Development of an artificial intelligence-based primary headache diagnosis model
Not Applicable
- Conditions
- Headache
- Registration Number
- JPRN-UMIN000048543
- Lead Sponsor
- Tominaga Hospital
- Brief Summary
76% correct, 56% sensitivity, 92% specificity for the 1200 patients' testdata. Non-specialist headache diagnostic accuracy for new 50 patients without AI was 46%, but improved to 83% with AI.
- Detailed Description
Not available
Recruitment & Eligibility
- Status
- Complete: follow-up complete
- Sex
- All
- Target Recruitment
- 4050
Inclusion Criteria
Not provided
Exclusion Criteria
Patients who cannot answer the questionnaire sheet
Study & Design
- Study Type
- Observational
- Study Design
- Not specified
- Primary Outcome Measures
Name Time Method Diagnostic accuracy of headache diagnostic based on headache questionnaires by non-headache specialists with and without AI-based diagnostic models.
- Secondary Outcome Measures
Name Time Method Diagnostic accuracy of AI-based diagnostic models in the test data.