Automated quantification of glenoid bone loss in CT scans for shoulder dislocation surgery planning

Area of Science:

• Radiology
• Orthopedic Surgery
• Medical Imaging Analysis

Background:

• Shoulder dislocation often results in glenoid bone loss, necessitating accurate quantification for surgical planning.
• Current methods for assessing glenoid bone loss can be time-consuming and subjective.

Purpose of the Study:

• To present a novel, fully automatic method for computing glenoid bone loss from computed tomography (CT) scans.
• To provide an accurate and efficient tool for quantifying bone loss to assist in surgical decision-making for shoulder instability.

Main Methods:

• A four-step model-based pipeline was developed, including oblique plane computation, glenoid slice selection, best-fit circle computation using Glenoid Clock Circle Constraints, and bone loss quantification.
• The method was evaluated on 51 shoulder CT scans, with manual measurements from three clinicians serving as ground truth.

Main Results:

• The automated pipeline achieved a mean absolute error of 4.67 ± 3.32% for glenoid bone loss quantification.
• Oblique CT slice selection accuracy was comparable to inter-observer variability.
• The method demonstrated high accuracy in measuring glenoid bone loss, with errors close to observer variability.

Conclusions:

• This is the first fully automatic method for quantitative analysis of glenoid bone loss in CT scans.
• The automated glenoid bone loss report can significantly assist orthopedic surgeons in planning surgical interventions for shoulder dislocations.