Clinical Performance Evaluation of AI-Enabled Automated Gram Staining Device
Study Details
Study Description
Brief Summary
We have developed an analysis AI for Gram staining. In this study, we will compare the testing accuracy of automated Gram staining equipment with AI with the testing accuracy of laboratory technicians. Based on the results, we will examine the possibility of clinical application of the automated Gram staining device.
Condition or Disease | Intervention/Treatment | Phase |
---|---|---|
|
Detailed Description
Gram staining,whitch provides crucial information for selecting antibiotics in the diagnosis and treatment of infectious diseases, is often manually performed in many laboratories, and the microscopic findings are subjectively interpreted and reported based on the experience of laboratory technicians. Automating the entire process from staining to microscopic examination, as achieved by the AI-equipped automatic Gram staining device, can reduce the labor required for Gram staining. This automation enables even less experienced technicians to quickly report Gram staining results.
The aim of this study is to evaluate the accuracy of the Auto microscopic examination of Gram-stained samples using anonymized processed slides created from clinical specimens. This evaluation will be compared with the accuracy of microscopic examinations performed by hospital laboratory technicians.
If the accuracy of the machine is demonstrated, it can be registered as a medical device and proceed to the implementation stage in actual clinical settings. The implementation of this device in clinical laboratories is expected to reduce workload, expedite processes, and contribute to the optimization of antibiotic use based on Gram staining results.
Study Design
Outcome Measures
Primary Outcome Measures
- Primary Outcome [only 1 time of measurement]
The difference in accuracy rates between the Gram staining examination performed by laboratory technicians and the automated Gram staining device, including the approximate bacterial quantity based on Gram staining and morphology (categorized as -, 1+, 2+, 3+, 4+), presence of white blood cells, presence of phagocytic images, and, in the case of sputum, the Geckler classification.
Secondary Outcome Measures
- Secondary Outcome [only 1 time of measurement]
Impact of specimen type and identified bacterial species on testing accuracy.
Other Outcome Measures
- Exploratory evaluation items [only 1 time of measurement]
The difference in accuracy rates of microscopic examination findings by technicians at each facility based on the technicians' years of experience.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
Cases where specimens for smear, culture, and identification testing of sputum, urine, and blood were submitted based on clinical necessity.
-
Cases where a sufficient amount of specimen was available, and the creation of two slide samples did not impact the conduct of the test.
Exclusion Criteria:
- Cases in the study where there was a refusal to use specimens for research, based on information provided about opting out, prior to the implementation of anonymization.
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- GramEye
Investigators
None specified.Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- K22581