Validation of new clinical quantitative analysis software applicable in spine orthopaedic studies

Area of Science:

• Radiology
• Medical Imaging Analysis
• Biomechanical Analysis

Background:

• Accurate measurement of spinal alignment from X-ray films is crucial for diagnosing and managing conditions like scoliosis.
• Software tools aim to improve the precision and reproducibility of these measurements.
• Evaluating the reliability and validity of such software is essential for clinical adoption.

Purpose of the Study:

• To evaluate the reliability and validity of a specific X-ray films analysis software for clinical measurements.
• To assess the software's precision, reproducibility, and clinical applicability.
• To identify factors affecting measurement accuracy, particularly in complex cases.

Main Methods:

• Conducted precision and reproducibility tests on dynamic and postural modules of the software.
• Calculated mean reproducibility coefficients for angular and linear parameters.
• Assessed software validity by comparing reference values from healthy subjects with literature data.
• Analyzed inter-observer reproducibility on X-ray films of severe scoliotic patients.

Main Results:

• Software precision was +/-2 degrees (lumbar) and +/-3 degrees (cervical) for dynamic modules.
• Mean reproducibility coefficients were approximately 4 degrees (angular) and 3 mm (linear) for postural modules.
• Reference values for healthy subjects agreed with existing literature.
• Inter-observer reproducibility was lower in severe scoliosis cases due to poor landmark visibility.

Conclusions:

• The X-ray analysis software demonstrates good reliability and validity for clinical measurements in general.
• Measurement reliability decreases in severe scoliosis, primarily due to challenges in identifying anatomical landmarks on X-ray films.
• The software is a useful tool, potentially facilitating multicentric studies and data exchange.