The constitution system of diagnostic characters usingoutcome analysis of boundary circular-shaped masses in lungs

Vesnina Zh.V., Krivonogov N.G., Arsenjeva Yu.A., Nesterov E.A., Lishmanov Yu.B.

Purpose. The article investigates the possibility of using the information about the images contour characteristics of pathological spherical formations in lungs for differential illness diagnosis.

Materials and methods. The two-dimensional array of densitometric parameters obtained during the processing of tomographic medical images of patients with verified diagnoses of lung diseases (cancer and tuberculosis, 49 and 26 patients respectively) was analyzed. The technique of constructing the contours of images of such formations is developed. To introduce quantitative characteristics of the contour, the conditional center of mass of the pathology image is calculated and two one-dimensional signature functions are constructed: of the radius-vector module drawn from the center of the image of the spherical formation to its boundary, as well as the polar angle of rotation of the radius-vector. The argument of both functions is the conditional number of the pixel belonging to the path. With the help of spectral analysis of the signature function, such characteristics of contours as spectral entropy, multiplicity, number of local maximums, signature functions of the polar angle of rotation of the radius vector are proposed and calculated in the paper. The statistical significance of differences in values of these characteristics of contours of spherical formations in different diseases (cancer and tuberculosis) was analyzed.

Results. There were statistical differences in the characteristics of the contour for different pathologies proposed in the work, which can allow to perform differential diagnosis of the investigated lung diseases.

Conclusion. The obtained quantitative characteristics of the spherical formations in lungs can enter the system of assistance to the doctor in diagnosis. It should be remembered that there is no single diagnostic feature based on the quantitative characteristic of the image, which would immediately determine the type of pathology. Only a set of such features will allow to carry out high-quality diagnostics.

1 - Altai Regional Oncological Center.

2 - Diagnostic Center of the Altai Krai.

3 - Altai State Medical University" of the Ministry of Health of the Russian Federation.

4 - Altai State University.

6 - Polzunov Altai State Technical University.

Barnaul, Russia.

 

Keywords: computer tomography, medical images, contour characteristics, differential diagnosis, cancer, tuberculosis.

 

Corresponding author: Molodkin I.V., e-mail: Этот e-mail адрес защищен от спам-ботов, для его просмотра у Вас должен быть включен Javascript

 

For citation: Molodkin I.V., Yakunin A.G., Shayduk A.M., Konovalov V.K.,

Ostanin S.A., Leonov S.L., Borisenko O.V. The constitution system of diagnostic characters usingoutcome analysis of boundary circular-shaped masses in lungs. REJR 2019; 9(3):79-85. DOI:10.21569/2222-7415-2019-9-3-79-85.

 

 

Статья получена:       23.04.19                 Статья принята:       14.06.19