VolgendeVolgende gids
AI in fondsenwerving voor non-profitorganisaties en donorbeheer
Industrieën
Gids voor industrieën
AI in diabetes management uses data from continuous glucose monitors, insulin pumps and retinal images to predict blood sugar changes, automate insulin dosing and screen for complications such as diabetic retinopathy.
Luister naar deze gidsAI-verteld · ~3 min
It matters because diabetes requires constant daily decisions, and algorithms can reduce dangerous highs and lows and catch sight-threatening eye disease earlier, especially where specialists are scarce.
Diabetes management revolves around keeping glucose within a target range. Continuous glucose monitors (CGMs), such as those from Dexcom and Abbott's FreeStyle Libre line, measure glucose in the fluid under the skin every few minutes, producing a data stream well suited to prediction. Many CGM systems show trend arrows and predictive alerts that warn of an impending low or high. The biggest application is automated insulin delivery (AID), often called an artificial pancreas. An algorithm reads CGM data and adjusts an insulin pump's delivery. The FDA approved Medtronic's MiniMed 670G, the first hybrid closed-loop system, in 2016. Later systems include Tandem's Control-IQ, Insulet's Omnipod 5 and Beta Bionics' iLet, which starts dosing from body weight rather than traditional pump settings. 'Hybrid' means users generally still announce meals or give meal boluses, because insulin acts more slowly than food raises glucose. A community open-source movement, including OpenAPS and Loop, built do-it-yourself systems before and alongside commercial ones. Most AID algorithms are not deep learning. They are control algorithms, such as model predictive control or proportional-integral-derivative control, often with adaptive features that adjust to a user's needs over time. Machine learning is more prominent in glucose forecasting research and decision-support apps. In eye care, IDx-DR (now called LumineticsCore, from Digital Diagnostics) received FDA De Novo authorization in 2018 as the first autonomous AI diagnostic system, detecting more-than-mild diabetic retinopathy without a clinician interpreting the image. Other systems, including Eyenuk's EyeArt, followed, and Google has studied retinopathy screening in India and Thailand. A common misconception is that closed-loop systems cure diabetes or make it fully automatic. They reduce burden and improve time in range for many users, but still require sensor changes, meal input and attention to device problems.
A person with type 1 diabetes wears a CGM and an insulin pump running a hybrid closed-loop algorithm that increases insulin when glucose is predicted to rise overnight and suspends it when a low is forecast.
A CGM app warns that glucose is predicted to fall below a set level within about 20 minutes, giving the user time to eat carbohydrates before symptoms start.
A primary care clinic photographs patients' retinas with a fundus camera, and autonomous AI software returns a refer-or-rescreen result during the same visit without an eye specialist reading the image.
A diabetes educator reviews a CGM report showing time in range and recurring after-dinner spikes, then adjusts the patient's meal insulin settings.
AID systems are expanding to more users, including some people with type 2 diabetes who use insulin, and research continues toward fully closed-loop control that does not need meal announcements, possibly using faster insulins or additional hormones. Interoperability, where pumps, CGMs and algorithms from different makers can work together under FDA pathways for compatible components, gives users more choice. Retinopathy screening AI may help close gaps in regions with few eye specialists, as long as programs guarantee follow-up care for referred patients. Cost, insurance coverage and access remain major barriers, so benefits may keep reaching better-resourced patients first unless deployment addresses equity directly.
Gratis nieuwsbrief
Drie verifieerde AI-verhalen elke weekdagmorgen, geschreven in eenvoudig Engels. Gratis voor altijd, geen advertenties.
Eén e-mail elke weekdag. Afmelden met één klik. We verkopen of delen je adres nooit.
Test jezelf
Directe feedback op elke antwoord, en een deelbare certificaat met een verifieerbare ID zodra je een cursus hebt afgerond.
Ondersteun gratis AI-onderwijs. AI Understanding is een 501(c)(3) non-profit — geen advertenties, geen betaalmuur, nooit. Doe een donatie
AI in diabetes management uses data from continuous glucose monitors, insulin pumps and retinal images to predict blood sugar changes, automate insulin dosing and screen for complications such as diabetic retinopathy. It matters because diabetes requires constant daily decisions, and algorithms can reduce dangerous highs and lows and catch sight-threatening eye disease earlier, especially where specialists are scarce.
CGM's meten met regelmatige tussenpozen glucose in interstitiële vloeistof, waardoor een gegevensstroom ontstaat die trendpijlen, waarschuwingen en prognoses ondersteunt.
De MiniMed 670G was het eerste hybride gesloten-lussysteem dat door de FDA werd goedgekeurd; de anderen kwamen later.
Omdat insuline langzamer werkt dan voedsel de glucose verhoogt, vertrouwen de meeste systemen nog steeds op de gebruiker om maaltijden aan te kondigen, terwijl achtergrondaanpassingen worden geautomatiseerd.
De meeste AID-systemen vertrouwen op besturingsalgoritmen, vaak met adaptieve functies, in plaats van op diepgaand leren.
IDx-DR, nu LumineticsCore, zou een screeningresultaat kunnen opleveren voor meer dan milde diabetische retinopathie zonder dat een arts het beeld leest.
Blijf leren
Er zijn meer handleidingen voor dit onderwerp geselecteerd
VolgendeVolgende gids
AI in fondsenwerving voor non-profitorganisaties en donorbeheer
Industrieën