Nrecon

Each calvaria was scanned using a micro-CT scanner (Skyscan 1076, Bruker, Belgium). The resolution was set at 18 µm with the rotational step of 0.40, beam hardening was reduced by the use of 0.025 mm titanium filter, while the frame averaging was set at 2. The obtained images were reconstructed using the NRecon (Bruker, Kontich, Belgium) software and analyzed using the CTAn (Bruker, Belgium) software. For analysis, a 5 mm diameter circular region was delineated along the margins of the initial defect area. To delineate newly formed bone from grafting material, we set specific thresholds for grafting material, while it was kept constant for bone tissue. The threshold for Cerabone^®, Cerabone^® + Al. bone, and Cerabone^® with magnesium was 200–255; for OsteoBiol^® 70–200; and for newly formed bone, it was 50–255 (Figure 5). Due to the difference in material density, threshold range delineated different graft material. After this was done, subtraction was performed to separate the values for newly formed bone versus biomaterial. The calculated parameters included bone volume fraction (BV/TV, %) and trabecular bone parameters such as trabecular thickness (Tb.Th, mm), trabecular number (Tb. N, 1/mm), trabecular separation (Tb.Sp, mm), total porosity (Po (tot), %), and connectivity density (Conn.D, 1/mm³) along with the percentage of residual biomaterial (RB, %).

Tibiae were carefully excised and fixed in 4% paraformaldehyde for 24 h, rinsed in PBS and stored in PBS. Tibiae were scanned using the Skyscan 1276 (micro-computed tomography imager, Bruker, Kontich, Belgium) at 9 μm resolution, 0.25 mm aluminium filter, 56 kV voltage, 200 μA, 560 ms exposure time, and 0.4° step rotation with frame averaging of 2. Images were reconstructed and analyzed using NRecon (v1.7.4.6, Bruker), Dataviewer (v1.5.6.2, Bruker), CT Analyzer (CTAn; v1.18.8.0, Bruker) and CTVox (v3.3.0 r1403, Bruker). Tibial lengths were determined after scanning. Regions of interest were determined as described previously (Chan et al., 2021 (link)). The trabecular bone was assessed in the proximal region commencing at 3% of bone length from the growth plate and extended distally for a total of 13.5% (equivalent to 0.5-3 mm of the growth plate). For cortical bone, a region of interest beginning at 50% of bone length and extending distally for 2% (approximately 0.5 mm) was used for three-dimensional cortical bone analyses using CT Analyzer. Representative images were taken at the mid-diaphysis (50% of bone length) using a pseudodensity filter in CTVox. The length of the tibial crest was measured using a custom script written in FIJI/ImageJ using individual cross-sectional images spanning 15 to 40% of the tibial length, as described previously (Chan et al., 2023 (link)).

As previously described [6 (link)] 5 mm diameter cylindrical punched samples were analysed using a Skyscan 1272 (Bruker, Belgium) desktop MicroCT system. An initial scan pixel size of 1.5 μm was selected (no applied camera binning), with an operating voltage of 25 kV. These data have been previously published [6 (link)] but scanning was then repeated of the same samples with 2× and 4× camera binning applied, resulting in pixel sizes of 3 and 6 μm.
Resulting projections were reconstructed in NRecon (Bruker, Belgium) and systematic VOIs selected as described previously [6 (link)]. A three-dimensional analysis was carried out in after automatic Otsu thresholding and sweep despeckling in CTAnalyzer.

Following euthanasia, the femoral heads were bisected coronally and fixed in 10% neutral buffered formalin. After fixation, all femoral heads were scanned using a μCT (Skyscan 1172, Bruker-μCT, Kontich, Belgium) at a setting of 100 kV and 100 µA and a resolution of 13.3 µm/pixel as previously described^{37 (link)} and reconstructed with NRecon (Bruker-μCT). The reconstructed images were binarized to a common threshold using CTAn (Bruker μCT), where the region of interest was defined to capture the original necrotic epiphysis. The region of interest was outlined within the epiphysis, avoiding the subchondral region of the calcified epiphyseal cartilage. The CTAn software was used to calculate the three-dimensional morphometric values for percent bone volume, trabecular thickness, trabecular number, trabecular separation, and percent bone void volume in the epiphysis of the femoral heads. To determine a bone void volume, spaces with a trabecular separation larger than 570 μm were defined as bone voids, as the control group had a maximum trabecular separation of 570 μm.