Thoracic and abdominal trauma
Conditions
Interventions
Gold Standard:Each case was double-blind diagnosed by at least two senior ultrasound diagnostic doctors with more than 5 years of experience, and then reviewed by one senior ultrasound diagnostic doct
Index test:Artificial intelligence diagnostic models, mainly including Unet, etc.
Sponsors
Department of Military Health Statistics, Naval Medical University
Eligibility
Sex/Gender
All
Age
18 Years to 90 Years
Inclusion criteria
Inclusion criteria: Ultrasound images of abdominal hematoma (fluid), pleural effusion, pericardial effusion, and pneumothorax caused by various reasons; individuals aged 18 years and above.
Exclusion criteria
Exclusion criteria: Poor image quality; patients with encapsulated effusion; pyogenic effusion caused by infectious etiology.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Efast-based ultrasound video;Positive rate;Misdiagnosis rate; | — |
Secondary
| Measure | Time frame |
|---|---|
| Efast-based ultrasound iamge;Evaluation of reading efficiency;Performance analysis of area under the ROC curve (AUC) of ascites (blood) detection;The performance evaluation of AI-assisted diagnosis system in detecting ascites (blood) independently;Overall ease of use of AI-assisted diagnosis system prototype; | — |
Countries
China
Contacts
Public ContactHe Jia; Chen Qi
Department of Military Health Statistics, Naval Medical University
Outcome results
None listed