Research Article
No access
Published Online: 30 May 2024

Neural-Net Artificial Pancreas: A Randomized Crossover Trial of a First-in-Class Automated Insulin Delivery Algorithm

Publication: Diabetes Technology & Therapeutics
Volume 26, Issue Number 6


Background: Automated insulin delivery (AID) is now integral to the clinical practice of type 1 diabetes (T1D). The objective of this pilot-feasibility study was to introduce a new regulatory and clinical paradigm—a Neural-Net Artificial Pancreas (NAP)—an encoding of an AID algorithm into a neural network that approximates its action and assess NAP versus the original AID algorithm.
Methods: The University of Virginia Model-Predictive Control (UMPC) algorithm was encoded into a neural network, creating its NAP approximation. Seventeen AID users with T1D were recruited and 15 participated in two consecutive 20-h hotel sessions, receiving in random order either NAP or UMPC. Their demographic characteristics were ages 22–68 years old, duration of diabetes 7–58 years, gender 10/5 female/male, White Non-Hispanic/Black 13/2, and baseline glycated hemoglobin 5.4%–8.1%.
Results: The time-in-range (TIR) difference between NAP and UMPC, adjusted for entry glucose level, was 1 percentage point, with absolute TIR values of 86% (NAP) and 87% (UMPC). The two algorithms achieved similar times <70 mg/dL of 2.0% versus 1.8% and coefficients of variation of 29.3% (NAP) versus 29.1 (UMPC)%. Under identical inputs, the average absolute insulin-recommendation difference was 0.031 U/h. There were no serious adverse events on either controller. NAP had sixfold lower computational demands than UMPC.
Conclusion: In a randomized crossover study, a neural-network encoding of a complex model-predictive control algorithm demonstrated similar performance, at a fraction of the computational demands. Regulatory and clinical doors are therefore open for contemporary machine-learning methods to enter the AID field.
Clinical Trial Registration number: NCT05876273.

Get full access to this article

View all available purchase options and get full access to this article.


1. Brown SA, Kovatchev BP, Raghinaru D, et al.; for the iDCL Trial Research Group. 6-month randomized multicenter trial of closed-loop control in type 1 diabetes. N Engl J Med 2019;381:1707–1717.
2. Breton MD, Kanapka LG, Beck RW, et al.; for the iDCL Trial Research Group. A randomized trial of closed-loop control in children with type 1 diabetes. N Engl J Med 2020;383:836–845.
3. Burnside MJ, Lewis DM, Crocket HR, et al. Open-source automated insulin delivery in type 1 diabetes. N Engl J Med 2022;387(10):869–881;
4. Wadwa RP, Reed ZW, Buckingham BA, et al. Trial of hybrid closed-loop control in young children with type 1 diabetes. N Engl J Med 2023;388(11):991–1001;
5. Bionic Pancreas Research Group; Russell SJ, Beck RW, Damiano ER, et al. Multicenter, randomized trial of a bionic pancreas in type 1 diabetes. N Engl J Med 2022;387(13):1161–1172;
6. Boughton CK, Allen JM, Ware J, et al. Closed-loop therapy and preservation of C-peptide secretion in type 1 diabetes. N Engl J Med 2022;387(10):882–893;
7. Kovatchev BP, Singh H, Mueller L, et al. Biobehavioral changes following transition to automated insulin delivery: A large real-life database analysis. Diabetes Care 2022;45(11):2636–2643;
8. Arrieta A, Battelino T, Scaramuzza AE, et al. Comparison of MiniMed 780G system performance in users aged younger and older than 15 years: Evidence from 12 870 real-world users. Diabetes Obes Metab 2022;24(7):1370–1379;
9. Forlenza GP, Carlson AL, Galindo RJ, et al. Real-world evidence supporting tandem control-IQ hybrid closed-loop success in the medicare and medicaid type 1 and type 2 diabetes populations. Diabetes Technol Ther 2022;24(11):814–823;
10. Benhamou PY, Adenis A, Lebbad H, et al. One-year real-world performance of the DBLG1 closed-loop system: Data from 3706 adult users with type 1 diabetes in Germany. Diabetes Obes Metab 2023;25(6):1607–1613;
11. Matejko B, Juza A, Kieć-Wilk B, et al. One-year follow-up of advance hybrid closed-loop system in adults with type 1 diabetes previously naive to diabetes technology: The effect of switching to a calibration-free sensor. Diabetes Technol Ther 2023;25(8):554–558;
12. Lombardo F, Passanisi S, Alibrandi A, et al. MiniMed 780G six-month use in children and adolescents with type 1 diabetes: Clinical targets and predictors of optimal glucose control. Diabetes Technol Ther 2023;25(6):404–413;
13. Wang XS, Dunlop AD, McKeen JA, et al. Real-world use of Control-IQ™ technology automated insulin delivery in pregnancy: A case series with qualitative interviews. Diabet Med 2023;40(6):e15086;
14. Phillip M, Nimri R, Bergenstal RM, et al. Consensus recommendations for the use of automated insulin delivery technologies in clinical practice. Endocr Rev 2023;44(2):254–280;
15. Doyle III FJ, Huyett LM, Lee JB, et al. Closed-loop artificial pancreas systems: Engineering the algorithms. Diabetes Care 2014;37:1191–1197.
16. Hovorka R, Chassin LJ, Wilinska ME, et al. Closing the loop: The ADICOL experience. Diabetes Technol Ther 2004;6:307–318.
17. Hovorka R, Canonico V, Chassin LJ, et al. Nonlinear model predictive control of glucose concentration in subjects with type 1 diabetes. Physiol Meas 2004;25:905–920.
18. Steil GM, Rebrin K, Darwin C, et al. Feasibility of automating insulin delivery for the treatment of Type 1 Diabetes. Diabetes 2006;55:3344–3350.
19. Kovatchev BP, Patek SD, Dassau E, et al. Control-to-range for diabetes: Functionality and modular architecture. J Diabetes Sci Technol 2009;3:1058–1065.
20. Grosman B, Dassau E, Zisser HC, et al. Zone model predictive control: A strategy to minimize hyper- and hypoglycemic events. J Diabetes Sci Technol 2010;4:961–975.
21. Atlas E, Nimri R, Miller S, et al. MD-logic artificial pancreas system: A pilot study in adults with type 1 diabetes. Diabetes Care 2010;33(5):1072–1076;
22. Collyns OJ, Meier RA, Betts ZL, et al. Improved glycemic outcomes with Medtronic Minimed advanced hybrid closed-loop delivery: Results from a randomized crossover trial comparing automated insulin delivery with predictive low glucose suspend in people with type 1 diabetes. Diabetes Care 2021;44(4):969–975;
23. Bergenstal RM, Nimri R, Beck RW, et al.; for Group FS. A comparison of two hybrid closed-loop systems in adolescents and young adults with type 1 diabetes (FLAIR): A multicentre, randomised, crossover trial. Lancet 2021;397:208–219;
24. El-Khatib FH, Russell SJ, Nathan DM, et al. A bihormonal closed-loop artificial pancreas for type 1 diabetes. Science Trans Med 2010;2:27ra27.
25. Castellanos LE, Balliro CA, Sherwood JS, et al. Performance of the insulin-only iLet bionic pancreas and the bihormonal iLet using dasiglucagon in adults with type 1 diabetes in a home-use setting. Diabetes Care 2021;44(6):e118–e120;
26. Castillo A, Villa-Tamayo MF, Pryor E, et al. Deep Neural Network Architectures for an Embedded MPC Implementation: Application to an Automated Insulin Delivery System. IFAC-PapersOnLine 2023;56(2):11521–11526.
27. Garcia-Tirado J, Diaz JL, Esquivel-Zuniga R, et al. Advanced closed-loop control system improves postprandial glycemic control compared with a hybrid closed-loop system following unannounced meal. Diabetes Care 2021:dc210932;
28. Place J, Robert A, Ben Brahim N, et al. DiAs web monitoring: A real-time remote monitoring system designed for artificial pancreas outpatient trials. J Diabetes Sci Technol 2013;7:1427–1435.
29. Kovatchev BP, Keith-Hynes PT, Breton MD, et al. Unified Platform For Monitoring and Control of Blood Glucose Levels in Diabetic Patients. U.S. Patent No. 10,610,154 Granted 2020. Available from:
30. Cobelli C, Renard E, Kovatchev BP, et al. Pilot studies of wearable artificial pancreas in type 1 diabetes. Diabetes Care 2012;35:e65–e67.
31. DeSalvo D, Keith-Hynes P, Peyser T, et al. Remote glucose monitoring in camp setting reduces the risk of prolonged nocturnal hypoglycemia. Diabetes Technol Ther 2013;16(1):1–7;
32. Kovatchev BP, Renard E, Cobelli C, et al. Safety of outpatient closed-loop control: first randomized crossover trials of a wearable artificial pancreas. Diabetes Care 2014;37:1789–1796.
33. Chernavvsky DR, DeBoer MD, Keith-Hynes P, et al. Use of an artificial pancreas among adolescents for a missed snack bolus and an underestimated meal bolus. Pediatr Diabetes 2016;17(1):28–35;
34. Brown SA, Kovatchev BP, Breton MD, et al. Multinight “Bedside” closed-loop control for patients with type 1 diabetes. Diabetes Technol Ther 2015;17(3):203–209;
35. Kovatchev B, Cheng P, Anderson SM, et al. Feasibility of long-term closed-loop control: A multicenter 6-month trial of 24/7 automated insulin delivery. Diabetes Technol Ther 2017;19(1):18–24.
36. Battelino T, Danne T, Bergenstal RM, et al. for the International Time-in-Range Consensus. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care 2019;42:1593–1603.
37. Schaller HC, Schaupp L, Bodenlenz M, et al. On-line adaptive algorithm with glucose prediction capacity for subcutaneous closed loop control of glucose: Evaluation under fasting conditions in patients with type 1 diabetes. Diabet Med 2006;23(1):90–93;
38. Shi D, Dassau E, Doyle FJ. Adaptive zone model predictive control of artificial pancreas based on glucose- and velocity-dependent control penalties. IEEE Trans Biomed Eng 2019;66(4):1045–1054;
39. Pinsker JE, Dassau E, Deshpande S, et al. Outpatient randomized crossover comparison of zone model predictive control automated insulin delivery with weekly data driven adaptation versus sensor-augmented pump: Results from the International Diabetes Closed-Loop Trial 4. Diabetes Technol Ther 2022;24(9):635–642;
40. Sun X, Rashid M, Hobbs N, et al. Incorporating prior information in adaptive model predictive control for multivariable artificial pancreas systems. J Diabetes Sci Technol 2022;16(1):19–28;
41. Ware J, Boughton CK, Allen JM, et al. Cambridge hybrid closed-loop algorithm in children and adolescents with type 1 diabetes: A multicentre 6-month randomised controlled trial. Lancet Digit Health 2022;4(4):e245–e255;
42. Brown SA, Forlenza GP, Bode BW, et al. Multicenter trial of a tubeless, on-body automated insulin delivery system with customizable glycemic targets in pediatric and adult participants with type 1 diabetes. Diabetes Care 2021;44(7):1630–1640;

Information & Authors


Published In

cover image Diabetes Technology & Therapeutics
Diabetes Technology & Therapeutics
Volume 26Issue Number 6June 2024
Pages: 375 - 382
PubMed: 38277161


Published in print: June 2024
Published online: 30 May 2024
Published ahead of production: 26 January 2024


Request permissions for this article.




    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
    Alberto Castillo
    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
    Elliott Pryor
    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
    Laura L. Kollar
    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
    Charlotte L. Barnett
    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
    Mark D. DeBoer
    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.
    Sue A. Brown
    Center for Diabetes Technology, University of Virginia School of Medicine, Charlottesville, Virginia, USA.


    Address correspondence to: Boris Kovatchev, PhD, Center for Diabetes Technology, University of Virginia School of Medicine, 560 Ray C Hunt Drive, Charlottesville, VA 22903, USA [email protected]

    Authors' Contributions

    B.K. co-created the NAP concept, contributed to the study design, was the sponsor of the Investigational Device Exemption by FDA, and wrote the first draft of this manuscript. A.C. and E.P. co-designed the NAP algorithm and contributed to the engineering study design, data analysis, and manuscript writing. L.K. was Project and Nursing Manager and contributed to the writing of this manuscript. C.B. was responsible for the data retrieval and contributed to the writing of this manuscript. M.D.D. and S.A.B. contributed to the study design, trial execution as the study physicians, and manuscript writing.

    Author Disclosure Statement

    B.K. declares research support from Dexcom, Novo Nordisk, and Tandem Diabetes Care, and patent royalties handled by the University of Virginia's Licensing and Ventures Group from Dexcom, Lifescan, Novo Nordisk, and Sanofi. M.D.B. has received research support to UVA from Dexcom, Tandem Diabetes Care, and Medtronic. S.A.B. has received research support to UVA from Dexcom, Insulet, Roche, Tandem Diabetes Care, and Tolerion.

    Funding Information

    National Institutes of Health/National Institute for Diabetes and Digestive and Kidney Diseases (NIDDK) Grant RO1 DK 133148. RedCap at the University of Virginia is supported in part by the National Center for Advancing Translational Sciences of the NIH under Award # UL1TR003015.

    Metrics & Citations



    Export citation

    Select the format you want to export the citations of this publication.

    View Options

    Get Access

    Access content

    To read the fulltext, please use one of the options below to sign in or purchase access.

    Society Access

    If you are a member of a society that has access to this content please log in via your society website and then return to this publication.

    Restore your content access

    Enter your email address to restore your content access:

    Note: This functionality works only for purchases done as a guest. If you already have an account, log in to access the content to which you are entitled.

    View options


    View PDF/ePub

    Full Text

    View Full Text







    Copy the content Link

    Share on social media

    Back to Top