Application of Machine Learning Algorithms in Automatic Anesthetic Drug Delivery Systems: A literature review
DOI:
https://doi.org/10.12775/QS.2026.51.68152Keywords
anesthetic drugs, machine learning, closed-loop system, artificial intelligenceAbstract
The dynamic advancements in artificial intelligence are enabling the integration of machine learning into medical practice. A significant challenge in modern anesthesiology is the need for precise, continuous adjustment of anesthetic drug doses in response to the patient's rapidly changing physiological state. Traditional manual methods of drug delivery often result in excessive variability in the depth of anesthesia and hemodynamic stability. Utilizing machine learning algorithms represents a modern, promising approach designed to enhance the overall quality and safety of anesthetic care.
Aim of the study: The objective of this study is to review the current state of knowledge concerning machine algorithms in automated anesthetic drug delivery systems and to compare these with conventional administration methods, evaluating the benefits and limitations of each approach.
Materials and methods: A review of selected literature in the PubMed, Google Scholar database was conducted, using the following keywords: “anesthetic drugs”, “machine learning”, “closed-loop system” “artificial intelligence”
Conclusions: The review demonstrates that Machine Learning-driven Closed-Loop Anesthesia Delivery Systems (CLADS) offer superior stability of anesthetic depth and hemodynamic control compared to manual administration. The implementation of AI allows for the personalization of therapy and a significant reduction in drug and gas wastage, supporting the "Green Anesthesia" initiative. Furthermore, automation contributes to better long-term outcomes, such as reduced incidence of postoperative delirium and cognitive dysfunction. However, despite high autonomy, the "human-in-the-loop" paradigm remains essential due to technical artifacts and unforeseen surgical events. Future research should focus on regulatory standardization and the integration of multi-loop systems managing hypnosis, analgesia, and neuromuscular blockade simultaneously.
References
1. Jiang F, Jiang Y, Zhi H, et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol. 2017;2(4):230–243. https://doi.org/10.1136/svn-2017-000101
2. Hashimoto DA, Witkowski E, Gao L, Meireles O, Rosman G. Artificial intelligence in anesthesiology: current techniques, clinical applications, and limitations. Anesthesiology. 2020;132(2):379–394. DOI: 10.1097/ALN.0000000000002960
3. Vutskits L, Xie Z. Lasting impact of general anaesthesia on the brain: mechanisms and relevance. Nat Rev Neurosci. 2016;17(11):705-717. DOI: 10.1038/nrn.2016.128
4. Mashour GA, Avidan MS. Intraoperative awareness: controversies and non-controversies. Br J Anaesth. 2015;115 Suppl 1:i20-i26. DOI: 10.1093/bja/aev034
5. Rinehart J, Chawla R, Bein B. A multicenter evaluation of a BIS-guided closed-loop anesthesia delivery system: a randomized controlled trial. Anesth Analg. 2016 Jan;122(1):106-14. doi: 10.1213/ANE.0000000000000769.
6. Hemmerling TM, Charabati S, Zaouter C, Minardi C, Mathieu PA A randomized controlled trial demonstrates that a novel closed-loop propofol system performs better hypnosis control than manual administration. Anesthesiology. 2010. Aug;57(8):725-35. doi: 10.1007/s12630-010-9335-z
7. Budic I, Jevtovic Stoimenov T, Pavlovic D, Marjanovic V, Djordjevic I, Stevic M, Simic D. Clinical Importance of Potential Genetic Determinants Affecting Propofol Pharmacokinetics and Pharmacodynamics. Front Med (Lausanne). 2022 Feb 28;9:809393. doi: 10.3389/fmed.2022.809393.
8. Sinha A, Singh A, Tewari A. The fatigued anesthesiologist: a threat to patient safety? J Anaesthesiol Clin Pharmacol. 2013;29(2):151–159. DOI: 10.4103/0970-9185.111657
9. Weinger MB, Slagle J. Human Factors Research in Anesthesia Patient Safety: Techniques to Elucidate Factors Affecting Clinical Task Performance and Decision Making. J Am Med Inform Assoc. 2002 Nov-Dec;9(6 Suppl 1):s58–63. doi: 10.1197/jamia.M1229.
10. Schüttler J, Schwilden H, Stoeckel H. Pharmacokinetics as applied to total intravenous anaesthesia. Theoretical considerations. Anaesthesia. 1983 Jul;38 Suppl:51-2. doi: 10.1111/j.1365-2044.1983.tb15178.x.
11. Absalom AR, Mani V, De Smet T, Struys MM. Pharmacokinetic models for propofol: defining and illuminating the devil in the detail. Br J Anaesth. 2009;103(1):26-37. DOI: 10.1093/bja/aep143
12. Hemmerling TM, Arbeid E, Wehbe M, et al. Evaluation of a novel closed-loop total intravenous anaesthesia drug delivery system: a randomized controlled trial. Br J Anaesth. 2013;110(6):1031-9. DOI: 10.1093/bja/aet001
13. Moore BL, Pyeatt LD, Kulkarni V, et al. Reinforcement learning for closed-loop propofol anesthesia: a study in human volunteers. J Mach Learn Res. 2014;15:655–696.
14. Rosow C, Manberg PJ. . 2001;19(4):947-66. Bispectral index monitoring DOI: 10.1016/s0889-8537(01)80018-3
15. Viertiö-Oja H, Maja V, Särkelä M, et al. Description of the Entropy algorithm as applied in the Datex-Ohmeda S/5 Entropy Module. Acta Anaesthesiol Scand. 2004;48(2):154-61. DOI: 10.1111/j.0001-5172.2004.00322.x
16. Dumont GA, Ansermino JM. Closed-loop control of anesthesia: a primer for anesthesiologists. Anesth Analg. 2013 Nov;117(5):1130-8. doi: 10.1213/ANE.0b013e3182973687.
17. Zaouter C, Joosten A, Rinehart J, Struys MMRF, Hemmerling TM. Autonomous Systems in Anesthesia: Where Do We Stand in 2020? A Narrative Review. Anesth Analg. 2020;130(5):1120-1132. DOI: 10.1213/ANE.0000000000004646
18. Gonzalez-Cava JM, Lam J, Karamanoglu M, Pascoal AMB. Robust PID control of propofol anaesthesia: Uncertainty limits performance, not PID structure. Comput Methods Programs Biomed. 2021;198:105783. 10.1016/j.cmpb.2020.105783
19. van Heusden K, Soltesz K, Cooke E, et al. Optimizing robust PID control of propofol anesthesia for children: design and clinical evaluation. IEEE Trans Biomed Eng. 2019;66(10):2918-2923. DOI: 10.1109/TBME.2019.2898194
20. Liu N, Chazot T, Hamada S, et al. Closed-loop coadministration of propofol and remifentanil guided by bispectral index: a randomized multicenter study. Anesth Analg. 2011;112(3):546-557.
21. Soltesz K, Hahn JO, Hägglund T, Dumont GA, Ansermino JM. Individualized closed-loop control of propofol anesthesia: A preliminary study. Biomed Signal Process Control. 2013;8(6):500-508. DOI: 10.1213/ANE.0b013e318205680b
22. Ionescu CM, Keyser RD, Torrico BC, et al. Robust predictive control strategy applied for propofol dosing using BIS as a controlled variable during anesthesia. IEEE Trans Biomed Eng. 2008;55(9):2161-2170. DOI: 10.1109/TBME.2008.923142
23. Struys MMRF, De Smet T, Glen JIB, et al. The history of target-controlled infusion. Anesth Analg. 2016;122(1):56-69. DOI: 10.1213/ANE.0000000000001008
24. van Heusden K, Dumont GA, Soltesz K, et al. Design and clinical evaluation of robust PID control of propofol anesthesia in children. IEEE Trans Control Syst Technol. 2014;22(2):491-501. DOI: 10.1109/TCST.2013.2260543
25. Singhal M, Gupta L, Hirani K. A Comprehensive Analysis and Review of Artificial Intelligence in Anaesthesia. Cureus. 2023;15(9):e45038. DOI: 10.7759/cureus.45038
26. Araujo H, Xiao B, Liu C, Zhao Y, Lam HK. Design of Type-1 and Interval Type-2 Fuzzy PID Control for Anesthesia Using Genetic Algorithms. J Intell Learn Syst Appl. 2014;6(2):70-93. DOI: 10.4236/jilsa.2014.62007
27. Lin CS, Li YC, Mok MS, et al. Neural network modeling to predict the hypnotic effect of propofol bolus induction. Proc AMIA Symp. 2002:450-3. PMCID: PMC2244570
28. Mirinejad H, Parastvand H, Hesamian MH. Feedback control of surgical anesthesia: A review. Annu Rev Control. 2020;50:1-26.
29. Wingert T, Lee C, Cannesson M. Machine Learning, Deep Learning, and Closed Loop Devices—Anesthesia Delivery. Anesthesiol Clin. 2021;39(3):565-581. DOI: 10.1016/j.anclin.2021.03.012
30. Li R, Wu Q, Liu J, et al. Monitoring Depth of Anesthesia Based on Hybrid Features and Recurrent Neural Network. Front Neurosci. 2020;14:26. doi: 10.3389/fnins.2020.00026
31. He X, Li T, Wang X. Research progress on the depth of anesthesia monitoring based on the electroencephalogram. Ibrain. 2024 Dec 6;11(1):32-43. doi: 10.1002/ibra.12186
32. Absalom AR, Sutcliffe N, Kenny GN. Closed-loop control of anesthesia using Bispectral index: performance assessment in patients undergoing major orthopedic surgery under combined general and regional anesthesia. Anesthesiology. 2002;96(1):67-73. DOI: 10.1097/00000542-200201000-00017
33. Ribba B, Bräm DS, Baverel PG, Peck RW. Model enhanced reinforcement learning to enable precision dosing: A theoretical case study with dosing of propofol. CPT Pharmacometrics Syst Pharmacol. 2022;11(11):1497-1510. DOI: 10.1002/psp4.12858
34. Tu Z, Jeffries S, Pelletier E, et al. Deep reinforcement learning for multi-targets propofol dosing. J Clin Monit Comput. 2025;39(3):613-623. DOI: 10.1007/s10877-025-01269-z
35. Malagutti N, McGinness G, Nithyanandam DA. Real-Time Personalised Pharmacokinetic-Pharmacodynamic Modelling in Propofol Anesthesia through Bayesian Inference. Annu Int Conf IEEE Eng Med Biol Soc. 2023. DOI: 10.1109/EMBC40787.2023.10339991
36. Viterbo JF, Lourenço AP, Leite-Moreira AF, et al. Prospective randomised comparison of Marsh and Schnider pharmacokinetic models for propofol during induction of anaesthesia. Eur J Anaesthesiol. 2012;29(10):477-83. DOI: 10.1097/EJA.0b013e3283542421
37. Hüppe T, Maurer F, Sessler DI, et al. Retrospective comparison of Eleveld, Marsh, and Schnider propofol pharmacokinetic models in 50 patients. Br J Anaesth. 2020;124(2):e22-e24. DOI: 10.1016/j.bja.2019.10.019
38. Eleveld DJ, Proost JH, Cortínez LI, et al. A general purpose pharmacokinetic model for propofol. Anesth Analg. 2014;118(6):1221-37. DOI: 10.1213/ANE.0000000000000165
39. Hall S, Ortmann L, Picallo M, Dörfler F. Real-time Projected Gradient-based Nonlinear Model Predictive Control with an Application to Anesthesia Control. IEEE CDC. 2022. DOI: 10.1109/CDC51059.2022.9992498.
40. Aubouin–Pairault B, Fiacchini M, Dang T. Online identification of pharmacodynamic parameters for closed-loop anesthesia with model predictive control. Computers & Chemical Engineering. 2024;191:108837. https://doi.org/10.1016/j.compchemeng.2024.108837
41. Cai X, Wang X, Zhu Y, et al. Advances in automated anesthesia: a comprehensive review. Anesthesiol Perioper Sci. 2025;3:3. https://doi.org/10.1007/s44254-024-00085-z
42. Khosravi S. Constrained model predictive control of hypnosis (Thesis). University of British Columbia. 2015. DOI:10.14288/1.0223114
43. Rezvanian S, Towhidkhah F, Ghahramani N. Increasing Robustness of the Anesthesia Process from Difference Patient's Delay Using a State-Space Model Predictive Controller. Procedia Engineering. 2011;15:928-932. https://doi.org/10.1016/j.proeng.2011.08.171
44. Parihar S, Shah P, Sekhar RA, Lagoo JY. Model Predictive Control and Its Role in Biomedical Therapeutic Automation: A Brief Review. Applied System Innovation. 2022. https://doi.org/10.3390/asi5060118
45. Pawłowski A, Schiavo M, Latronico N, et al. Event-based MPC for propofol administration in anesthesia. Comput Methods Programs Biomed. 2023;229:107289. DOI: 10.1016/j.cmpb.2022.107289
46. Puri GD, Mathew PJ, Biswas I, et al. A multicenter evaluation of a closed-loop anesthesia delivery system: a randomized controlled trial. Anesth Analg. 2016;122(1):106-114. DOI: 10.1213/ANE.0000000000000769
47. Puri GD, Kumar B, Aveek J. Closed-loop anaesthesia delivery system (CLADS) using bispectral index: a performance assessment study. Anaesth Intensive Care. 2007;35(3):357-362. DOI: 10.1177/0310057X0703500306
48. Struys MMRF, De Smet T, Versichelen LF, et al. Comparison of closed-loop controlled administration of propofol using Bispectral Index as the controlled variable versus "standard practice" controlled administration. Anesthesiology. 2001;95(1):6-17. DOI: 10.1097/00000542-200107000-00007
49. Varvel JR, Donoho DL, Shafer SL. Measuring the predictive performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm. 1992;20(1):63-94. DOI: 10.1007/BF01143186
50. Chan MT, Cheng BC, Lee TM, Gin T. BIS-guided anesthesia decreases postoperative delirium and cognitive decline. J Neurosurg Anesthesiol. 2013;25(1):33-42. DOI: 10.1097/ANA.0b013e3182712fba
51. Pasin L, Nardelli P, Pintaudi M, et al. Closed-Loop Delivery Systems Versus Manually Controlled Administration of Total IV Anesthesia: A Meta-analysis of Randomized Clinical Trials. Anesth Analg. 2017;124(2):456-464. DOI: 10.1213/ANE.0000000000001394
52. Sng BL, Tan HS, Sia AT. Closed-loop double-vasopressor automated system vs manual bolus vasopressor to treat hypotension during spinal anaesthesia for caesarean section: a randomised controlled trial. Anaesthesia. 2014;69(1):37-45. DOI: 10.1111/anae.12460
53. Liu N, Chazot T, Genty A, et al. Titration of propofol for anesthetic induction and maintenance guided by the bispectral index: closed-loop versus manual control. Anesthesiology. 2006;104(4):686-95. DOI: 10.1097/00000542-200604000-00012
54. Wu B, Zhu W, Wang Q, et al. Efficacy and safety of ciprofol-remifentanil versus propofol-remifentanil during fiberoptic bronchoscopy: A prospective, randomized, double-blind, non-inferiority trial. Front Pharmacol. 2022;13:1091579. DOI: 10.3389/fphar.2022.1091579
55. Eleveld DJ, Colin P, Absalom AR, et al. Pharmacokinetic-pharmacodynamic model for propofol for broad application in anaesthesia and sedation. Br J Anaesth. 2018;120(5):942-959. DOI: 10.1016/j.bja.2018.01.018
56. Singaravelu S, Barclay P. Automated control of end-tidal inhalation anaesthetic concentration decreases agent consumption in comparison with manual control. Br J Anaesth. 2013;110(4):561-566. DOI: 10.1093/bja/aes464
57. Hemmerling TM, Taddei R. Robotic anesthesia - a vision for the future of anesthesia. Transl Med UniSa. 2011;1:1-14. PMID: 23905028
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Jakub Michał Lichoń, Paweł Jan Kuna, Wojciech Kuna, Jakub Mateusz Pietrucha, Kamil Igor Turczynowski, Mateusz Józef Gołdyn, Konrad Olaf Turczynowski
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Stats
Number of views and downloads: 4
Number of citations: 0
