Acromegaly facial changes analysis using last generation artificial intelligence methodology: the AcroFace system
Authors' conclusion
AcroFace system shows a good performance to discriminate acromegaly and non-acromegaly facial traits that may serve for the detection of acromegaly at an early stage as a screening procedure at a population level.
Study context
- Population
- Facial photographs for detection of acromegaly; expert endocrinologist-based score as ground truth
- Comparison
- ResNet-50 + Support Vector Regression (SVR) vs VGG-16 + SVR
- Outcome
- Acromegaly detection performance
Methodological detailsAI-based, development and validation cohort
- AI-based
- Development and validation cohort
Performance
As reported in the abstract
| Test or model | Accuracy δ1 | Accuracy δ3 |
|---|---|---|
| ResNet-50 + SVR | 75% | 89% |
ResNet-50 + SVR
Accuracy δ175%
Accuracy δ389%
Further findings
Performance in validation cohort
- Precision
- 0.90
- Accuracy
- 0.93
- F1-Score
- 0.92
- Sensitivity
- 0.93
- Specificity
- 0.93
Original publication
Acromegaly facial changes analysis using last generation artificial intelligence methodology: the AcroFace system. Pituitary. Published 21 April 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →