HE METHOD OF VOLUMETRIC VISUALIZATION OF BONE DENSITY

Valerii Albertovich Dzhidzalov¹, Maryana Arturovna Bzhihova², Zaurbek Salambekovich Agaev¹, Gadzhimurat Magomedzapirovich Umarov^1*, Bolatkhan Abdurashidovich Tovzerkhanov¹, Linda Ruslanbekovna Djamaldinova²

¹Department of Therapy, Faculty of Dentistry of North Ossetian State University named after K.L. Khetagurov, Vladikavkaz, Republic of North Ossetia-Alania, Russia. [email protected]

²Department of Therapy, Faculty of Dentistry, North Ossetian State Medical Academy, Vladikavkaz, Republic of North Ossetia-Alania, Russia.

https://doi.org/10.51847/2pnOTO5Qu6

ABSTRACT

The work aimed to optimize the technique of spatial digital analysis of bone density For the experiment, spiral computed tomography of the facial skeleton of 20 patients aged 10 to 22 years with maxillofacial pathology was performed. Spiral computed tomography technique (CT): the plane of the CT slice is parallel to the plane of the Frankfurt horizontal or the plane of the bite. The proposed technique makes it possible to visualize and, depending on the density, divide the bone tissue within each scan into a certain range of spectra, which makes it possible to better identify the zones with the highest and lowest density of bone tissue. The possibility of selecting a separate bone layer with a certain density range allows its reconstruction and tracing of its boundaries on a computer volumetric model. This technique can be successfully used in maxillofacial surgery, orthodontics, periodontics, and implantology to determine the condition of the bone tissue of the jaws to plan therapeutic measures.

Key words: Bones, X-ray density, Computed tomography, 3D visualization.

Introduction

To draw up a plan for both orthodontic and surgical treatment of bone pathologies of the dental system, criteria such as the patient's age, concomitant somatic diseases, localization of the pathological process, and the condition of the bone tissue in the affected area are increasingly taken into account . The latter is the most important in determining the type and scope of surgical intervention. Existing methods of X-ray studies allow us to assess the condition of the facial skeleton, both in two-plane projection (radiography) and in three-plane projection (spiral computed tomography), which produces a detailed high-resolution image of bone tissue on axial, frontal, and sagittal sections and in a volumetric SSD computer reconstruction [1-3]. This makes it possible to determine not only the exact localization and prevalence of the pathological process in the bone tissue but also to differentiate it by density [4, 5]. The software that currently exists allows marking bone tissue with a color index on the axial section according to the density gradient, which makes it possible to analyze bone density only within the axial plane [6-8]. This disadvantage does not provide a full-fledged stereoscopic spatial representation of the structure of bone tissue, and in some cases creates a false idea of the location of density zones due to the different angles of inclination of the scanning plane [9, 10]. If the latter error can be solved by observing the standard placement of the axial plane during scanning, then the spatial placement of bone density fields requires the development of additional techniques [11, 12]. The purpose of the study was to optimize the methodology of spatial digital analysis of bone density.

Materials and Methods

Archival data from spiral computed tomography (CT) studies of 20 patients aged 10 to 22 years with maxillofacial pathology were analyzed. To diagnose and plan treatment, spiral computed tomography of the facial skeleton was performed using the following technique: the CT plane of the slice is parallel to the plane of the Frankfurt horizontal or the plane of the bite (Figure 1).

The computed tomography protocol for bone tissue identification was formed in accordance with another work [13]. The Gantry tilt angle was 0°. The reconstruction algorithm was billed as "bone" or "high resolution". The expansion of the matrix was 512 × 512. the thickness of the slice is 1 mm, the rotation step is 1 mm, and the step during the reconstruction of the slice is 1 mm. The archived data was stored in DICOM format, and the analysis was carried out in the MATLAB R2006a software package and DICOM Works 1.3.5. more than 300 CT sections were analyzed, which were subjected to special processing in a computer program that included the darkening of soft tissues and marking of bone tissue (Figure 2).

Each range of bone density was assigned its color. Thus, the bone tissue was the largest in density and the teeth were marked in black, and the smallest in density was yellow. In the range between them, in ascending order, bone density was marked from green to blue in accordance with the color differentiation presented in the Guidelines for the diagnostic management of incidental solitary bone lesions on CT and MRI [14].

To map the spatial placement of bone density zones based on the MATLAB software platform, a mathematical algorithm has been developed for the degree of darkening of the image matrix with an upper and lower border, which makes it possible to isolate bone tissue of a certain density range and carry out their computer reconstruction independently of tissues of another density range. In this way, a solid-state computer reconstruction with ring-marked zones of four degrees of bone density was obtained. In the computer program DICOM Works 1.3.5, densitometric analysis of a series of axial slices was carried out to assign the units of Hounsfield (HU) to the corresponding color spectra.

Results and Discussion

To further determine the correspondence of bone density to certain color indicators, the image of the axial section of bone tissue is marked in the HSV color range. On the first series of axial sections, an image with the lowest density of bone tissue, marked with yellow and its shades, with a range of bone density of 400-162 HU was obtained (Figure 3a).

With the degree of dimming of the matrix of the second series of slices with an upper boundary of 100 units and lower boundary of 50 units in HSV mode, an image marked with a green color spectrum and its shades with a density range of 678-469 HU was obtained (Figure 3b). With the degree of dimming of the matrix of the third series of axial sections with an upper boundary of 190 units and a lower boundary of 100 units in HSV mode, an image marked with a blue color spectrum and its shades with a bone density range of 1176-775 HU was obtained (Figure 4a).

The selection of the last series of axial slices encoded by the HSV color format is not possible due to the fusion of the background and the bone marked with red with its shades of color, which significantly complicates visual analysis and makes further mathematical calculations not reliable (Figure 4b). In this regard, the last range of bone density was encoded with the RGB (jet) color format. With the degree of dimming of the matrix of the fourth series of axial slices with an upper boundary of 280 units and a lower boundary of 190 units in RGB mode with excluded median and adaptive filtering, an image of bone tissue marked with a spectrum of red with its shades of color was obtained (Figure 5).

To obtain a color-marked volumetric image of bone tissue, SSD will reconstruct each color component that corresponds to a certain range of bone density (Figure 6a). Combining SSD reconstruction of each color component, a volumetric reconstruction of the bones of the facial skeleton was obtained, marked in red, which corresponds to bone tissue with a density of 2107-1369 HU and yellow with a density of 400-162 HU (Figure 6b).