Related Experiment Video
Updated: Jun 13, 2026
06:45
Automated Joint Space Detection Improves Bone Segmentation Accuracy
Published on: November 28, 2025
Interrater Reliability of Partially Automated Segmentation of Spinal Radiographs in Adult Spinal Deformity Patients
Alyssa A Federico1, Manjot Birk1, Rémi Pelletier-Roy2
1Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada; Division of Orthopaedic Surgery, Department of Surgery, University of Calgary, Calgary, Alberta, Canada.
World Neurosurgery
|June 11, 2026
Summary
The Knowledge Environment for Orthopedic Personalized Surgery (KEOPS) software shows excellent reliability for measuring spinopelvic parameters in adult spinal deformity (ASD) patients. However, image quality issues can affect accuracy in the cervical spine.
Area of Science:
- Orthopedics
- Radiology
- Spine Surgery
Background:
- Spinopelvic radiographic measurements are crucial for adult spinal deformity (ASD) evaluation and treatment.
- Automated software like KEOPS improves measurement speed and reliability.
- Variable image quality can impact the clinical reliability of these measurements.
Purpose of the Study:
- To assess the inter-rater reliability of partially automated spinopelvic parameter measurement using KEOPS software in ASD patients.
- To evaluate the consistency of measurements across different spine specialists.
Main Methods:
- Full-length spinal radiographs of five ASD patients were analyzed.
- Five spine surgeons, a fellow, and a resident performed manual segmentation.
- KEOPS software automated the measurement of spinopelvic parameters.
- Inter-rater reliability was determined using intraclass correlation coefficients (ICCs).
Main Results:
- Excellent inter-rater reliability was found for all spinopelvic parameters, with ICCs ranging from 0.954 to 0.994.
- Pelvic tilt showed the highest reliability (ICC=0.994), while C2-C7 lordosis had the lowest (ICC=0.954).
- Segmentation success rates were lower for the cervical spine (sagittal: 88-97%, coronal: 51-61%) due to poor visualization.
Conclusions:
- The KEOPS software demonstrates excellent inter-rater reliability for spinopelvic parameters in adult spinal deformity.
- Automated measurements using KEOPS can enhance consistency in clinical practice.
- Cervical spine segmentation remains a challenge, highlighting the impact of image quality on automated measurements.
