prospect of identifying those at most risk earlier than we are currently able to is incredibly tantalizing as it may help to motivate people to work towards preventing its onset through behavioral change.
In addition, predictive analysis afforded through AI and machine learning can help PWDs to identify whether they have higher risk for developing comorbidities and complications, such as cardiovascular disease or neuropathy, before they emerge or progress. For example, companies have previously investigated how machine learning could be used to automate screenings for diabetic retinopathy, which is recognized to be one of the leading causes of blindness in working-age adults (20-65 years). Hopefully, using data in this way will help us connect the dots between various conditions and healthcare settings, creating a more holistic healthcare ecosystem than we have today. This is particularly important as touch points between PWDs and HCPs evolve to facilitate care more virtually, with telehealth and healthcare portals serving as key opportunities to transfer data for remote patient monitoring and electronic medical records.
Diabetes management
Daily diabetes management lends itself well to AI, as it largely involves recognizing patterns in PWDs, whether it be in behavior or blood sugars. This is especially true as PWDs increasingly turn to digital tools such as apps to help them manage their condition.
For glucose management, in particular, machine learning can help with automating insulin needs or bolus calculations based on data from continuous glucose monitors (CGM), systems that provide continuous measurement of blood sugar.
DreaMed Diabetes, for example, uses cloud-based analytics software to provide recommendations on optimal insulin doses, getting more proficient at doing so as data accumulates. Furthermore, algorithms across a range of apps and platforms, including our CONTOUR®Diabetes App, can harness data to learn about a person’s habits and help identify patterns in their glucose profile.
Personalization and usability
Another way to improve the diabetes journey is to tailor the tools available to the needs of individual PWDs. In terms of both what diabetes management tools deliver and how they are delivered, it’s certainly not one size fits all. There are countless considerations for the treatment of PWDs, ranging from the infrastructure available and behavior of people in a particular geography, to the preferences and needs of the individual.
Machine learning and AI can help individualize treatments by using data to provide PWDs with personalized regimes and coaching. The information output can then be evolved and adapted automatically, with algorithms continuously pulling on the data to learn about PWDs.
A great example of this is Virta Health’s individualized nutrition therapy for people with T2D. On this platform, machine learning is used to provide PWDs with recipe recommendations and nutrition tips based on the patterns they have recorded in the Virta app, with the goal of minimizing dependence on diabetes medication.
Furthermore, usability data from apps and other mobile health tools can help developers work out how to design their products to be more helpful and effective. As we move forward, it is important to capitalize on the data we have at our disposal to simplify how we deliver services and solutions. For example, the use of algorithms to better understand the behavior and patterns of consumers, can allow us to pre-empt their needs and make their lives as easy and uninterrupted as possible. Simplicity is key in the adoption of technology and, therefore, key in unlocking the potential of AI.
Across the healthcare landscape we are starting to see the acceleration of improvements in how people are treated due to better use of data, and this can only be a good thing. With the number of PWDs across the globe in the millions, the data pool doesn’t get much bigger and there are undoubtedly so many crucial learnings to be discovered from within it.
The automation afforded through AI and machine learning means that this huge data set can be processed and subsequently utilized at speeds that we have never seen before, with data sharing potentially optimizing these insights even further. We are excited to see how this data-driven future of diabetes can further empower people living with the condition and arm them with the tools to achieve the best possible health outcomes.
But diabetes is just one piece of the puzzle. Data is omnipresent in healthcare and, as we harvest more of it, our AI and machine learning capabilities will be able to grow exponentially. AI has the potential to facilitate personalized treatment approaches across multiple conditions, enable efficient healthcare delivery, and lead to a more connected healthcare ecosystem. We are confident that by embracing the data revolution, health outcomes can be dramatically improved, in diabetes and beyond.
About Athar Siddiqui: Global Strategy Lead, Digital Connected Solutions – Ascensia Diabetes Care
Athar Siddiqui is a forward-thinking strategic leader, with a focus on innovating digitally connected solutions in healthcare. Athar is currently Ascensia Diabetes Care’s Global Strategy Lead for Digital Connected Solutions, bringing over 15 years’ progressive experience as a decision-maker and leader. He champions the development and commercialization of customer-centred digital healthcare solutions and is integral part of the team behind Ascensia’s CONTOUR®DIABETES App