Simulation of Three-Dimensional Images and Estimation of Lung Volumes from Two-Dimensional MRI and CT Images

• Published in: ICCSCE 2022 - Proceedings: 2022 12th IEEE International Conference on Control System, Computing and Engineering
• Main Author: Binti Indera Putera S.H.; Bin Dzulkifli M.R.; Sidek N.B.; Abu Bakar Z.B.; Binti Mohammad N.N.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85142439330&doi=10.1109%2fICCSCE54767.2022.9935628&partnerID=40&md5=c5161cf2c4aab6261185f2498f7db66a

Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) are standard imaging techniques used for diagnosis of various medical conditions in clinical settings. They are used to generate two-dimensional (2D) images of internal organs and tissues. Digital Imaging and Communications in Medicine (DICOM) is the standardized practice for processing, transferring, and storing medical images such as CT, x-ray, and MRI. This paper proposes a method to produce three-dimensional (3D) descriptions of the lungs and estimation of the length and volumes of the lungs. The 3D image generation and estimation of lung volumes were performed using simple image processing tools in Matlab® on two sets of 2D DICOM protocol images of the thorax taken from different healthy volunteers. Two sets of images are used in this paper; a set of 2D MRI slices a set of of 2D CT images. The DICOM images are obtained from the Sheffield Royal Hallamshire Hospital, United Kingdom. Generation of the 3D images of the lungs were performed by determining the grey-scale equivalent values for the lung tissues and setting the threshold levels for the lung tissues. The grey-scale images are converted into binary images and the estimated 3D images are rendered. Information from the DICOM image metafile such as the pixel equivalent area, calibration factor, slice thickness, and the size of the reconstructed areas were used to estimate the lengths and volumes of the lungs. Extrapolation of the estimated lungs were made using linear regression and second order polynomial regression analysis to ensure all areas of the lungs were considered in the lung volume estimations. The resulting volume estimations were between 2588ml and 3273ml for the MRI images and between 1891.55ml and 2223.84ml for the CT images. © 2022 IEEE.