Reverse shoulder arthroplasty is effective surgery because most of patients have positive long-term results. However, the search for the «perfect» endoprosthesis continues. Objective. To justify the dimensions of a new modular reverse shoulder endoprosthesis using additive technologies based on spiral computed tomography data. Methods. Two data sets of healthy shoulder joints (right — R, left — L) of 100 patients obtained on a spiral computed tomography AQUILION 128 sections (Toshiba, Japan) were processed. Each set consisted of 11 morphometric parameters — linear and angular values. For each of them, three data samples (combined, R and L) are calculated: minimum, maximum, mode, median, mean, standard deviation, distribution asymmetry coefficient. Pearson’s correlation coefficient was calculated, cluster analysis was performed. Results. It is proved that most of the parameters of R and L data sets can be considered homogeneous and can be analyzed as a combined group of 200 cases. It was found that the width and height of the glenoid are more homogeneous data sets, and the value of the endosteal diameter of the humerus decreases in the distal direction. The cervical-diaphyseal angle averages 137.4° ± 4.66°. The correlation between different parameters is more pronounced within most clusters than in the sample as a whole. Conclusions. It is necessary to create different sizes of the distal part of the conical stem, to which securely fix a wide proximal part, as well as in different sizes, in the form of a cup for fixing the liner. The height of the proximal part of the reverse shoulder endoprosthesis should be not less than 20 mm, the diameter of the base of the proximal parts of the stem — 38, 40, 42 mm. It is proposed to use a conical stem of the implant with a wider proximal part, to create the angle 135° between the cup of the proximal part and the stem. Three standard sizes of basic glenoid plates with a diameter of 26, 30, 32 mm are defined. Key words. 3D-printing, arthroplasty of the shoulder joint, glenoid, cluster analysis, correlation analysis.
On Correlated Sources of Uncertainty in Four Dimensional Computed Tomography Data Sets
