Clinical relevance of augmented statistical shape model of the scapula in the glenoid region.

摘要: Abstract Objective: To illustrate (a) whether a statistical shape model (SSM) augmented with anatomical landmark set(s) performs better fitting and provides improved clinical relevance over non-augmented SSM (b) which set the best augmentation strategy for predicting glenoid region of scapula. Methods: Scapula was built using 27 dry bone CT scans three sets (16 landmarks each) resulting in SSMs (aSSMproposed, aSSMset1, aSSMset2). The were then used non-rigid registration (regression) algorithm to fit six external scapular shapes. prediction error by each type evaluated goodness (mean error, root mean square Hausdorff distance Dice similarity coefficient) four angles (critical shoulder angle, lateral acromion inclination, glenopoar angle). Results: Inter- intra-observer reliability selection moderate excellent (ICC>0.74). Prediction significantly lower SSMnon-augmented (0.9 mm) (1.15 mm) distances. coefficient higher (0.78) aSSMproposed compared all other types. lowest critical angle (3.4°), inclination (2.6°), (3.2°). Conclusion Significance: conventional robustness criteria or do not guarantee accuracy may be crucial certain applications pre-surgical planning. This study insights into how region-specific can provide relevance.