Skip to content

A deep-learning based automatic framework for full-length multi-pose radiographs of lower limbs: K-L grading, posture classification, and ergonomic parameters measurement

A deep-learning based automatic framework for full-length multi-pose radiographs of lower limbs: K-L grading, posture classification, and ergonomic parameters measurement

Status
Recruiting
Phases
Early Phase 1
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2100051224
Enrollment
Unknown
Registered
2021-09-16
Start date
2021-09-20
Completion date
Unknown
Last updated
2022-05-30

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Biomechanically related diseases of the lower limbs

Interventions

Gold Standard:1. Full-length CT of both lower extremities + three-dimensional reconstruction: the gold standard
2. Full-length coronal X-rays of both lower extremities: the most commonly used diagnostic method, but the parameters measured by "non-standard" coronal X-rays are inaccurate
3. The full-length coronal X-ray film of both lower
Index test:A&#32
based&#32
framework&#32
for&#32
full-length&#32
multi-pose&#32
radiographs&#32
of&#32
limbs.

Sponsors

The Affiliated Hospital of Guizhou Medical University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
10 Years to 85 Years

Inclusion criteria

Inclusion criteria: 1. Inpatients requiring full-length coronal X-ray examination of lower extremities, aged 10-85 years old, with no gender restrictions; 2. Inpatients who need full-length CT and X-ray examination of lower extremities, age 10-85 years old, and gender is not limited.

Exclusion criteria

Exclusion criteria: 1. The quality of full-length X-ray films of both lower extremities or full-length CT of both lower extremities is poor, which obviously affects the image data of automatic frame model training and testing; 2. Patients who are unwilling to participate in the trial.

Design outcomes

Primary

MeasureTime frame
Multipose classification of full-length X-ray radiographs of lower limbs;Multiple biomechanical parameters(Multiple ergonomic parameters);K-l grading of knee joint;sensitivity;specificity;precision;reliability;ACC;ROC, AUC;

Countries

China

Contacts

Public ContactWang Jianji

The Affiliated Hospital of Guizhou Medical University

1063802482@qq.com+86 18786088092

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026