Systemic sclerosis & Raynaud's
Automated Hand Flexor Tendon–Thickness Measurement in Systemic Sclerosis
Authors' conclusion
Automated analysis of longitudinal hand ultrasonography images enables accurate measurement of flexor tendon thickness.
Study context
- Population
- Patients with systemic sclerosis (SSc); longitudinal hand US images from the multicenter HANDSOME cohorttraining 59, testing 15 individuals
- Comparison
- Automated deep learning-based flexor tendon-thickness measurement vs manual expert measurement
- Outcome
- Agreement and accuracy of automated tendon-thickness measurement (testing dataset)
Methodological detailsproof-of-concept, multicentre cohort, deep learning (nnU-Net), independent testing dataset
- Proof-of-concept
- Multicentre cohort
- Deep learning (nnU-Net)
- Independent testing dataset
Performance
As reported in the abstract
- Automated measurements successfully obtained
- 100%
- Dice similarity coefficient, flexor tendon
- 0.90
- Dice similarity coefficient, proximal phalanx base
- 0.83
- Dice similarity coefficient, metacarpal head
- 0.76
- ICC, automated vs manual thickness
- 0.83
- Mean absolute measurement error
- 0.25 mm
- Relative measurement error
- 8.0%
- Automated segmentations clinically acceptable (visual expert evaluation)
- 90%
Original publication
Automated Hand Flexor Tendon–Thickness Measurement in Systemic Sclerosis. Arthritis Care Res (Hoboken). Published 25 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →