Computed tomography images preliminary processing for their analysis by artificial intelligence methods

Nataliia I. Limanova; Sergey V. Palmov; Dmitry A. Morozov

doi:10.1117/12.2680857

4 October 2023 Computed tomography images preliminary processing for their analysis by artificial intelligence methods

Nataliia I. Limanova, Sergey V. Palmov, Dmitry A. Morozov

Proceedings Volume 12743, Optical Technologies for Telecommunications 2022; 127430Y (2023) https://doi.org/10.1117/12.2680857
Event: Optical Technologies for Telecommunications 2022, 2022, Ufa, Russian Federation

Abstract

In this paper we consider method, proposed by the authors, for pre-processing X-ray images based on the convolutional neural network. Multilayer computed tomography scans of the lungs are ubiquitous, used to detect pathology, such as covid-19 diseases. The proposed method consists of computed tomography lung image segmentation, the volume of interest localization at these scans, distinguishing between denser structures, such as bones, and less dense ones, such as blood vessels. A comparison of the pathological formations classification quality was made on images with different degrees of preprocessing on large data sets. Image Classifier software was developed for demonstration of x-rays preliminary processing opportunities in the context of their class association forecasting. The developed software showed quite good results. Maximum value of the average probability reached 77.5%. The number of epochs required to achieve the specified quality level is four.

Citation Download Citation

Nataliia I. Limanova, Sergey V. Palmov, and Dmitry A. Morozov "Computed tomography images preliminary processing for their analysis by artificial intelligence methods", Proc. SPIE 12743, Optical Technologies for Telecommunications 2022, 127430Y (4 October 2023); https://doi.org/10.1117/12.2680857

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available