# Internal Density Calibration in the Proximal Humerus to Estimate Bone Stiffness for Stemless Shoulder Arthroplasty

## TL;DR

This study shows that internal calibration methods can accurately estimate bone stiffness in the proximal humerus for stemless shoulder arthroplasty.

## Contribution

The study introduces and validates an internal calibration method for estimating bone stiffness in the proximal humerus without using phantoms.

## Key findings

- Internal calibration methods showed strong correlations with phantom-calibrated vBMD, with ACM having the lowest error.
- Estimated stiffness and vBMD were strongly correlated across calibrations, with ACM showing the lowest error.
- The findings support using internal calibration and FEMs for stiffness estimation in the proximal humerus.

## Abstract

End‐stage osteoarthritis (OA) alters bone density in the humeral head, complicating implant fixation during stemless shoulder arthroplasty. Current preoperative assessments fail to consider the mechanical properties of bone directly supporting the humeral component. While in‐scan phantom calibration is used to determine volumetric bone mineral density (vBMD), phantoms are rarely used clinically. An internal density calibration method has been developed, but not yet applied in the proximal humerus. This study examined correlations between phantom and internal density calibration, and between vBMD and estimated stiffness in the proximal humerus. Nonpathologic cadaveric CT images containing a K2HPO4 phantom were used to analyze a 10 mm region of interest below the anatomic neck. Phantom calibration was performed. Internal calibration used air (A), adipose (Ad), skeletal muscle (M), and cortical bone (C) in three combinations (AAdCM, ACM, AAdC). Finite element models (FEMs) were generated from each. Strong correlations were observed between phantom‐ and internally calibrated vBMD (AAdC R² = 0.80; AAdCM R² = 0.88; ACM R² = 0.90), with ACM showing the lowest error (9.98%). Estimated stiffness and vBMD were strongly correlated across calibrations (R² = 0.61–0.66), with ACM showing the lowest error (5.46%). Findings support internal calibration for determining vBMD and FEMs for estimating stiffness in the proximal humerus.