Skip to content
type1.science

Concept: closed loops that act on ketones

What we should build — the sensor half now exists (Abbott's Libre Duo, the first dual glucose-and-ketone wearable, CE-marked in Europe in May 2026), but no automated insulin delivery system reads the ketone channel and no standard says how one should respond. Until an algorithm can see ketones climbing and act, a dual sensor is a warning light nobody is wired to.

What we should buildEarly evidencecgmketone-sensingconcept

The scorecard

Interoperability5

Zero, and that is the point of this entry: no commercial closed loop reads a ketone channel today, and no standard defines how one should respond when ketones rise. Sensor makers are racing to add analytes; no algorithm has been cleared to act on them.[4]

Ketone sensing95

No longer hypothetical — the Libre Duo measures interstitial beta-hydroxybutyrate every minute and is CE-marked in Europe. The hardware answer to today's biggest sensing gap exists.[2]

Alerts & prediction88

Continuous ketone alerts address one of pump therapy's most dangerous failure modes. But an alert asks a person to act; the concept here is a loop that acts on its own.[1]

Accuracy80

Glucose accuracy rides on the well-established Libre 3 platform the Duo is built on; a published, peer-reviewed accuracy figure for the ketone channel is not out yet, which is exactly the number an algorithm would need before it is allowed to dose on it.[2]

No calibration90

Factory-calibrated on the Libre 3 platform — no routine fingersticks, so the ketone signal arrives without extra user work.[2]

Low lag65

Interstitial lag is unchanged; the value of this concept is a new analyte and a loop that uses it, not a faster glucose reading.[2]

Editor’s take

We used to rank this as the most important sensing upgrade on the roadmap. Abbott has now built the sensor — so the argument moves one layer up. Ketone blindness is why a kinked cannula can escalate to the ER, and a continuous ketone channel turns that from emergency into early warning only if something is listening. Today nothing is: no closed loop consumes ketones, and no standard defines what it should do when they climb. That algorithm, not the sensor, is now the thing we would fund.

The full picture

This entry used to argue for a sensor. That argument has been won: Abbott's Libre Duo reads glucose and ketones from one wearable, every minute, and took a CE Mark in Europe in May 2026 (US review still pending).12 So the concept moves up a layer, to the half that nobody has built.

The remaining gap: nothing is listening. When an infusion site fails or illness strikes, ketones (beta-hydroxybutyrate) can climb toward diabetic ketoacidosis — sometimes while glucose still looks reassuring. A wearable ketone monitor has caught exactly that: sustained ketosis above 1.0 mmol/L for 14 hours while CGM showed normal glucose.3 Today the best a dual sensor can do is tell you. No commercial automated insulin delivery (AID) system reads a ketone channel, and no standard defines what a loop should do when it sees one rising — increase basal, prompt an injection and a site change, refuse to trust the pump, or simply escalate loudly.4 A warning light is only worth as much as the thing wired to it.

What we want built. A loop that treats ketones as a first-class input: it notices ketones rising while glucose creeps up, infers "this insulin is not getting in", stops quietly compensating, and tells you to change the site and inject by pen — the reasoning a person does today, done automatically and hours earlier. The prerequisites are a published accuracy figure for the ketone channel (you cannot let an algorithm act on a signal whose error bars are unpublished), an agreed action policy, and a regulator willing to clear it. None of those exist yet.

Who is actually building the sensing side. Abbott shipped first with the Libre Duo.12 PercuSense is starting two Helmsley-funded type 1 diabetes studies in summer 2026, measuring glucose, lactate, ketones and tissue oxygen from a single probe.4 Biolinq is adding lactate, with ketones and cortisol under consideration.4 Dexcom has said it will ship potassium first (around 2027-2028) and ketones after.4 Sequel has a public agreement to connect its twiist AID system to Abbott's dual sensor once it is US-cleared — the closest anyone has come to closing this loop, and still an agreement rather than a product, with no published rule for what the loop would actually do about a rising ketone.5 The multi-analyte hardware race is real and crowded; the algorithm race has not started.

An international expert consensus called for precisely this class of sensor, noting how poorly fingerstick ketone testing is actually done and how SGLT2 inhibitors raise the risk of ketosis at normal glucose levels.6 A McGill trial is now studying continuous ketone monitoring in people with type 1 diabetes using SGLT2 inhibitors.7 We keep this as a concept to hold the unbuilt half visible — the sensing is arriving, the intelligence is not (see the algorithm-intelligence gap).

References

  1. Abbott. Abbott secures CE Mark for world's first dual glucose-ketone sensing technology for people with diabetes. Abbott Media Room (2026). https://abbott.mediaroom.com/2026-05-27-Abbott-secures-CE-Mark-for-worlds-first-dual-glucose-ketone-sensing-technology-for-people-with-diabetes ↩ ↩2

  2. Reuter E. Abbott receives CE mark for dual glucose-ketone sensor. MedTech Dive (2026). https://www.medtechdive.com/news/abbott-receives-ce-mark-for-dual-glucose-ketone-sensor/821296/ ↩ ↩2

  3. Continuous Ketone Monitoring Captures Overt Ketosis in Type 1 Diabetes Caused by Empagliflozin, Low-Carbohydrate Diet, and Exercise. Journal of Diabetes Science and Technology (2026). https://doi.org/10.1177/15209156261420167 ↩

  4. MedTech Dive. 4 diabetes tech execs on the future of multi-analyte glucose sensors (2026). https://www.medtechdive.com/news/4-diabetes-tech-execs-on-the-future-of-multi-analyte-glucose-sensors/823401/ ↩ ↩2 ↩3 ↩4

  5. Sequel Med Tech. Sequel Med Tech to Integrate twiist Automated Insulin Delivery (AID) System with Abbott's Future Dual Glucose-Ketone Sensor. GlobeNewswire (2025). https://www.globenewswire.com/news-release/2025/05/22/3086535/0/en/Sequel-Med-Tech-to-Integrate-twiist-Automated-Insulin-Delivery-AID-System-with-Abbott-s-Future-Dual-Glucose-Ketone-Sensor.html ↩

  6. Virdi N, Poon Y, Abaniel R, Bergenstal RM. Prevalence, Cost, and Burden of Diabetic Ketoacidosis. Diabetes Technol Ther (2023). https://doi.org/10.1089/dia.2023.0149 ↩

  7. Continuous Ketone Monitoring in Participants With Type 1 Diabetes Using SGLT2 Inhibitors as Adjunctive Therapy (NCT06753994). ClinicalTrials.gov. https://clinicaltrials.gov/study/NCT06753994 ↩

Coming soon

ETA · The sensor is real: Libre Duo is CE-marked in Europe (May 2026), with US FDA review pending. A closed loop that acts on the ketone channel has no timeline.

  • →Sensor half delivered — Abbott's Libre Duo took the CE Mark in May 2026 and measures glucose and ketones every minute; US FDA clearance still pending
  • →More analytes coming — PercuSense begins two Helmsley-funded type 1 studies in summer 2026 sensing glucose, lactate, ketones and tissue oxygen on one probe; Dexcom plans potassium first and ketones after
  • →Still missing — no automated insulin delivery system consumes a ketone channel, and no standard defines how a loop should respond to rising ketones

Sources

  1. [1]Prevalence, Cost, and Burden of Diabetic Ketoacidosis · peer-reviewed — Reports an international expert consensus recommending a continuous ketone monitoring system, ideally one combining CGM with 3-β-OHB in a single sensor; also documents poor adherence to fingerstick ketone testing and euglycemic DKA risk with SGLT2 inhibitors.
  2. [2]Abbott secures CE Mark for world's first dual glucose-ketone sensing technology for people with diabetes · manufacturer — Libre Duo measures glucose and ketones every minute for early DKA warning; CE-marked May 2026, not yet cleared in the US.
  3. [3]Abbott receives CE mark for dual glucose-ketone sensor · news — MedTech Dive, 27 May 2026 — independent confirmation of the CE mark, wear and age specs, phased EU launch, and pending US FDA review.
  4. [4]4 diabetes tech execs on the future of multi-analyte glucose sensors · news · 2026-06-22 — MedTech Dive, 22 June 2026 — PercuSense is starting two Helmsley-funded type 1 diabetes studies in summer 2026 measuring glucose, lactate, ketones and tissue oxygen on a single probe; Biolinq is adding lactate with ketones and cortisol under consideration; Dexcom plans potassium first (2027-2028) and ketones after.
  5. [5]Sequel Med Tech to Integrate twiist Automated Insulin Delivery (AID) System with Abbott's Future Dual Glucose-Ketone Sensor · manufacturer — Sequel/GlobeNewswire, 22 May 2025 — an agreement to integrate the twiist AID system with Abbott's dual sensor once US-cleared. An intention, not a shipping product; no published policy for how the loop would act on ketones.
  6. [6]Continuous Ketone Monitoring Captures Overt Ketosis in Type 1 Diabetes Caused by Empagliflozin, Low-Carbohydrate Diet, and Exercise · peer-reviewed — Wearable continuous ketone monitor detected sustained ketosis (>1.0 mmol/L for 14 h) while CGM showed euglycemia — exactly the blind spot glucose-only sensing misses.
  7. [7]Continuous Ketone Monitoring in Participants With Type 1 Diabetes Using SGLT2 Inhibitors as Adjunctive Therapy (NCT06753994) · registry — McGill University trial evaluating wearable continuous ketone monitoring in T1D.