Development and Accuracy Evaluation of Gram Staining Analysis AI
ConditionBacterial Infections
Trial statusRecruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
AgeNot listed
SponsorGramEye
About this trial
The investigators use Gram-stained specimens provided from clinical settings to develop and evaluate the accuracy of Gram staining analysis AI
Eligibility criteria
This trial does not accept healthy volunteersQualifiers
Patients in whom a Gram stain test was ordered
Disqualifiers
No exclusion criteria
Trial population
Specimens for microbiological testing, such as Gram staining, ordered at the hospital.
Trial design
Design model
Case-only
Time perspective
Retrospective
Treatments tested in this trial
Sample collection without intervention
Other interventionNo Intervention
Treatment groups
18,000 Participants
are divided into 1 treatment groupGroup A: Groups providing specimens for development of Gram Stain analysis AI1 intervention
Trial outcomes
Primary outcomes
1
Advanced level of blood culture Gram stain analysis AI accuracy
Analyzing Gram-stained microscopic images of blood culture-positive specimens using AI and evaluating the accuracy of bacterial species estimation For example, classify GPC clusters as S. aureus or CNS.
Time frame
Day12
Ordinary level of blood culture Gram stain analysis AI accuracy
Gram-stained microscopic images of blood culture-positive specimens are analyzed using AI, and bacteria are classified based on morphological classification. For example, gram-positive cocci (GPC) are classified into gram-positive staphylococci (GCP clusters) and gram-positive streptococci (GPC chains).
Time frame
Day1 Source ClinicalTrials.gov