Real-time detection of acromegaly from facial images with artificial intelligence
Authors' conclusion
The present study provided evidence that the proposed AcroEnsemble Model might detect acromegaly from facial images with high performance. This highlights that artificial intelligence programs are promising methods for detecting acromegaly in the future.
Study context
- Population
- 77 acromegaly patients (52.56 ± 11.74, 34 males/43 females; 56/77 (73%) in remission) and 71 healthy controls (48.47 ± 8.91, 39 males/32 females), gender- and age-compatible
- Comparison
- Deep learning models for detecting acromegaly from facial images
- Outcome
- Detection of acromegaly from facial images
Methodological detailscross-sectional, single-centre
- Cross-sectional
- Single-centre
Performance
As reported in the abstract
| Test or model | Sensitivity | Specificity | PPV | Accuracy | Precision |
|---|---|---|---|---|---|
| ResNet50 | — not stated in the abstract | — not stated in the abstract | 0.958 | — not stated in the abstract | — not stated in the abstract |
| DenseNet121 | — not stated in the abstract | — not stated in the abstract | 0.965 | — not stated in the abstract | — not stated in the abstract |
| InceptionV3 | — not stated in the abstract | — not stated in the abstract | 0.962 | — not stated in the abstract | — not stated in the abstract |
| Ensemble Method (EM) | 0.997 | 0.997 | 0.997 | 0.997 | 0.998 |
Sensitivity
ResNet50— not stated in the abstract
DenseNet121— not stated in the abstract
InceptionV3— not stated in the abstract
Ensemble Method (EM)0.997
Specificity
ResNet50— not stated in the abstract
DenseNet121— not stated in the abstract
InceptionV3— not stated in the abstract
Ensemble Method (EM)0.997
PPV
ResNet500.958
DenseNet1210.965
InceptionV30.962
Ensemble Method (EM)0.997
Accuracy
ResNet50— not stated in the abstract
DenseNet121— not stated in the abstract
InceptionV3— not stated in the abstract
Ensemble Method (EM)0.997
Precision
ResNet50— not stated in the abstract
DenseNet121— not stated in the abstract
InceptionV3— not stated in the abstract
Ensemble Method (EM)0.998
— not stated in the abstract
Original publication
Real-time detection of acromegaly from facial images with artificial intelligence. Eur J Endocrinol. Published 13 January 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →