Abstract
This study assessed automated bone density measurement technologies in pediatric groups, focusing on lumbar spine localization and spine segmentation models initially trained on adult data. The research involved three phases: training models using YOLOv5 and U-Net on adult images, adapting these models with pediatric data via transfer learning, and external validation categorized by age to account for anatomical variances. The adult-trained model showed decreased sensitivity in younger ages, with the lowest performance in the youngest group. Conversely, the pediatric-trained model achieved high sensitivity, over 90% in children under 10, and perfect scores in the 10-12 group, demonstrating improved accuracy. Qualitative analysis for segmentation indicated better performance in the pediatric model across all age groups, particularly in those under 13. The study concludes that transfer learning enhances the performance and generalizability of models for pediatric spine analysis, suggesting a potential for more accurate diagnostics.
Original language | English |
---|---|
Title of host publication | Medical Imaging 2024 |
Subtitle of host publication | Image Processing |
Editors | Olivier Colliot, Jhimli Mitra |
Publisher | SPIE |
ISBN (Electronic) | 9781510671560 |
DOIs | |
State | Published - 2024 |
Event | Medical Imaging 2024: Image Processing - San Diego, United States Duration: 19 Feb 2024 → 22 Feb 2024 |
Publication series
Name | Progress in Biomedical Optics and Imaging - Proceedings of SPIE |
---|---|
Volume | 12926 |
ISSN (Print) | 1605-7422 |
Conference
Conference | Medical Imaging 2024: Image Processing |
---|---|
Country/Territory | United States |
City | San Diego |
Period | 19/02/24 → 22/02/24 |
Bibliographical note
Publisher Copyright:© 2024 SPIE.
Keywords
- Age-Dependent Generalizability
- Pediatric
- Spine Localization
- Spine Segmentation
- Transfer Learning