In vivo kinetics X-ray imaging

OData support
Supervisor:
Dr. Szigeti Krisztián
SE Biomedical Engineering MSc

Background:

The early diagnosis of different sorts of pulmonary diseases, such as chronical obstructive pulmonary disease (COPD) and idiopathic pulmonary fibrosis (IPF), can raise the rate of successful therapy for these progressive and high mortality diseases. The diagnosis strongly depends on the result of the HRCT scan, but patients are usually sent for scan in a late stage, because the nonspecific symptons of these diseases. Taking these reasons into account, the importance of finding a fast and cheap diagnostic method, which can be used for early diagnosing of these pulmonary diseases, is unquestionable. The goal of my master thesis is to examine whether kinetic imaging could satisfy these needs.

Methods:

During the kinetic imaging technique a new method based on x-ray imaging is used to detect movements created by life functions and the pathology correlated differences of such movements. First the method of kinetic imaging in silico was tested with a software code (MATLAB).

As the next step preclinical experiments with COPD small animals had been done. For the laboratory modell of COPD, mice were injected intratracheal with elastase (13ul 3,6 mg/ml elastase + 87ul salsol, C57BL/6 mice), which made emphisema and inflammatory lesion in the lung. 4 weeks after the injection, kinetic imaging (with a detector which software was made by our research group), normal x-ray and CT scan (NanoSPECT/CT plusz, Mediso) were made. Several kinetic series were recorded with different integration time (1-10 ms), all series comprise of 20 pictures.

The method were tested in clinically as well. Digital x-ray screening data of patients with IPF were used for the clinical tests. The kinetic images were processed by a MATLAB code that was developed and tested. The diaphragmatic movement and the elasticity of the lung were quantitatively characterised with two parameters. These parameters were compared with the results of pulmonary function tests (TLV, FVC, FEV1).

Results:

In silico methods proved, that the kinetic imaging could be suitable for visualizing the movement of the living organisms with low level of random noise. When analysing the preclinical measurements an optimal integration time that matches the expected value of respiratory rate can be chosen. In the kinetic images there was a remarkable difference between the COPD animals and the healthy control group. During the analysis of the clinical data the stability of the respiratory cycle was measured. The parameters of the diaphragmatic movement and the lung elasticity were calculated for all patients' several respiratory cycles from the kinetic images. For the IPF group the value of the parameters describing the diaphragma movement and the parameters of the lung elasticity show a linear correlation with the pulmonary function test's parameters (TLV, FVC, FEV1). The kinetic images which describe the moovements could be coregistered with the x-ray images that display the anatomy of the human body. The coregistration could help physicians establish the diagnosis.

Conclusion:

During the evaluation of the measurements, a new method has been worked out, which is suitable for quantitatively describing the main movement of the lung. Furthermore the parameters calculated from the kinetic images can eliminate the often false conclusions based on the morphological parameters. As a result, the method could be suitable for setting early diagnose. Based on these researches a new diagnostic protocol could be developed.

Downloads

Please sign in to download the files of this thesis.