Next Article in Journal
Feasibility of a Short-Stay Lumboperitoneal Shunt Pathway Based on Perioperative Optimization and Individualized Discharge Decision-Making: A Pilot Before–After Study
Next Article in Special Issue
Artificial Intelligence for Personalized Management of Acute Myeloid Leukemia
Previous Article in Journal
Bridging the Gap Between Pharmacogenomic Discovery and Clinical Implementation: Insights from Selected Studies on Inter-Individual Variability in Drug Response
Previous Article in Special Issue
Integrating AI Segmentation, Simulated Digital Twins, and Extended Reality into Medical Education: A Narrative Technical Review and Proof-of-Concept Case Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JPM
Volume 16
Issue 4
10.3390/jpm16040222
jpm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Clinical Application of Artificial Intelligence in Anesthesiology: A Multicenter Retrospective Comparison Between Human Anesthetic Decisions and Algorithmic Recommendations in Non-Cardiac Surgery

by
Gilberto Duarte-Medrano
1,*,
Natalia Nuño-Lámbarri
2,3,*,
Octavio Gonzalez-Chon
1,
Rebeca Garazi Elguezabal Rodelo
1,
Carmelo Calvagna
4,
Daniele Paternò
5,
Luigi La Via
4,6 and
Massimiliano Sorbello
7
1
Anesthesiology Department of Hospital Medica Sur, Mexico City 14050, Mexico
2
Translational Research Unit, Medica Sur Clinic & Foundation, Mexico City 14050, Mexico
3
Department of Surgery, Faculty of Medicine, The National Autonomous University of Mexico (UNAM), Mexico City 04510, Mexico
4
Department of Anesthesia and Intensive Care, University Hospital Policlinico “G. Rodolico–San Marco”, 95121 Catania, Italy
5
Department of Anesthesia and Intensive Care, Hospital “Giovanni Paolo II”, ASP Ragusa, 97100 Ragusa, Italy
6
Department of General Surgery and Medical Surgical Specialties, University of Catania, 95131 Catania, Italy
7
Faculty of Medicine and Surgery, University of Enna “Kore”, 94100 Enna, Italy
*
Authors to whom correspondence should be addressed.
J. Pers. Med. 2026, 16(4), 222; https://doi.org/10.3390/jpm16040222
Submission received: 10 March 2026 / Revised: 30 March 2026 / Accepted: 15 April 2026 / Published: 17 April 2026
(This article belongs to the Special Issue Artificial Intelligence for Personalized Medicine: Bridging Innovative Technologies and Patient-Centric Care)
Browse Figure
Versions Notes

Abstract

Background: Artificial intelligence (AI) is progressively entering perioperative medicine; however, its role in preoperative anesthetic decision-making remains insufficiently characterized. We evaluated the concordance between anesthesiologist-selected anesthetic techniques and algorithm-generated recommendations in a cohort of adult patients undergoing non-cardiac surgery. Methods: This retrospective observational study included adult patients (≥18 years) undergoing elective non-cardiac surgery between January 2024 and January 2025 at two international centers (Mexico and Italy). Clinical, demographic, and surgical variables were extracted from electronic medical records. For each case, a structured anonymized vignette was submitted to ChatGPT (version 5.0, medical configuration) to obtain an independent recommendation regarding anesthetic technique. Concordance between AI-generated and clinician-selected techniques was assessed using agreement analysis and stratified by country and surgical specialty. Results: A total of 1965 patients were analyzed. Overall concordance between ChatGPT recommendations and anesthesiologist-selected techniques was 84.6%. Agreement remained stable across centers (Mexico 84.3%; Italy 88.7%). Disagreement rates varied by surgical specialty, with the highest values observed in vascular and proctologic surgery (28.6%), followed by urology (21.1%) and thoracic surgery (18.8%). Orthopedic procedures—particularly shoulder arthroscopy—accounted for a relevant proportion of divergences, where AI frequently favored regional techniques over general anesthesia. No specialty demonstrated discordance exceeding 30%. Conclusions: AI-generated anesthetic recommendations demonstrated substantial concordance with expert clinical decision-making across heterogeneous surgical settings. These findings support the potential integration of AI within a hybrid decision-making framework, complementing—rather than replacing—anesthesiologist expertise in contemporary perioperative care.
Keywords: anesthesia; IA; ChatGPT; hybrid model anesthesia; IA; ChatGPT; hybrid model

Share and Cite

MDPI and ACS Style

Duarte-Medrano, G.; Nuño-Lámbarri, N.; Gonzalez-Chon, O.; Elguezabal Rodelo, R.G.; Calvagna, C.; Paternò, D.; La Via, L.; Sorbello, M. Clinical Application of Artificial Intelligence in Anesthesiology: A Multicenter Retrospective Comparison Between Human Anesthetic Decisions and Algorithmic Recommendations in Non-Cardiac Surgery. J. Pers. Med. 2026, 16, 222. https://doi.org/10.3390/jpm16040222

AMA Style

Duarte-Medrano G, Nuño-Lámbarri N, Gonzalez-Chon O, Elguezabal Rodelo RG, Calvagna C, Paternò D, La Via L, Sorbello M. Clinical Application of Artificial Intelligence in Anesthesiology: A Multicenter Retrospective Comparison Between Human Anesthetic Decisions and Algorithmic Recommendations in Non-Cardiac Surgery. Journal of Personalized Medicine. 2026; 16(4):222. https://doi.org/10.3390/jpm16040222

Chicago/Turabian Style

Duarte-Medrano, Gilberto, Natalia Nuño-Lámbarri, Octavio Gonzalez-Chon, Rebeca Garazi Elguezabal Rodelo, Carmelo Calvagna, Daniele Paternò, Luigi La Via, and Massimiliano Sorbello. 2026. "Clinical Application of Artificial Intelligence in Anesthesiology: A Multicenter Retrospective Comparison Between Human Anesthetic Decisions and Algorithmic Recommendations in Non-Cardiac Surgery" Journal of Personalized Medicine 16, no. 4: 222. https://doi.org/10.3390/jpm16040222

APA Style

Duarte-Medrano, G., Nuño-Lámbarri, N., Gonzalez-Chon, O., Elguezabal Rodelo, R. G., Calvagna, C., Paternò, D., La Via, L., & Sorbello, M. (2026). Clinical Application of Artificial Intelligence in Anesthesiology: A Multicenter Retrospective Comparison Between Human Anesthetic Decisions and Algorithmic Recommendations in Non-Cardiac Surgery. Journal of Personalized Medicine, 16(4), 222. https://doi.org/10.3390/jpm16040222

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop