Related Experiment Video
Updated: Jul 22, 2026
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.7K
Bulldogs stenosis degree classification using synthetic images created by generative artificial intelligence.
Gustavo da Silva Andrade1, Gabriel Toshio Hirokawa Higa2, Jarbas Felipe da Silva Ribeiro2
1Universidade Federal de Mato Grosso do Sul, Campo Grande, Brazil. gustavo.s.andrade@ufms.br.
Scientific Reports
|March 22, 2025
Summary
This study created an AI model to diagnose nasal stenosis in bulldogs. The deep learning approach achieved human-level accuracy, aiding in better treatment and animal welfare.
Area of Science:
- Veterinary Medicine
- Artificial Intelligence
- Animal Anatomy
Background:
- Nasal stenosis significantly impacts bulldog quality of life.
- Early diagnosis is critical for effective treatment of this condition.
Purpose of the Study:
- To develop an automated deep learning model for classifying nasal stenosis severity in bulldogs.
- To compare the model's diagnostic performance against trained veterinary anatomists.
Main Methods:
- Utilized a dataset of 1020 bulldog nostril images, including real and AI-generated samples.
- Tested five neural network architectures, with DenseNet201 showing the highest performance.
- Evaluated model accuracy against human expert diagnosis.
Main Results:
- DenseNet201 achieved a median F-score of 54.04%.
- The deep learning model demonstrated comparable accuracy and reliability to human evaluators.
- AI model performance shows potential for clinical application.
Conclusions:
- Automated deep learning models can achieve human-level diagnostic performance for canine nasal stenosis.
- This technology can enhance diagnostic speed and accuracy, improving treatment planning.
- The study highlights the potential of AI to advance veterinary care and animal welfare.
