|
|
| Identifier |
000450976 |
| Title |
Signal to Noise Ratio reproducibility measurements in clinical MRI examinations |
| Alternative Title |
Επαναληψιμότητα μετρήσεων λόγου σήματος πρός θόρυβο σε κλινικές εικόνες μαγνητικού συντονισμού. |
|
Author
|
Μακρής Τσαλίκης, Γεώργιος
|
|
Thesis advisor
|
Μαρής, Θωμάς
|
|
Reviewer
|
Καρατζάνης, Αλέξανδρος
|
| Abstract |
During this thesis we will attempt to evaluate the performance , in terms of SNR , of the equipment used for Magnetic Resonance Imaging of a brain by the homonymous department of the University Hospital of Herakleion. This procedure usually is part of a much larger Quality Assurance program but it the one that is repeated most frequently due to the fact that the SNR as metric is straight forward indicator of the image quality produced by the equipment.
This as a practice occupies valuable scantime that could be invested in patient care or forces medical physists and radiographers to perform the procedure after hours when the equipment is not in clinical use ( if that time exists). Performing such a crucial procedure after a long day is an error prone procedure. A viable alternative solution could be a procedure that runs with the already acquired datasets and acts as an indicator of the system status and does not mandate the use of the equipment itself.
In order to evaluate the performance of the aforementioned equipment and hence feasibility of this alternative method of performance status monitoring (in terms of SNR measurement) we designed an image evaluation process that consists of nine areas of measurments (ROIs) one in the background area of the image and eight measurements in four specific areas of the imaged brain.
These four areas were selected carefully to meet certain criteria , to be in key areas of the brain that pathology is easily depicted in order to be excluded in such case and the areas should cover the center of the image but also the edges. By combining these eight brain measurements with the background measurement (that acts as a noise measurement) via the practical SNR measurement technique as described by the American Association of Physicists in Medicine (AAPM) will be having data tabulated that reflect the performance in actual conditions and not in ideal conditions that are used in the usual Quality Assurance program.
This image evaluation process was applied in a single T2 Fast Spin Echo image that was acquired during the scanning of a T2 FSE imaging series when routine brain protocol was used in longitudinal period of three years. The review of those images was performed retrospectively in Evorad PACS-HIS workstation and the statistical analysis of the data was performed under MedCalc software.
Thorough analysis of the data revealed consistent recurrence of values with no statistical difference between the sets of variables proving the stability of the imaging equipment during the inspected period of time and providing us with the initial data that we could be standing infront of a viable alternative method for SNR performance status monitoring if performed under certain circumstances.
|
| Language |
English, Greek |
| Issue date |
2022-07-29 |
| Collection
|
School/Department--School of Medicine--Department of Medicine--Post-graduate theses
|
|
|
Type of Work--Post-graduate theses
|
| Views |
285 |
Signal to Noise Ratio reproducibility measurements in clinical MRI examinations
