Sensitivity analysis of hip joint centre estimation based on three-dimensional CT scans

Area of Science:

• Orthopedic surgery
• Biomechanical engineering
• Medical imaging analysis

Background:

• Accurate hip joint centre (HJC) determination is crucial for femur morphological analysis.
• Current HJC methods rely on 3D femoral head models from medical images, but image selection can introduce variability.
• The influence of image segmentation on HJC calculation accuracy requires investigation.

Purpose of the Study:

• To evaluate the impact of image segmentation thresholds on three HJC calculation methods, including a novel approach.
• To assess the sensitivity of HJC results to variations in the selected femoral head portion.
• To compare the performance of different HJC calculation methods, particularly in cases of hip deformity.

Main Methods:

• Utilized cadaver CT images to isolate femoral heads using distance thresholds between femur and acetabulum models.
• Quantified the sensitivity of HJC calculations to varying segmentation thresholds.
• Compared a newly developed HJC method against two existing techniques.

Main Results:

• For healthy hips and thresholds between 6-9 mm, HJC differences were <1 mm, showing method insensitivity.
• Higher thresholds led to inaccuracies due to the fovea capitis femoris affecting HJC.
• The novel method demonstrated superior performance in two cases of deformed hips.

Conclusions:

• Existing and novel HJC calculation methods yield valid results for normal hips within specific segmentation parameters.
• The fovea capitis femoris can disrupt HJC accuracy at higher segmentation thresholds.
• The introduced method shows promise for HJC determination in hips with deformities.