Avizo 9

Color-coded depth projections of 3D images were created via ImageJ. Blinking signal in Fig. 2d was rendered in Imaris (Bitplane). We further used the Volume Rendering tool in Avizo 9.4 (Thermo Fisher Scientific Inc) for 3D visualization of whole brain dopamine neurons (Fig. 5). For single neuron segmentation, as depicted in Fig. 5, we used the Brush tool in the segmentation mode of Avizo 9.4 (Thermo Fisher Scientific Inc) to manually select the boundary of each single neuron to generate a single neuron boundary mask. We further used the arithmetic tool to intersect the original data with the manually segmented single neuron mask to generate single neuron images with the original gray scale. In Figs. 5d and 6, we used the Material Statistics module in Avizo 9.4 (Thermo Fisher Scientific Inc) to quantify the volume of the MB sectors, DPM neurons, and VMAT protein expression. This module calculates the voxel numbers inside a labeled area, which can be transformed into volumes by multiplying with a known voxel size in each image. We manually segmented the boundaries of the MB sectors using the Lasso tool in segmentation mode. We used the Magic Wand tool to select one seed within the DPM neuron and to determine a reasonable threshold for selecting the connecting voxels. We directly used the Threshold tool for whole-volume VMAT images, setting the lower bound to 78 (8-bit; 0-255).

Thermo Scientific Avizo 9.5 software (Huston, TX, USA) was used to filter, segment, and quantify all phases present in the image: new bone, biomaterial, connective tissue (CT), and background. Before segmentation, a non-local means filter was applied in order to reduce the reconstruction artifacts and facilitate targeting [34 (link)]. The attenuation of the X-ray beam when passed through the sample, depending on the density of each sample structure, resulted in differentiation in the grayscale images. Thus, we obtained different grayscale ranges for resident bone together with the newly formed bone, the biomaterial, the connective tissue, and the background.
The background was removed with an interactive threshold tool. In the first segmentation step, a quick segmentation was performed by single thresholding to create seeds attributed to each phase present in the image. The previously created seeds were the input for the fine-tuned watershed-based segmentation of the bone and biomaterial phases [35 (link)]. A Lasso 3D tool masked the new bone region located in the defect cavity, separating it from the pre-existing bone. CT was segmented using a magic wand tool, masking the corresponding threshold range. After the segmentation, all phases (pre-existing bone, new bone, biomaterial, and connective tissue) were visualized using Thermo Scientific Avizo 9.5.