Recommended antibiotic treatment agreement between infectious diseases specialists and ChatGPT®

被引：0

作者：

Montiel-Romero, Santiago ^{[1
]}

Rajme-Lopez, Sandra ^{[1
]}

Roman-Montes, Carla Marina ^{[1
,2
]}

Lopez-Iniguez, Alvaro ^{[1
]}

Rivera-Villegas, Hector Orlando ^{[1
,2
]}

Ochoa-Hein, Eric ^{[1
,3
]}

Gonzalez-Lara, Maria Fernanda ^{[1
,2
]}

Ponce-de-Leon, Alfredo ^{[1
]}

Tamez-Torres, Karla Maria ^{[1
,2
]}

Martinez-Guerra, Bernardo Alfonso ^{[1
,2
]}

机构：

[1] Inst Nacl Ciencias Med & Nutr Salvador Zubiran, Dept Infect Dis, 15 Vasco Quiroga,Belisario Dominguez Secc 16 Tlalp, Mexico City 14080, Mexico

[2] Inst Nacl Ciencias Med & Nutr Salvador Zubiran, Clin Microbiol Lab, Dept Infect Dis, 15 Vasco Quiroga,Belisario Dominguez Secc 16 Tlalp, Mexico City 14080, Mexico

[3] Inst Nacl Ciencias Med & Nutr Salvador Zubiran, Hosp Epidemiol Dept, 15 Vasco Quiroga,Belisario Dominguez Secc 16 Tlalp, Mexico City 14080, Mexico

来源：

BMC INFECTIOUS DISEASES | 2025年 / 25卷 / 01期

关键词：

Artificial intelligence; Machine learning; Infectious diseases; Antimicrobial resistance; STAPHYLOCOCCUS-AUREUS BACTEREMIA; ARTIFICIAL-INTELLIGENCE; CONSULTATION;

D O I：

10.1186/s12879-024-10426-9

中图分类号：

R51 [传染病];

学科分类号：

100401 ;

摘要：

BackgroundAntimicrobial resistance is a global threat to public health. Chat Generative Pre-trained Transformer (ChatGPT (R)) is a language model tool based on artificial intelligence. ChatGPT (R) could analyze data from antimicrobial susceptibility tests in real time, especially in places where infectious diseases (ID) specialists are not available. We aimed to evaluate the agreement between ChatGPT (R) and ID specialists regarding appropriate antibiotic prescription in simulated cases.MethodsUsing data from microbiological isolates recovered in our center, we fabricated 100 cases of patients with different infections. Each case included age, infectious syndrome, isolated organism and complete antibiogram. Considering a precise set of instructions, the cases were introduced into ChatGPT (R) and presented to five ID specialists. For each case, we asked, (1) "What is the most appropriate antibiotic that should be prescribed to the patient in the clinical case?" and (2) "According to the interpretation of the antibiogram, what is the most probable mechanism of resistance?". We then calculated the agreement between ID specialists and ChatGPT (R), as well as Cohen's kappa coefficient.ResultsRegarding the recommended antibiotic prescription, agreement between ID specialists and ChatGPT (R) was observed in 51/100 cases. The calculated kappa coefficient was 0.48. Agreement on antimicrobial resistance mechanisms was observed in 42/100 cases. The calculated kappa coefficient was 0.39. In a subanalysis according to infectious syndromes and microorganisms, Agreement (range 25 - 80%) and kappa coefficients (range 0.21-0.79) varied.ConclusionWe found poor agreement between ID specialists and ChatGPT (R) regarding the recommended antibiotic management in simulated clinical cases.

引用

页数：6

共 20 条

[1] Impact of Infectious Disease Consultation on Quality of Care, Mortality, and Length of Stay in Staphylococcus aureus Bacteremia: Results From a Large Multicenter Cohort Study [J].