Wyniki wyszukiwania - BazTech

1

Wpływ wyboru elementu strukturalnego użytego w przetwarzaniu morfologicznym na skuteczność nowej metody usuwania tła z sekwencji MRI

Kalicka R., Lipiński S., Pietrenko-Dąbrowska A.

Zeszyty Naukowe Wydziału ETI Politechniki Gdańskiej. Technologie Informacyjne

|

2008

|

T. 15

165-170

PL

W artykule zaprezentowano nową metodę usuwania zaszumionego tła z sekwencji skanów MRI. Na każdy skan składają się dwa obszary: przekrój mózgu i tło, oba zawierające szum. Tło powinno być odseparowane i wykluczone z dalszej analizy. Cel ten został osiągnięty dzięki zastosowaniu opisanego algorytmu, aplikującego podstawowe operacje morfologiczne: dylację, erozję, otwarcie i zamknięcie do uprzednio zbinaryzowanych obrazów MRI. Pokazano rezultaty przetwarzania z użyciem różnych kształtów i rozmiarów elementów strukturalnych użytych w operacjach morfologicznych. Celem było takie dobranie parametrów tych elementów by opisaną metodę uczynić możliwie uniwersalną i odporną na różnice w wyglądzie i jakości przetwarzanych sekwencji MRI.

EN

The paper presents a new method of removing the noisy background from the sequence of magnetic resonance imaging (MRI) scans. The scans contain the noisy head-cross data and also the noisy background data. The latter have to be removed and excluded from a further analysis. It is achieved by applying same basic morphological operations of dilation, erosion, opening and closing to the previously binarized MRI scans. The results of removing the background from the sequence of scans with the use of same different shapes and sizes of structuring elements are presented in the paper. The aim of this paper is to choose the best structuring elements to make the new method the most immune to various qualities and appearances of the processed MRI images.

2

Registration and normalization of MRI/PET images

Rumiński J., Suchowirski M.

Journal of Medical Informatics & Technologies

|

2005

|

Vol. 9

159--166

EN

Parametric imaging is more and more popular in dynamic brain studies. It enables to quantitatively or semi-quantitatively estimate physiological state and processes in brain. Parametric images represent spatial distribution of parameter values calculated for chosen mathematical model of the process or object. This work compares different methods of geometrical transformations for image registration and normalization. Appropriate method for image registration and normalization (in reference to atlases) is extremely important for common visualization of structural and parametric images in MRI and PET studies. Rigid and elastic geometrical transformations are implemented and compared. Additionally Delaunay triangulation and image morphing methods are used. Manual and proposed automatic registration and normalization methods are presented and compared based on MRI/PET and Talairach atlas images. Concluding, the proposed automatic image normalization method is accurate and using the combination of Delaunay and morphing methods can produce even better results.

3

Parametric imaging in dynamic glucose metabolism studies in brain

Rumiński J., Karczewski B.

Journal of Medical Informatics & Technologies

|

2005

|

Vol. 9

201--208

EN

Parametric imaging is more and more popular in dynamic brain studies. It enables to quantitatively or semi-quantitatively estimate physiological state and processes in brain. This work analyse the dynamic 18FDG-PET studies for estimation of brain glucose metabolism. The influence of the signal noise is analysed to estimate its influence on the final glucose metabolism parameter values. The LCMRGlc parameter is under investigation. It is based on three compartmental model proposed by Phelps. Using different 18FDG-PET data series obtained from independent sources the Gaussian noise was introduced (with different variance). Then the quality of the model fitting results were estimated. The final results clearly indicates than the noise is highly compensated in microparameter used in calculation of LCMRGlc. Concluding, it is possible to estimate the LCMRGlc parameter value even in the presence of noise.

4

Limitations of DSC-MRI for quantitative brain perfusion

Rumiński J., Bobek-Billewicz B.

Journal of Medical Informatics & Technologies

|

2004

|

Vol. 8

MM49--58

EN

Synthesis of quantitative parametric images in DSC-MRI is presented. Critical review of major limitations of the DSC-MRI method is discussed. It includes investigation of measurement procedures/conditions as well as parametric image synthesis methodology. Simulations, as well as phantom studies were used to verify theoretical limitations of the DSC-MRI. Especially, estimation of the contrast (Gd-DTPA) concentration by EPI measurements, the role of a phantom and its pipes orientation, influence of a bolus dispersion, bolus arrival time, and other signal parameters on an image quality. As a conclusion testing software package is proposed.

5

Obrazowanie parametryczne z wykorzystaniem dynamicznych badań tomograficznych

Bogorodzki P., Wolak T., Orzechowski M., Piątkowski A.

Prace Naukowe Politechniki Warszawskiej. Elektronika

|

2001

|

z. 130

73-89

PL

W niniejszym artykule przedstawiono metodę tworzenia obrazu czynnościowego z serii czasowej obrazów tomograficznych (CT, MRI). Obraz parametryczny wyliczany jest za pomocą modelu otrzymanego z referencyjnego obszaru mózgu z zastosowaniem analizy korelacyjnej. Przyjęcie takiego sposobu postępowania umożliwia wyeliminowanie najbardziej zawodnego i czasochłonnego etapu tworzenia obrazu czynnościowego, jakim jest procedura aproksymacyjna.

EN

Parametric images can be obtained from CT or MRI modalities through a special sequential scanning after the injection of a contrast medium. This kind of examination, called dynamic study, forms an image sequence that monitors the first pass transit of injected boluses of the contrast agents, through an organ of interest. The new proposed method applies correlation and discrimination analysis to each concentration-time curve instead of fitting them to any a priori and arbitrary tracer kinetic model. We have previously shown that temporal correlation methods can be used to analyse functional images of the brain, kidney and heart (Bogorodzki, 1996, Lo 1996).