Outer-Boundary Assisted Segmentation and Quantification of Trabecular Bones by an Imagej Plugin

The overall goal of this procedure is to quantify structural measurements of trabecular bones automatically with accuracy and efficiency. This method can help answer key questions in the image analysis field, such as how to measure irregular structures accurately with efficiency. The main advantage of this technique is that structural measurements of two-dimensional or three-dimensional objects by such a technique are more accurate and efficient than the current available quantification approaches.

To begin this procedure, first install the ImageJ software and trabecular analysis plugins, as outlined in the text protocol. Open the ImageJ software. Under Plugins, BoMomics, Simulate Objects, click the Circle button.

In the resulting popup window, enter 200 as the diameter, and then click OK to generate a simulated circle with a diameter of 200 pixels. Save the generated circle in TIFF format. Next, in Plugins, BoMomics, Simulate Objects, click the Square button.

In the popup, enter 200 and the side length, and click OK to generate a simulated square with a side length of 200 pixels. Save it in TIFF format. Then, in the same menu, click the Sphere button.

Enter 30 as the diameter, and click OK to generate the simulated sphere. Click the Plugins, 3D, Volume Viewer to view the sphere, and save it in TIFF format. Finally, in Plugins, BoMomics, Simulate Objects, click the Cylinder button.

Enter 30 as the diameter and 100 as the height, and then click OK to generate the simulated cylinder. After this, click Plugins, 3D, Volume Viewer to view the cylinder, and save it in TIFF format. First, open the ImageJ software, and open or import a scanned image.

Slide the bottom scroll bar to choose a slice, and then click the Image, Adjust, Threshold button. In the Threshold popup window, adjust the minimum and maximum threshold values to ensure that the bones are well separated from the background. Record the minimal threshold value as the cortical bone threshold value.

Next, click the Plugins, BoMomics, Trabecular Parameter Profiling button. In the popup window, set the Slice Index to the position of the representative slice. Set the Cortical Bone, Range, and Step values to calculate a set of cortical thresholds for profiling segmentation parameters.

After this, set the Noise Diameter, Step, and Range values to specify a set of noise values for the analysis. Set the Hole Diameter, Step, and Range values for calculating a set of hole values. Click OK to perform the parameter profiling.

In the Parameter Profiling Results window, visually check the segmentation results, and select a slice layer where the bone's outer boundary is outlined accurately. Then, retrieve the profiling parameters from the corresponding entry in the Parameter Profiling Results table. To begin analyzing the trabecular bones, open the ImageJ software and open or import a scanned image.

Click the Plugins, BoMomics, Trabecular Segmentation button. Fill in appropriate analysis parameters, as shown here. Next, click OK to perform the trabecular segmentation.

Visually inspect the results in the Trabecular Segmentation Results window. Then, save the extracted trabecular bones shown in the Segmented Trabecular Bones window in TIFF format for further analysis. After this, click the Plugins, BoMomics, Trabecular Analysis button, and fill in the appropriate analysis parameters, as outlined in the text protocol.

In the Results Reporting section, select one or more parameters to be measured where the options are trabecular bone volume, total volume, and thickness measured either two-dimensionally or three-dimensionally. To perform trabecular analysis, select the checkboxes for both 2D and 3D, and then click OK.To begin quantifying objects, open a simulated image in the ImageJ software. Select the Plugins, BoMomics, Trabecular Analysis button, and fill in the appropriate analysis parameters, keeping the default values for Start, End, Outline Boundary, and Trabecular Bones, while setting Noise Reduction Diameter, Hole Filling Diameter, and Cortical Thickness Diameter to zero.

In the Results Reporting section, select both 2D and 3D as the parameters to be measured. Then, click OK to perform trabecular analysis on the simulated object. After this, calculate the calibrated bone volume, total volume, bone mineral content, bone volume fraction, and bone mineral density in spreadsheet columns, and analyze the data, as outlined in the text protocol.

In this study, an ImageJ plugin is used to automatically segment and quantify trabecular bones. Representative parameter profiling analysis using different parameter conditions reveals that some combinations are more accurate when delineating a bone's outer boundaries than others. Next, segmentation and analysis is performed to quantify measurements of the trabecular bones.

The results of this segmentation can be visually checked slice-by-slice. The raw quantifications of the bone volume, the total volume, the sum of gray values, and the thickness, either in two or three dimensions, can be reported depending on which options are selected during analysis. The calibration information is extracted from the scanned micro-CT dataset, and calibrated measurements of bone volume, total volume, bone mineral content, bone volume fraction, and bone mineral density are then calculated.

Their distributions can then be profiled in the selected analyzing region, layer-by-layer, against the layer positions. Once mastered, this technique can quantify trabecular bones of 500 image layers in 10 to 20 minutes if it's performed properly. We first had the idea for this method when we found that the results reported by the micro-CT vendor software were not reproducible when the same dataset was analyzed by the same experienced operator.

After watching this video, you should have a good understanding of how to choose appropriate bone segmentation parameters and perform analysis of long bones with minimal user interactions.