Quantification of fat in the posterior sacroiliac joint region applying a semi-automated segmentation method.

摘要: Abstract Background and objective Fat within the posterior sacroiliac joint region (PSIJ) is thought to compensate for incongruent surfaces of sacrum ilium posteriorly. Knowledge on presence fat in SIJ could provide useful information about physiology clinical kinematic implications its presence. This study aimed at quantifying PSIJ, using a semi-automated method, compare results manual segmentation method based data from frozen cadaveric sections computed tomography (CT). The may quicker more volume quantification. Methods Seventy-eight hemipelves were used. Frozen obtained photographed CT subsamples. A MATLAB routine was deployed assess serial scans, masks derived color thresholds Hounsfield units, respectively. Regions interest created isolate PSIJ before computed. Friedman test used comparison between all Kruskall-Wallis comparing with Bland-Altman plots express result differences these methods. Results averaged 3.9 ± 2.2, 4.9 ± 2.5, 3.7 ± 2.3 7.2 ± 7.3 cm3 1 (fat mask), 2 (no-fat 3 (‘control’ mask) CT, All significantly different (p Conclusions Manual relative underestimation total compared or CT-based methods, as might not be sufficiently distinguished surrounding structures. However, resulted vastly higher variation warrants further study. approach quantify presented here investigator-independent, time efficient applicable which provides opportunity use this technique various tissue types vivo.