Possibilities of Automated Diagnostics of Odontogenic Sinusitis According to the Computer Tomography Data

Abstract

Individual anatomical features of the paranasal sinuses and dentoalveolar system, the complexity of physiological and pathophysiological processes in this area, and the absence of actual standards of the norm and typical pathologies lead to the fact that differential diagnosis and assessment of the severity of the course of odontogenic sinusitis significantly depend on the measurement methods of significant indicators and have significant variability. Therefore, an urgent task is to expand the diagnostic capabilities of existing research methods, study the significance of the measured indicators, and substantiate the expediency of their use in the diagnosis of specific pathologies in an automated mode. Methods of digital filtering, image segmentation and analysis, fluid dynamics, and statistical and discriminant analysis were used. Preliminary differential diagnosis of odontogenic sinusitis can be performed by densitemetric analysis of tomographic images of the maxillary sinuses, performed using frontal multiplanar reconstructions according to a given algorithm. The very manifestation of the characteristic changes in the densitography of the maxillary sinus allows for the initiation of certain pathological processes and permits the development of the effectiveness of the diagnosis of the pathology of the sinus sinuses, which can be realized automatically in real life.

Keywords: bone densitography; discriminant analysis; pathology diagnosis; sinusitis.

Figure 1

Examples of densitographic analysis according to spiral computed tomography: normal (a) and in the presence of a cyst of odontogenic origin in the maxillary sinus (b) (above is a tomographic section in coronary projection with the trajectory of the densitogram).

Figure 2

This is a figure. Schemes follow the same formatting. If there are multiple panels, they should be listed as: (a) Description of what is contained in the first panel; (b) Description of what is contained in the second panel. Figures should be placed in the main text near to the first time they are cited. A caption on a single line should be centered.

Figure 3

Segmentation of the airways of the nasal cavity according to tomographic data and a schematic illustration of the overlap of the natural anastomosis of the maxillary sinus: (a) segmentation of the airways of the nasal cavity according to tomographic data in the axial projection at the level of the middle nasal passage: 1—airways, 2—maxillary sinus, 3—natural anastomosis of the maxillary sinus sinuses (for the left side of the nasal cavity); (b) a schematic illustration of the overlap of the anastomosis of the maxillary sinus: from left to right, with a working (unblocked) anastomosis, 60% free, 30% free, and with completely blocked anastomosis.

Figure 4

Schematic illustration of the maxillary sinus in a sagittal projection: (a) at a conditional rate, (b) at 60% filled sinus cavity with polyposis content; (c) when filling the sinus by 40% liquid content; (d) filling of the sinus with liquid and polyposis contents by 30% and 20%, respectively.

Figure 5

Illustration of a set of frontal sections of the air channel of the nasal cavity in a sagittal projection obtained as a result of segmentation of tomographic data.

Figure 6

The results of discriminant data analysis when comparing the norm with acute serous odontogenic sinusitis: (a) the dependence of the increase in the normalized Euclidean distance as signs are added to the discrimination model $\mathsf{\delta }=\mathrm{f}(j)$; (b) the dependence of the decrease in the probability of a decision error as signs are added to the discrimination model ${\mathrm{p}}_{\mathrm{er}}^{}=\mathrm{f}(j)$: j = 5 is the dimension of the space of informative parameters.

Figure 7

The results of discriminant data analysis when comparing the norm with acute purulent odontogenic sinusitis: (a) the dependence of the increase in the normalized Euclidean distance as signs are added to the discrimination model $\mathsf{\delta }=\mathrm{f}(j)$; (b) the dependence of the decrease in the probability of a decision error as signs are added to the discrimination model ${\mathrm{p}}_{\mathrm{er}}^{}=\mathrm{f}(j)$: j = 5-is the dimension of the space of informative parameters.

Figure 8

Results of discriminant analysis of data when comparing the norm with chronic odontogenic sinusitis: (a) dependence of the increase in the normalized Euclidean distance as signs are added to the discrimination model $\mathsf{\delta }=\mathrm{f}(j)$; (b) -the dependence of the decrease in the probability of error in decision-making as features are added to the discrimination model ${\mathrm{p}}_{\mathrm{er}}^{}=\mathrm{f}(j)$: j = 5-is the dimension of the space of informative parameters.

Figure 9

The results of discriminant data analysis when comparing the norm with exacerbated chronic odontogenic sinusitis: (a) the dependence of the increase in the normalized Euclidean distance as signs are added to the discrimination model $\mathsf{\delta }=\mathrm{f}(j)$; (b) the dependence of the decrease in the probability of decision-making errors as features are added to the discrimination model ${\mathrm{p}}_{\mathrm{er}}^{}=\mathrm{f}(j)$: j = 5-is the dimension of the space of informative parameters.

Figure 10

Results of discriminant analysis of data in the differential diagnosis of forms of acute serous and purulent odontogenic sinusitis (a,b), forms of chronic and acute purulent odontogenic sinusitis (c,d), forms of acute purulent and exacerbated chronic odontogenic sinusitis (e,f), and forms of chronic and exacerbated chronic odontogenic sinusitis (g,h): first column—the dependence of the increase in the normalized Euclidean distance as features are added to the discrimination model $\mathsf{\delta }=\mathrm{f}(j)$; second column—the dependence of the decrease in the probability of error in decision-making as features are added to the discrimination model ${\mathrm{p}}_{\mathrm{er}}^{}=\mathrm{f}(j)$: j = 5 is the dimension of the space of informative parameters.

Figure 10

Results of discriminant analysis of data in the differential diagnosis of forms of acute serous and purulent odontogenic sinusitis (a,b), forms of chronic and acute purulent odontogenic sinusitis (c,d), forms of acute purulent and exacerbated chronic odontogenic sinusitis (e,f), and forms of chronic and exacerbated chronic odontogenic sinusitis (g,h): first column—the dependence of the increase in the normalized Euclidean distance as features are added to the discrimination model $\mathsf{\delta }=\mathrm{f}(j)$; second column—the dependence of the decrease in the probability of error in decision-making as features are added to the discrimination model ${\mathrm{p}}_{\mathrm{er}}^{}=\mathrm{f}(j)$: j = 5 is the dimension of the space of informative parameters.