Table 1 Eight clinical cases used for evaluation.

From: Pre-trained ChatGPT for report generation in automated microbial identification and antibiotic susceptibility testing systems

Case no.

Question

Purpose

Correct solution

1

A clinical isolate requires accurate identification for diagnosis.

Assess ability to identify organisms based on Vitek-2 outputs.

Correct bacterial species identification.

2

Susceptibility testing is performed on a mixed culture.

Evaluate recognition of purity requirements for valid testing.

Report that a pure bacterial culture is required for accurate results.

3

A rare NDM-producing strain not present in the expert database.

Test handling of database limitations and resistance interpretation.

Identify NDM enzyme presence and infer carbapenem resistance.

4

A CRE strain harboring blaOXA-48 resistance gene.

Evaluate detection and reporting of carbapenemase type.

Correctly report blaOXA-48 and its associated resistance phenotype.

5

A mucoid strain is tested by automated susceptibility system.

Evaluate recognition of sample limitations in automated testing.

Report that mucoid phenotype may compromise susceptibility test accuracy.

6

Contradictions are found in detected resistance mechanisms.

Assess ability to reconcile conflicting susceptibility results.

Provide a consistent interpretation aligned with known resistance mechanisms.

7

Determination of a specific enzyme type is required.

Evaluate ability to identify enzyme-mediated resistance.

Correctly identify and report the enzyme type.

8

No matching resistant expression type is identified.

Test handling of novel or unclassified resistance patterns.

Report as “unclassified resistance mechanism; further testing required.”