1087-P: AI as a Clinical Advisor—Benchmarking AI’s Recommendations against Clinicians’ Practices in Type 2 Diabetes Management

Abstract

Objective: To compare clinical recommendations on type 2 diabetes management based on the ADA’s Standards of Medical Care - 2021 generated by two generative artificial intelligent (AI) platforms for internal consistency (vertically within a platform) and external consensus (horizontally between platforms and with clinicians’ clinical decisions). Methods: A complex clinical case was chosen because the patient has developed comorbid conditions related to type 2 diabetes (e.g., myocardial infarction, stroke, chronic kidney disease). Before entering the case into each AI platform for analysis, the clinical note - chief complaint, subjective, and objective information - was edited and organized to maximize clarity and standardize the language. Following the analyses by the two AI platforms, the outcomes was analyzed qualitatively by comparing the assessment and plan written by a clinician. Results: Conclusion: The two generative AI platforms achieved a similar internal consistency. Regarding external consensus, both platforms focused on key points in the effectiveness theme for secondary cardiovascular prevention. Although this abstract shows one pilot case, this qualitative approach sets a foundation for future analyses as the clinical recommendations from the two generative AI platforms exhibited alignment with the notes written by clinicians (diabetes specialists). Disclosure C.F. Young: Other Relationship; Abbott. S. Wong: None.

Ähnliche Arbeiten