Review Paper

Artificial Intelligence–enabled Decision Support in Surgery

State-of-the-art and Future Directions

Loftus, Tyler J. MD^*,†; Altieri, Maria S. MD, MS^†,‡; Balch, Jeremy A. MD^*,†; Abbott, Kenneth L. MD, MS^*,†; Choi, Jeff MD, MSc^†,§; Marwaha, Jayson S. MD, MSc^†,∥,¶; Hashimoto, Daniel A. MD, MS^#,†,**; Brat, Gabriel A. MD, MPH^∥,†,¶; Raftopoulos, Yannis MD, PhD^†,††; Evans, Heather L. MD^†,‡‡; Jackson, Gretchen P. MD, PhD^†,§§; Walsh, Danielle S. MD^∥∥,†; Tignanelli, Christopher J. MD, MS^{†,¶¶,##,***}

Author Information

^*Department of Surgery, University of Florida Health, Gainesville, FL

^†American College of Surgeons Health Information Technology Committee and Artificial Intelligence Subcommittee, Chicago, IL

^‡Department of Surgery, University of Pennsylvania, Philadelphia, PA

^§Department of Surgery, Stanford University, Stanford, CA

^∥Department of Surgery, Beth Israel Deaconess Medical Center

^¶Department of Biomedical Informatics, Harvard Medical School, Boston, MA

^#Department of Surgery, University of Pennsylvania Perelman School of Medicine

^**General Robotics, Automation, Sensing, and Perception Laboratory, University of Pennsylvania School of Engineering and Applied Science, Philadelphia, PA

^††Weight Management Program, Holyoke Medical Center, Holyoke, MA

^‡‡Department of Surgery, Medical University of South Carolina, Charleston, SC

^§§Digital, Intuitive Surgical, Sunnyvale, CA; Departments of Pediatric Surgery, Pediatrics, and Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN

^∥∥Department of Surgery, University of Kentucky, Lexington, KY

^¶¶Department of Surgery

^##Institute for Health Informatics

^***Program for Clinical Artificial Intelligence, Center for Learning Health Systems Science, University of Minnesota, Minneapolis, MN

This article was prepared by the following members of the American College of Surgeons Health Information Technology Committee. T.J.L., M.S.A., J.A.B., K.L.A., J.C., J.S.M., D.A. H., G.A.B., Y.R., H.L.E., G.P.J., D.S.W., C.J.T.: Conception and Design, Literature Review, Interpretation of Data, Writing and Critical Revision. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

T.J.L. was supported by the National Institute of General Medical Sciences (NIGMS) of the National Institutes of Health under Award Number K23GM140268 and by the Thomas H. Maren Junior Investigator Fund. Y.R. is the founder of RightBMI, LLC, a company that has developed a fully automated responsive web software application that generates individualized and structured nutritional and exercise plans and adjusts antihypertensive and antidiabetic medications. He has received research grants from Lexington Medical, Heron Therapeutics, and Allurion Technologies. He is also in the Speaker Bureau of Heron Therapeutics. G.P.J. is an employee of Intuitive Surgical and a shareholder of Intuitive Surgical, IBM, and Kyndryl. The remaining authors declare no conflicts of interest.

[email protected].

Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal’s website, www.annalsofsurgery.com.

Annals of Surgery 278(1):p 51-58, July 2023. | DOI: 10.1097/SLA.0000000000005853

Abstract

Objective:

To summarize state-of-the-art artificial intelligence–enabled decision support in surgery and to quantify deficiencies in scientific rigor and reporting.

Background:

To positively affect surgical care, decision-support models must exceed current reporting guideline requirements by performing external and real-time validation, enrolling adequate sample sizes, reporting model precision, assessing performance across vulnerable populations, and achieving clinical implementation; the degree to which published models meet these criteria is unknown.

Methods:

Embase, PubMed, and MEDLINE databases were searched from their inception to September 21, 2022 for articles describing artificial intelligence–enabled decision support in surgery that uses preoperative or intraoperative data elements to predict complications within 90 days of surgery. Scientific rigor and reporting criteria were assessed and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines.

Results:

Sample size ranged from 163–2,882,526, with 8/36 articles (22.2%) featuring sample sizes of less than 2000; 7 of these 8 articles (87.5%) had below-average (<0.83) area under the receiver operating characteristic or accuracy. Overall, 29 articles (80.6%) performed internal validation only, 5 (13.8%) performed external validation, and 2 (5.6%) performed real-time validation. Twenty-three articles (63.9%) reported precision. No articles reported performance across sociodemographic categories. Thirteen articles (36.1%) presented a framework that could be used for clinical implementation; none assessed clinical implementation efficacy.

Conclusions:

Artificial intelligence–enabled decision support in surgery is limited by reliance on internal validation, small sample sizes that risk overfitting and sacrifice predictive performance, and failure to report confidence intervals, precision, equity analyses, and clinical implementation. Researchers should strive to improve scientific quality.

Secondary Logo

Journal Logo

Artificial Intelligence–enabled Decision Support in Surgery

State-of-the-art and Future Directions

Abstract

Objective:

Background:

Methods:

Results:

Conclusions:

Full Text Access for Subscribers:

Institutional Users

Not a Subscriber?

Article Level Metrics

Article Keywords

Related Links

Related Videos

Artificial Intelligence–enabled Decision Support in Surgery: State-of-the-art and Future Directions : Annals of Surgery

Abstract

Objective:

Background:

Methods:

Results:

Conclusions:

Full Text Access for Subscribers:

Individual Subscribers

Institutional Users

Not a Subscriber?