From Artificial Intelligence to Smart Pens: Technological Advancements in Diabetes Care
While there are many new innovations on the horizon, these three areas seem particularly ripe for this technological revolution.
Technological advancements in the past decade have percolated into diabetes care and the forthcoming decade promises to totally change the face of diabetes management. While there are many new innovations on the horizon, these three areas seem particularly ripe for this technological revolution.
After the advent of smart phones and smart watches, a new addition to the smart lexicon is the smart insulin pen. Simply put, a smart pen means an insulin pen that can record the amount and timing of each insulin dose administered by it and transmit the information wirelessly via Bluetooth to a linked mobile app.
Presently, insulin users face three main impediments to have a better control of their diabetes – skipped and forgotten insulin doses, inadequate increases in insulin doses and risk of hypoglycemia (low blood glucose levels). By creating an interface through which the pen can keep a track of insulin doses, store the data in an app and share the data with users, care givers and health care professionals, the smart pen seeks to create a diabetes care ecosystem that can track not only the doses of insulin, but also link up with the glucose sensor and suggest appropriate doses of insulin based on blood glucose levels – in effect, a low cost insulin pump.
While this sounds very attractive in theory, there are a few practical barriers that need to be surmounted. Like all new medical innovations, smart pens will have to demonstrate their utility in real-world settings outside the lab, fidelity in blood sugar reporting and a cost benefit.
A small pen-sized device, insulin pumps replace the need for frequent injections. The next big push in insulin pumps in the next couple of years is going to be in the areas of patch pumps.
In addition to convenience, another advantage with these wearable pumps has been the decrease in total daily insulin dose after prolonged use of these pumps for more than 3-4 months.
For the past few years, artificial intelligence has been touted to be the game changer in diabetes care, but somehow AI has not lived up to its initial promise. Lack of actionable data, high cost of AI and the lack of a cogent association between the AI developers and the current medical stakeholders in diabetes care have hamstrung the development of an AI framework that can supplement or even replace health care providers in management of diabetes.
But recent trends indicate that diabetes AI is on the cusp of a major revolution in diabetes care. The main areas of thrust and utility for AI in the near future will be in – dose adjustment of insulin pumps – especially pumps that are linked to a glucose sensor to manage fluctuations in blood glucose levels, earlier diagnosis of diabetes and diabetes related complications, smarter algorithms to predict high and low blood glucose levels and their association with a particular meal of even particular foods.
The adoption of many new technologies for diabetes has been slower than their developers had envisaged, classic examples being, insulin pumps (as opposed to daily injections), continuous glucose monitors (as opposed to glucometer monitoring), and insulin dosing mobile apps (as opposed to pen-and-paper charts). Technology inertia might also be a barrier, especially in older people, but the cost of these technologies also needs to be addressed before one can expect them to be the norm rather than the exception in diabetes management.