Age- and sex-related changes in proximal humeral volumetric BMD assessed via chest CT with a deep learning–based segmentation model
Authors' conclusion
We present a novel, automated method for quantifying proximal humeral vBMD from chest CT, revealing distinct, sex-specific aging patterns. Males' humeral vBMD declines linearly, while females experience an earlier, accelerated loss. Moreover, the peak humeral vBMD in females occurs later than that of the lumbar spine, and spinal measurements cannot reliably substitute for humeral BMD in clinical assessment.
Study context
- Population
- Adults with chest CT scans; lumbar spine volumetric bone mineral density (vBMD) measured in 1460 for correlation
- Participants
- 2,675 adults
- Exposure
- Males vs Females
- Outcome
- Segmentation performance and age-related proximal humeral vBMD changes
Methodological detailsdeep learning-based segmentation (nnU-Net), phantom-calibrated vBMD, generalized additive models for location, scale, and shape (GAMLSS) percentile modelling
- Deep learning-based segmentation (nnU-Net)
- Phantom-calibrated vBMD
- GAMLSS percentile modelling
Main finding
As reported in the abstract
Dice Similarity Coefficient (DSC)
98.42 ± 0.20%
Intersection over Union (IoU)
96.89 ± 0.42%
95HD
1.12 ± 0.37 mm
Average Symmetric Surface Distance (ASSD)
0.94 ± 0.31 mm
Observational data: associations, not effects.
Further findings
- In females, vBMD declined earlier in the lumbar spine than in the proximal humerus
- Correlations between proximal humeral and lumbar spine vBMD low to moderate overall, weakened after age 50
Original publication
Composed by AI from the published abstract and checked against it. How Briefs are made →