Artificial Intelligence in Perioperative Medicine: A Proposed Common Language With Applications to FDA-Approved Devices

Abstract

As implementation of artificial intelligence grows more prevalent in perioperative medicine, a clinician's ability to distinguish differentiating aspects of these algorithms is critical. There are currently numerous marketing and technical terms to describe these algorithms with little standardization. Additionally, the need to communicate with algorithm developers is paramount to actualize effective and practical implementation. Of particular interest in these discussions is the extent to which the output or predictions of algorithms and tools are understandable by medical practitioners. This work proposes a simple nomenclature that is intelligible to both clinicians and developers for quickly describing the interpretability of model results. There are three high-level categories: transparent, translucent, and opaque. To demonstrate the applicability and utility of this terminology, these terms were applied to the artificial intelligence and machine-learning-based products that have gained Food and Drug Administration approval. During this review and categorization process, 22 algorithms were found with perioperative utility (in a database of 70 total algorithms), and 12 of these had publicly available citations. The primary aim of this work is to establish a common nomenclature that will expedite and simplify descriptions of algorithm requirements from clinicians to developers and explanations of appropriate model use and limitations from developers to clinicians.

Strengths-weaknesses-opportunities-threats analysis of artificial intelligence in anesthesiology and perioperative medicine

“…AI is currently in its infancy within the specialty of anesthesiology, as demonstrated by the relatively small number of approved algorithms compared to, say, Radiology ( 7 ). This review presents several different schools of thought regarding where AI can go in anesthesiology.…”

“…Diagnostic Radiology and Pathology are two specialties that have embraced significant adoption of AI technologies. In addition to diagnostic radiology and pathology, there have been increasing AI and ML-based technologies spanning oncology, ophthalmology, and general medicine decision making ( 7 , 8 ). In terms of more simple image-based specialties, diagnostic radiology and pathology are the two clinical specialties were a most obvious entry-point for AI applications in healthcare, as they both utilize one of AI's most well-vetted skills: deep learning for computer vision ( 9 ).…”

“…Black box models provide decisions and predictions without access to model details (weights) making it unclear how these conclusions are derived. In this sense, there is a lack of “context awareness,” in regard to how a black box will operate ( 7 ). Solutions for model interpretability are available, such as saliency maps and SHAP values, which better delineate how AI functions within its boundaries.…”

Artificial Intelligence-Supported Systems in Anesthesiology and Its Standpoint to Date—A Review

“…Recently the transparency of these algorithms are discussed and reviewed. Unfortunately, there is no FDA approved algorithm with anesthesiology utility found [57].…”

Large language models in physical therapy: time to adapt and adept

“…The potential benefits of AI and ML in physical therapy are vast and encompass several crucial areas ( Table 2 ). Medical and health sciences practitioners and educators have delved into the use of language models in ophthalmology, radiology, orthopedics, dermatology, surgery, obstetrics, and gynecology for educational and clinical practice development ( 20 – 24 ). Wilhem et al evaluated the performance of four language models, i.e., GPT-3.5.…”